So you want to build an open source community...

Just finished reading Matthias Sturmer's (153-page!) paper, "Open Source Community Building." Sturmer does a decent job of providing a high-level view of what attributes successful open source projects share. I'm not sure that Sturmer breaks new ground, but I'm also unaware of anyone else that has attempted to synthesize commonalities between successful open source projects. Thanks for doing the work, Matthias.

Sturmer reviews the following open source projects: Plone (CMS framework), Magnolia (CMS), Cocoon (Web application framework), Kupu (WYSIWYG browser editor), Lenya (CMS), Typo3 (CMS), eZ Publish (CMS), and Xaraya (CMS and application framework). Aside from a common content management theme, with a mostly European base, Sturmer chooses a nice cross-section of open source projects.

His findings (with my comments)?

• "Success" [which I'll define as "lots of people use it, commercially and non-commercially] is always somewhat serendipitous, as are market timing (unless you're Tim O'Reilly) and the mix of people involved. As is to be expected, Sturmer finds that positive human character traits (Commitment, experience, helpfulness, patience, etc.) all bode well for a project's chances. He also points to public demand for a project's features which, as with proprietary software, is very difficult to predict or fulfill. Moving on to things over which one has more control....
  
  
• Architecture of participation. Sturmer doesn't state it this way, but much of his findings have Tim's classic phrase all over it. Want lots of people to contribute (bug fixes, extensions, etc., since I think it's pretty clear that most code on a given project will be created by a small, core group of developers)? Write good documentation, use an accessible programming language (C and PHP are good candidates), have a modular framework, etc. Each of these things facilitates the influx of newbies and casual, drive-by developers/bug finders/bug reporters/bug fixers.
  
  
• Choose your license carefully. As Gregor Rothfuss, core developer on the Xaraya Project, relates to Sturmer,

  "The GPL sort of forces community building for legal reasons because people have to give back their code changes of GPL projects. That's different with the Apache License. Somewhat exaggerated, GPL communities are sort of forced communities and Apache communities are voluntary ones." (46)

  The thought that comes out of Sturmer's research is that for a newbie project, a GPL approach is more likely to succeed than a more open, BSD approach. The GPL operates much like traditional copyright, except it's inverted. But the same rigid control persists by forcing competitors to divulge modifications.
  
  
• Great initial source code. It's hard to attract a community to rubbish. You need to start with quality code to attract quality contributions. Eric Raymond has said that a project must begin with "plausible promise," even if it's buggy and incomplete at first. Developers have to catch the vision of good things to come, or they won't help to kickstart the project.
  
  
• Degree of applicability. An interesting point that should be obvious, but it's not. Sturmer writes that "the breadth of applicability determines the potential user community of the open source project" (e.g., supporting various platforms increases the range of developers who will have use for and interest in the project) (104).
  
  
• Modularity of code. If I had to rank these attributes, modularity might well take the number one spot. It is critical to a project's success that it scale well, both in terms of functionality (and the code that goes with it) and in terms of outside contributions. Apache has been phenomenally successful, at least in part, because the core web server is kept so small with various other pieces (Cocoon, Lenya, Geronimo, etc.) serving as separate but integrated (or, rather, integrate-able) extensions. Linux? The kernel is kept small, and is expanded through classes, etc. Mozilla? Same thing - plug-ins, themes, and extensions. Good projects lower the bar to outside participation, and modularity is at the very heart of this principle. Even Microsoft is getting the modularity religion, as Jim Allchin recently told the WSJ.

Of course, even with all these things going for a project, serendipity will have a lot of sway. I once complained to Tim O'Reilly that his "architecture of participation" wasn't prescriptive - it is descriptive. What can I do with it, I asked?

The answer, of course, is that you can create the highest possible chance that your project will succeed, once chance/luck/serendipity drop by. Of course, as Sturmer's open source interviewees also said, great code and a good architecture are only the beginning: old-fashioned marketing (done, perhaps, in new-fashioned ways over the 'Net) is indispensable. Ultimately, there are no shortcuts in open source.

Analyst: Open source (corporate) community?

Jason Matusow of Microsoft has an interesting slide in his current deck which corresponds to a conversation I had this morning with Joel West, as well as a few of my recent posts. I've included Jason's slide below (which he has presented at OSBC and a range of other places).

In Jason's presentation, he makes the very interesting (and valid) point that open source is increasingly corporatized. By this I mean I don't mean "commercialized," which has also happened. I mean that open source communities have increasingly been overcome by corporate interests. Using the Linux kernel as an example (see below), it's clear that the community is no longer a mishmash of volunteers writing code they love.

Talk to these developers, and they will sharply dispute what I just said. "Code first, corporation second!" they'll say. And, no doubt, most (if not all) of them mean it. That's not the point.



The point is that the more corporatization, the less free-wheeling freedom. This isn't a bad thing (unless you're RMS), but it does call into question the idyllic open source community in which people persist in believing.

For my part, I think the corporatization is a great thing. Novell, Red Hat, IBM, HP, and others have dramatically helped to solidify the Linux kernel into an enterprise-class operating system. IBM's early involvement in Apache no doubt helped to ramp its quality and distribution.

My point is that we shouldn't be naive on where the code comes from (i.e., who is developing it). We also should seriously question the premise that open source offers a novel, effective way to coordinate individual firms. It hasn't. As Joel indicated, Eclipse offers the only credible example of a true "corporate community" project. The rest - JBoss, MySQL, SugarCRM, Alfresco, etc. - are run by individual companies that, theoretically (a la the GPL) make it possible for others to collaborate with them on equal terms, but in reality don't invite such semi-forking.

All of which means we need to rethink what open source actually gives us that closed-source development doesn't. Lots of eyeballs to find bugs? Probably not. Vendor-neutral, code Darwinism (i.e., best code wins, not corporate politics)? Not that one, either.

I think it comes down to greater customer intimacy (again, to quote John Powell of Alfresco), higher trust with customers (they can see the code, if they choose), lower barriers to distribution and adoption, and a different, cheaper way to buy software. Any one of these would justify the rapid rise of open source, in and of itself. Together, they make for a very compelling value proposition.

News: GPL 3.0 may close the web distribution loophole (ZDNet)

Two years ago, I complained that the Free Software Foundation might be aiming GPL 3.0, in part, at the so-called "ASP loophole." Bradley Kuhn, executive director of the FSF, slammed me for assuming the worst without asking him first.

Well, two years later, it seems like I was right to complain. ZDNet is reporting on an O'Reilly interview in which Richard Stallman states that the FSF would like to close the loophole. To what am I referring? Have a read:

The next version of the General Public License may tackle the issue of Web companies that use free software in commercial Web-based applications but don't distribute the source code.

At present, companies that distribute GPL-licensed software must make the source code publicly available, including any modifications they've made. Though the rule covers many businesses that use GPL-licensed software for commercial ends, it doesn't cover Web companies that use such software to offer their services through the Web, as they're not actually distributing the software.

GPL 3, the next version of the free software license, a draft of which is expected to be released in early 2006, may close this loophole, GPL author and Free Software Foundation head Richard Stallman said in an interview with publisher O'Reilly Media.

Stallman said developers may be encouraged to add a command to their GPL-licensed Web application that lets users download the source code. The inclusion of this command in modified versions of the program will then be enforced by an additional clause in GPL 3.

"We're looking at an approach where programs used (on a public server) will have to include a command for the user to download the source for the version that is running," Stallman said. "If you release a program that implements such a command, GPL 3 will require others to keep the command working in their modified versions of the program."

This change would have no effect on existing software but could be added by developers to future versions of a particular program, according to Stallman. He said this was only a "tentative plan" as it has not yet been studied fully to see whether it would work.

I think this would be a huge mistake. It's unclear to me why RMS wants so desperately to devour every piece of software he can, but this idea goes well beyond his standard "free as in freedom" pitch.

Why? In part because suddenly all of that "free" (as in cost and as in freedom) code that built the web would be suddenly less free. Vendors (like Google, Amazon, etc.) who would suddenly have to divulge much of their code - Tim O'Reilly would say that such disclosure would not hurt them, or not much, since their value is not in their code but rather in their participation architectures, but I'm not so sanguine.

More importantly, this broad view of copyright would creep into the user base. At present, a Morgan Stanley can heavily modify Linux or other GPL'd code all it wants, and keep these changes for itself. Because it is not a distributor of software, in the classic sense, it need not divulge its modifications. But if one defines distribution in the sense that RMS would like it to be, suddenly internal "distribution" (over intranets, for example) would require forfeiture of that internal code.

This would put a huge chilling effect on open source.

Now, mind you, RMS (last time I checked) had no authority over US copyright law, and his expansive views of "freedom" are not necessarily (or even probably) consistent with US copyright law. So, this may just be a case of RMS overreaching ("free" as in "free license"). I certainly hope so. I don't think the open source (or "free source," if you prefer) community is well served with hard-line approaches that rigidly insist on one particular view of freedom.

For example, I think Massachusetts' recent decision would have yielded more freedom had it offered OpenDocument, PDF, and Microsoft Office formats. Free as in multiple choice,

We don't need GPL 3.0 to try to consume the web.

News: Keeping Linux (desktop) from the mainstream (eWeek)

eWeek has a report on the "Open Source Goes Mainstream" panel at MIT's Emerging Technologies Conference. It's mostly pedestrian, but one paragraph caught my eye - actually, it made me burst out laughing, because it unintentionally captures the culture of Novell's old guard. Hilarious, in a Dilbert sort of way....

Even within companies, such as Novell, that have embraced open source software, there are not always alternatives to closed source products, said Miguel de Icaza, founder of the GNOME (GNU Network Object Model Environment) project and now a vice president at Novell.

Novell moved all its employees from Microsoft Office to Open Office and around 70 percent of its users from Windows to Linux.

However, some critical business applications such as those used for finance or purchasing can only run on Windows, he said.

Incorporating open source development practices into traditional corporations can also pose challenges, said de Icaza.

At Novell, managers who were used to resolving problems in meetings and through face-to-face encounters had to learn to accomplish the same tasks using e-mail [Imagine that! :-) ], after the company bought de Icaza's company Ximian, which had a team of 40 open source developers working from home offices scattered across the globe, de Icaza said.

Novell does many, many things extremely well. Meetings are not one of them. It is a meeting culture but, as Miguel points out, that's starting to change. People have email addresses now.... ;-)

Commentary: Adam Bosworth on "reactive development" (ZDNet)

Phil Wainewright has a great post on how Google, Salesforce.com, and other web services-type businesses have changed the way vendors interacts with customers. Transcribing from Bosworth's DreamForce keynote, Wainewright writes:

[One] factor [Bosworth] brings out is a commitment to an iterative approach that continually delivers features to users and tests their reactions before deciding whether to continue in a particular direction. This is one of the most distinctive characteristics of on-demand developers and product managers, [Bosworth] says:

"They are very intimately involved with operational data. They are trying things out. Rather than figuring out for two years what you might want, and then spending another two years building it, and then spending a year trying to figure out, maybe it'll work for you - or more likely just ignore you - they're trying it after two weeks and then watching with fascination to see how many people actually used it and [saying] 'Oh, good that one looks like it's working for people, let's do more of that' - or 'No, that one's not actually working. It seemed like a good idea, but no one wants it. Let's not do more of that.'

"...A model like that, where you watch, you try things out, you learn and you iterate, is a model that very quickly stays in touch with what you want."

What's interesting to me is that this same principle largely follows for open source. While not iterative in exactly the same sense, open source also provides an "intimacy" with customers (as John Powell of Alfresco calls it) - a less monolithic, take-it-or-leave-it approach to product development.

Bosworth also says,

"Software is not about intelligent design, it's about intelligent reaction. It's about figuring out what works for people -— which will change over time, as costs go down, as mobile VoIP phones suddenly become pervasive - the things that we need to do for you will not be the same as the things we needed to do for you before a laptop was the standard operating piece of equipment for everybody, which was not the same as when you installed enormous disks that were 100k disks. You have to learn and you have to change with the times. You have to follow your customers in real-time."

Amen.

Announcement: Last chance for free Sony PSP (Register for OSBC by 9/30/05)

Shameless Plug

Just thought I should announce that tomorrow (Friday, September 30, 2005) is the last day to pick up a free Sony PlayStation Portable by registering for the Open Source Business Conference. Consider it OSBC's first present of the holiday season. :-)



Try getting one of these from LinuxWorld....



/Shameless Plug

Analyst: If I were a betting man...(Linux vs. Windows server shipments)

Just found this (of all places, on Novell's website [Click on "Figure 4" to get to the text below] - you'd think someone would have shared this with me...?). As the IDC analysis goes,

IDC expects Linux to continue biting at the heels of Windows market share and increase from 24 percent to 33 percent within two years forcing Microsoft to swallow even smaller profit margins just to hold on to their share.

Reasons for this?

1. Linux is starting to move beyond Unix - it's slowly starting to take a bite out of Windows. Why?
   • Whatever the relative cost of Linux talent (Robert Frances Group says it's cheaper than Windows system administrator talent, while Forrester says it's pricier), Linux requires fewer support resources. In Boscov's case (a SUSE Linux customer - full disclosure), they have all 200 instances of Linux in the enterprise supported by one person. If you read Microsoft's TCO case studies, its customers only state that Microsoft is cheaper because they didn't want to invest in any new training, which is a bit of a canard since Unix skills transfer automatically to Linux skills and, last time I checked, Unix predates Windows (so, in many cases, it is Unix skill, not Windows skill, that needs transferring....);
     
   • Linux hardware is much cheaper than Unix hardware (old news - Cendant indicates it saved $100 million by moving from proprietary Unix hardware), but also significantly cheaper than Windows hardware (Higher utilization rates, more apps/server)
     
   • "Windows installations require twice the number of administrator hours...spent patching systems and dealing with other security-related issues." (Robert Frances Group 2002) Even the Yankee Group, which has been bearish (neutral, they'd say) on Linux and open source indicates that only "18 percent [of its surveyed enterprises] report[] that [Windows Server 2003] is comparable to Linux reliability in terms of unnecessary reboots." This came right after the analyst's statement that Windows Server 2003 has come light years in terms of security and stability. Great progress, no doubt, but obviously a long way to go.
2. Microsoft customers have been pushing back on price, resulting in a better pricing structure/program, as both Yankee Group and SearchWin2000 have reported.
   
3. I suspect that choice really is an issue for CIOs. Some prefer Microsoft's "integrated innovation" model, which aims to remove complexity (and, necessarily, choice) from the IT purchasing equation by tightly integrating a broad suite of products. The problem with this "one ring to rule them all" approach is that whether you think Gates is Sauron or not (and I, for one, do not), the fact remains that many CIOs prefer to manage their vendors, rather than have the vendors manage them.

Anyway, I'm sure there are many more reasons for Linux's rise and Microsoft's leveling off. But it's a head-to-head battle that hasn't been won - by either side - and it's no time for complacency and back-slapping.

Analyst: The Myths of Open Source (CIO Magazine Special Report)

Good article that dispels some myths around open source. Here's a teaser:

MYTH 1
THE ATTRACTION IS THE PRICE TAG
One of open source's most touted benefits is its price. Download the software, install it - and don't pay a penny. That's the theory. But to a surprising number of open-source user companies, the price tag - —or lack of one - —is irrelevant. "It's not about being cheap," insists Employease's Alberg. "It's about doing our jobs effectively—and we're willing to pay quite a bit for that. We want stable software that does what it says it will do."

What Alberg finds fascinating about moving to open source is the performance improvement that resulted. The move to Linux, for example, dramatically cut the rate of server failure experienced by the company. Typically, under NT, one of the company's servers would fail each working day. Now, he says, "we get at most two failures a month, —and often don't get any in a month."

Linux also runs Alberg's applications faster than NT, a fact that has meant that despite more than doubling its business since 2000, the company hasn't needed to buy more servers. "Linux increased our capacity by between 50 percent and 75 percent," says Alberg.

Even so, Alberg is careful to make clear that his commitment to open source isn't the blind buying behavior of a zealot. He wouldn't, for example, go open source if it were more expensive than proprietary code. "Solaris is a strong commercial operating system. We'd choose it over open source if we found it to be less expensive," he says. "[While] cost is a huge driver for our decision-making process, we cannot risk choosing an inferior solution to save money. We couldn't even consider open source if it weren't at par with - or in some cases better than - —commercial alternatives."

Ask many users of open source and a similar story emerges. "Cost savings weren't really a factor in our decision to go open source," says John Novak, CIO of 330-plus hotel chain La Quinta, which is moving its online booking system - previously on BEA's WebLogic - —to a combination of Apache, JBoss and Tomcat. "What got us into it was that it was simply the best technology open to us."

Analyst: Tech support cheap, venture-backed CEOs? Not so cheap (Kedrosky)

Paul has an interesting post on the rising pay of venture-backed CEOs. I remember the day when startup CEOs were working for the equity. Apparently, that's a bit passe. Be sure to check out the VentureSource report (in PDF, no less! Hurray for open standards! :-) in Paul's posting:

I had an interesting conversation with the CEO of a venture-backed company recently where he "explained" why CEOs of such firms need to be paid more now than an inflation-adjusted increase from, say, five years ago. He argued that in the absence of public market exits that the risk/return tradeoff for being a CEO at a venture-backed firm is no longer what it once was. Put differently, cash (in salary) is king, and anyone who relies on options is a chump in the face of the risk of running a startup staying constant, or even increasing.

I don't entirely disagree, but to the extent that the comp for venture-backed CEOs skews in any significant way toward cash it augurs ill for creating and financing cost-effective startups. But data today does corroborate his point, with the following having been released by VentureSource:

...the CEOs of [startup] companies earn about $10,000 more a year than they did in 2004, and about $35,000 more than in 2003. Overall, the CEOs are earning $260,000 in total salary, bonus and commission compensation, based on an annualized median.

I understand the point that in the absence of easy exits, hard cash becomes more attractive. But, as Paul implies, isn't that what VCs are paying for? Entrepreneurs who wouldn't settle for an easy exit, anyway, and will push through until their company commands an IPO or big exit? Am I missing something?

News: OpenOffice 2.0 ready for download

Heise Online is reporting that OpenOffice 2.0 is officially ready for download:

This Thursday, the Open Source project OpenOffice presented Release Candidate 1 of its free Office suite; this test version -- possibly the last one -- was, however, already available for downloading from the OpenOffice web site before the official release on Thursday: all of the mirror servers had been prepared since Wednesday. With the release of RC1, the developers are entering the last test phase. The results of this test will show whether additional release candidates will have to be produced before the Final Release....

OpenOffice 2.0 almost exactly corresponds to the commercial StarOffice 8 presented yesterday by Sun, with the main difference being that the free version does not contain as many templates and has different correction tools.

Give it a try, and let me know what you think. I'm hoping the NeoOffice people will have it ready (native) for the Mac soon. In the meantime, I'm going to try it out on my Linux and Windows machines.

Analyst: Tech support benchmarks

As open source drives more and more value to services, it's important to keep tech support and related services (including training) costs in check. Below is data from the Association of Support Professionals (ASP), derived from a 2005 survey. Are you paying too much? Or, given that your entire business may increasingly rest on the intelligence and service of this group, are you paying too little?

News: Dell, the king of cheap, tries to go luxury(???)

That's right. As BusinessWeek reports, Dell is moving up-market (trying to, I should emphasize) into high-end computers (i.e., with hardware like Apple's, except that they're stuck in commodity parts land) and even high-end consumer electronics (plasma screen TVs).

Hmm.... Could anything be more incongruent than Dell pretending to be luxury? The company's hardware routinely breaks (but hey! They're cheap!!), it has recently had a slew of complaints over poor customer service (long wait times, poorly trained support staff - but hey! They're cheap!!), and it has generally shown itself to be fantastic at cobbling together cheap components, but not of delivering high quality.

So now it's planning to cobble together more expensive components and add a "premium" support line? Whatever. Try again.

If you're tired of playing the commodity game, you're going to need more than a dedicated support line for your 2-3 premium customers. You'll need a completely different corporate culture. In other words, you'll need to be Apple. Not going to happen. Stick with what you know.

News: Apple's not into digital convergence

Don't look for a wireless toaster anytime soon from Apple. Or so says Jon Rubinstein, an Apple leading light, as reported AppleInsider:

"Is there a toaster that can also brew coffee?," quipped the iPod chief during an interview with Germany's Berline Online. Answering his own question, Rubinstein explains that no such device exists because it would not provide any benefits over an individual toaster and coffee machine.

"Many companies believe in [the convergence theory], but I personally do not," said Rubinstein. "It's important to have specialized devices."

Well, I don't drink coffee, but I bet my kids would love the toaster to toast their waffles and mix up their hot chocolate....

Analyst: (More on) Defining successful open source communities

Audris Mockus (Avaya Labs), Roy Fielding (eBuilt), and James Herbsleb (Bell Laboratories) have posted interesting research on how open source development communities work, using Apache and Mozilla to frame hypotheses of successful open source projects. You can find the paper - "Two Case Studies of Open Source Software Development: Apache and Mozilla" - here [PDF]. It's worth a read, especially if you're a commercial entity (or a VC investing in such) looking to build an open source community around your company's project.

The authors of the report list a few hypotheses developed from their analysis of Apache, and then refine them in light of their Mozilla analysis (and, unless I misread, they also tested them against a few other projects). The results are interesting:

Hypothesis 1a: Open source developments will have a core of developers who control the code base, and will create approximately 80% or more of the new functionality. If this core group uses only informal, ad hoc means of coordinating their work, it will be no larger than 10-15 people.

Comment: See the chart at right. As the researchers note,

The chart "shows that the top 15 developers contributed more than 83% of the MRs [managed releases] and deltas [an MR generates one delta for each of the files it changes], 88% of added lines and 91% of deleted lines. Very little code and, presumably, correspondingly small effort is spent by non-core developers (for simplicity, in this section we refer to all the developers outside the top 15 group as non-core). The MRs done by core developers are substantially larger than those done by the non-core group." (14)

Hypothesis 2a: If a project is so large that more than 10-15 people are required to complete 80% of the code in the desired time frame, then other mechanisms, rather than just informal, ad hoc arrangements, will be required in order to coordinate the work. These mechanisms may include one or more of the following: explicit development processes, individual or group code ownership, and required inspections.

Hypothesis 3: In successful open source developments, a group larger by an order of magnitude than the core will repair defects, and a yet larger group (by another order of magnitude) will report problems.

As the authors report,

"[The chart at right] shows that participation of wider development community is more significant in defect repair than in the development of new functionality. The top 15 contributors produced only 66% of the fixes. The participation rate was 26 developers per 100 fixes and 4 developers per 100 code submissions, i.e., more than six times lower for fixes. These results indicate that despite broad overall participation in the project, almost all new functionality is implemented and maintained by the core group." (15)

The authors further found that "of the top 15 problem reporters only three are also core developers [on Apache]. It shows that the significant role of system tester is reserved almost exclusively to the wide community of Apache users." (18) This is interesting, because it points to the importance of developing a wide user base (which, as I indicated in my research posting yesterday, generally follows from the core team developing a significant body of code that others can use. Seems logical, but logic doesn't apparently follow the vast majority of projects on Sourceforge.net. It points to the need to free up core developers' time for writing new functionality - get a large body of beta testers, as MySQL, JBoss, and others have done.

Hypothesis 4: Open source developments that have a strong core of developers but never achieve large numbers of contributors beyond that core will be able to create new functionality but will fail because of a lack of resources devoted to finding and repairing defects.

Hypothesis 6: In successful open source developments, the developers will also be users of the software.

As the authors note,

"The reasoning behind this hypothesis was that low defect densities are achieved because developers are users of the software, hence they have considerable domain expertise. This puts them at a substantial advantage relative to many commercial developers who vary greatly in their domain expertise. This certainly appears to be true in the Mozilla case. While we did not have data on Mozilla use by Mozilla developers, it is wildly implausible to suggest that the developers were not experienced browser users, hence, "domain experts" in the sense of this hypothesis." (37-38)

The researchers make points about open source code quality, defect resolution response time, etc., but these are of secondary importance to me, so you'll have to read the full paper to see what they say.

Some parting thoughts:

• As the researchers found in Mozilla's case, good documentation, tutorials, and refined development tools and processes can help grow a community. It's tough for people to contribute if they haven't a clue as to where to begin....
  
• In larger projects (like Mozilla, unlike Apache - Apache's core code is kept very lean, with all new functionality farmed out to separate modules/projects), modularity is critical. Unfortunately, for projects that start out commercial and then try to go open source, the code base often is riddled with interdependencies (as was the case with Mozilla). Mozilla found a way around this by establishing module-by-module code ownership, with that one individual sufficiently knowledgeable about her module to ensure code conflicts don't arise within that module. These individual owners, then, must carefully coordinate with other module owners, unlike in a small core project like Apache, where the dozen core contributors are well aware of what's going on across the core, enabling them to contribute to various pieces of code within the core.
  
• As in commercial software, there is no free lunch. If you want people to use your code, you have to spend the time and effort to build something worth downloading, using, and commenting on. It's no easier than commercial software, and requires an equivalent amount of work. The payoff, however, is a user base that feels ownership in the project, and not merely a buyer. That's valuable.

Analyst: The 451's take on OpenLogic

I've long admired OpenLogic and BitRock, the early providers of the "glue" that holds open source projects together. SpikeSource and SourceLabs have really made the space seem credible, but it has been OpenLogic and BitRock who have been shipping product to customers for two to three years.

The 451 Group has an interesting analysis of OpenLogic, some of which I've excerpted below. [Note: I just shortened it because I felt I may have used too much of The 451's original source material in my blog. Sorry to the authors.]

OpenLogic positions itself as elder statesman in open source services

Analyst: Dennis Callaghan
Sector: Enterprise Software »»
Date: 27 Sep 2005

With a high-profile management team, blue-chip venture backers and a Silicon Valley pedigree, SpikeSource got a lot of attention when it launched into the open source services space earlier this year. Its primary competitor, OpenLogic, has none of those attributes, but does have a longer track record and a larger customer base. It aims to ride its BlueGlue [a rather unfortunate name, in my (Matt's) opinion] stack to new growth, especially in larger organizations, to raise more money and to position itself as an open source product, rather than services, company.

Impact assessment

The message
OpenLogic has quietly built a large customer base for its open source stack and related services and is preparing for faster growth, particularly among larger enterprises.

Competitive landscape
SpikeSource is OpenLogic's most direct competitor, although OpenLogic supports more projects and claims more customers. Free downloads that bypass stack and service offerings are an even bigger competitor.

The 451 Assessment
Though less heralded than rival SpikeSource, OpenLogic is a company to watch in the open source services space. But with increasing demand for its services from larger enterprises and more growth expected, the company needs to invest more in building out its indirect sales channels and expanding its direct sales capabilities beyond the Web if it intends to take advantage of those opportunities.

Context
OpenLogic can trace its roots back to 1998 when current CTO Rod Cope founded a consulting firm around Java application development consulting called EJB Solutions. Cope, a former General Electric and IBM executive, had been CTO at Anthem Inc. before starting the firm. In early 2002, the company changed its business model from consulting to open source products and changed its name to OpenLogic.

Later that year, it released the first version of its BlueGlue platform, which, true to the company's heritage, is focused around open source Java components for building applications. Roy Kligfield, a former geology professor who had been a VP at SignalSoft before it was bought by OpenWave, was brought in last year as president and CEO. While the company offers its BlueGlue open source stack on a subscription basis, it stresses that is a product company, not a services company, though it offers support and other services on top of its stack. Based in Colorado, OpenLogic has 20 employees plus subcontractors. All are in North America.

Customers
OpenLogic has nearly 450 customers in 21 countries, mostly North America and Europe. The company doesn't track average deal size, since deals can range from a small IT shop buying a single developer subscription to General Motors, which OpenLogic counts as a customer. Lockheed Martin and SAIC are also notable customers. The company is seeing increasing demand from larger firms – demand it expects will 'skyrocket' next year. OpenLogic has no real vertical focus, but it does see a lot of government agencies, universities, supply chain management firms, software development houses and consulting firms in its customer base.

Good analysis. I'm glad to see the company getting some recognition.

Analyst: The nature and size of the open source community

I'm doing an analysis of commonalities between successful open source projects (If you can point me to good research on the topic, please email me!!!), and came across an interesting academic paper (Requires purchase) related to the subject. Here are some of its findings (none of which will be surprising to those who have followed Joel West and Siobhan O'Mahony's work, but surprising if you still believe in the magical open source community, numbered in the millions, anxiously waiting to contribute code to your project):

1. It is interesting to compare horizontal applications (applications used to build other software, the end user is required to program and is, likely, a software professional) with vertical ones (applications used by an end user, no programming is required). Horizontal applications (categories Internet, System, Software development, Communications, Database, Security) account for 72%. The researchers interpret this data as evidence that the OS [open source] community is largely oriented to produce applications for the same community.
   
   [No news here. The difference only comes when we add commercial open source to the mix. Once we bring in the commercial entities, the only thing bounding open source development is the success of the companies' business models.]
   
   
2. Open source projects [at least, as housed on FreshMeat (the source for the researchers' data set), which tends to host newer and, hence, smaller projects] tend to be small (82% - suitable to one or two developers) and young. 60% of open source projects (as measured in February 2001 - admittedly, an ancient data set) had been in development less than a year, 22% from one to two years, 15% two to three years, and around 2% more than three years.
   
   
3. The GPL license prevails, at 77% of projects. The LGPL is second at 6%, and BSD trails in third at 5%. All other licenses account from 3% to 1%.
   
   
4. C is the most used programming language (41.5%), followed by C++ and Perl (~14% each), then PHP, Java, and Python (5% - 8%).
   
   
5. 49% of projects have only one person developing the application; 15% have two to three developers; 20% have four to 10; 9% have 11 to 20; and 6% have more than 20. Clearly, this calls into question the ideal of "community" in open source. Last time I checked, even with my multiple personalities, I'm not a community.
   
   
6. The researchers assumed that larger projects would have more developers. Wrong. "Instead we find that there is no meaningful increase of size with developers." Apparently, "[fewer] developers produce the same amount of code...." This isn't surprising - I take it as a given that a minority of people in any company/project/etc. will produce a majority of code/product/whatever. The interesting thing to note is that an open source project can be wildly successful without a massive community contributing code to it. The key is code quality and contributor productivity.
   
   
7. Related to the above, 73% of projects have only one stable developer. 10% more projects have two stable developers (defined as a developer with a "prolonged collaboration with the project"). That leaves just 17% of projects that have more than two committed developers.
   
   
8. Added to the above, the researchers found that 55% of projects have no transient developers at all ("Transient" defined as those providing at most one patch in the development of any section of a project or up to three patches to the same part of the code base). Of the remainder, 9% have one transient developer, 8% have two, and 20% have between two and 10.
   
   
9. How does a project attain a larger status, such that it can sustain 10 or more developers? The researchers find that such conditions include "a defined and clear architecture and an adequately appealing function offered; both conditions require a meaningful size of code." This means that the initial developer(s) must be committed to see the project through its young, immature phase. But this isn't surprising, as the same principle holds true for religious movements, political uprisings (the United States is one example), and various other projects. Bluntly put, you need a fanatic or two (in the nicest sense of the word) at the beginning to blindly push forward against all odds. Open source software development appears to be no different.
   
   
10. 80% of projects have less than 11 users (measured in terms of those who "subscribe" to a project - i.e., those who register and download a piece of code).
    
    
11. 15% of projects are actively developed - the remainder (85%) wither and die on the vine or are, at best, "lethargically" developed. Over the six months measured, 90% of the projects on FreshMeat did not change.

Undoubtedly, there are problems with the data set. But regardless of how many holes one can point out in these researchers' work, it corroborates very well with other academic work on the subject. The myth of a global, expansive open source development community is just that: a myth. The reality is more like severe clumping of development around Linux, Apache, and very few other projects. (Even JBoss and MySQL, as I've written before, are overwhelmingly developed by those respective companies, and not by a crowd of outside developers. 95% and 85%, respectively, I believe.)

Does this mean open source is a sham? Not at all. It is still a great way to engage prospective customers, incorporating them into one's development. And it's a great way to replicate Google's "perpetual beta" development methodology, which allows them to innovate and deliver code faster, because it artificially sets expectations low.

It's also a reminder that companies engaging in open source should not delude themselves into thinking that some amorphous community will do their work for them. There is no community to do this. Whether one is a company or an individual developer, the onus of code production is on you. The community only comes when the project initiator has done the grueling, constant work to make the project worthwhile.

In this way, open source really isn't so different from closed source software.

Analyst: More from Yankee Group

I just finished reading through Yankee Group's 2005 North American Linux and Windows TCO Comparison Report, Part 2, available here. The general take is that Windows is equal to Linux in terms of security, because of improvements Microsoft has made (both process and code-level) and because of improvements Linux has not yet made.

Now, I was a bit cavalier in dismissing Microsoft the other day. This is largely due to how the company markets itself - "Microsoft is good because you already use us, and might as well use more." That's weak, obviously, and not consistent with the value that the company actually provides (value that JBoss certainly sees.

What stood out to me in this report was a comment from a New York-based Linux and Windows customer, including Yankee Group's follow-on commentary:

"The bulk of the [Linux] vulnerabilities are based not on the Linux kernel itself but on the applications running on Linux, such as Apache, PHP, eCommerce software and SSH."...

The company's Linux security issues are directly attributable to poor documentation and resources to track and resolve multiplatform/application security issues in the Linux environment.

This is consistent with Yankee Group's finding that file and web servers tend to be recovered disproportionately faster on Windows networks than Linux networks.

What does this mean? It means that Linux vendors (like Novell) need to constantly improve the full support package. Microsoft promotes its "integrated innovation," which has a lot of negatives (it creates weaknesses at the intersection of every interlocking Microsoft application/middleware/etc.), but also positives, one being that Microsoft takes responsibility for security at the operating system level and above, as the OS interacts with its other technologies.

Linux needs to be a bit more like this - not necessarily in the integrated suite approach that Microsoft takes, though the average user tends to want this. No, I'm referring to the responsibility to provide more comprehensive support - it's no longer enough to simply support the OS and expect others to fend for themselves above the OS.

This is one reason, for example, that Novell has kept building its support for complementary open source products, like MySQL and JBoss. (Btw, I guess Novell ran out of words it could use in these press releases. The headlines sound suspiciously similar.... :-) ]

News: Recent executive moves....

Not sure if anyone catalogues this stuff, but since I find it interesting....

• r0ml (aka "Robert") Lefkowitz has left Optaros. He hasn't found a new home yet, so for those interested.... No bad blood between him and Optaros, on either side. r0ml is just trying to find his ideal job. (Being "just a bozo" doesn't count.) r0ml is extremely intelligent, knows open source cold, codes (was voted on of Wall Street's best back in his days at Merrill Lynch and elsewhere), and is generally an awesome person. You should hire him.
  
• Dave Rosenberg, who ran LinuxWorld's conference program (and actually turned LW into more than just a dull exhibition - the conference part of LW has become really interesting) and Comdex before that, left IDG. He's now CIO at Glass Lewis and also spends part of his time as Principal Analyst at OSDL. Another great, talented person.
  
• Angie Anderson, formerly Novell's VP of Linux Engineering, left to go to OpenWave. Not sure of what she's doing there, but I think she's a general manager.
  
• Pamela Roussos left Zend, where she was VP of Marketing. No idea where she has gone....
  
• Mike Evans, formerly VP of Partner Development at Red Hat is still with the company, but his job recently got bigger. He's still doing partner development, but has had corporate development, the run-time business, and business development added to his realm of responsibility. This is a good move on Red Hat's part - Mike is exceptionally capable and a hard worker.
  
• Lars Nordwall left Novell, where he was a senior member of our consulting team, to run sales (as VP of Sales) for SugarCRM. Great company, great guy. Good match.

That's it for now. I believe, however, that there is a range of open source startups (Funambol, Zend, JasperSoft, Greenplum, etc.) that are looking for executives right now. Why not you?

News: HP learns the hard way - doing business in France

It's hard to do business in France without doing business with France. At Lineo, I was charged with spinning out a French subsidiary (the original INUP team), which experience no one could pay me to do again. Lineo acquired the company; Lineo went through hard times; Lineo wanted to set the company free again (no strings attached); the subsidiary/company wouldn't go without massive guarantees (1-2 years of severance, for starters); I lost 10 years of my youth....

HP is now trying to reduce headcount in France by 1,240 (6,000 total in Europe). But France isn't letting go so easily. As The Register is reporting, the mayor of Grenoble is demanding 1.2 million Euros back, which it says (and HP denies) it paid out in subsidies.

Not sure how this will end up, but it's not going to do France any favors. It's just not the sort of place that companies want to set up shop anymore, given the huge costs of exit (which, unfortunately, is sometimes necessary).

Bifurcation for two weeks....

I'm going to be guest blogging for my friend, Dave Rosenberg, over at InfoWorld for the next two weeks. Dave is pretending to be an international playboy and has left me with his laundry. Oh, well.

I'll continue to blog in both places, and will try to keep the content both here and there. (Herethere.com - nice URL. Want to buy it? :-)

You can find my first post over there, starting with an elaboration on what I mean by "the desktop is dead." Matt Harrison just wouldn't leave me alone on that....

What money buys...

...hopefully, someday, a better seat at Arsenal's matches. (This was taken last Monday at the Arsenal match. They spanked Everton.) At least Mark Watson joined me. I think he was secretly hoping Arsenal would lose....

News: What's $100 million between friends?

Wow. CNET is reporting(originally reported by the WSJ, but since a link to it is worthless... :-) that Oracle's Ellison agreed to a $100 million settlement over allegations of insider trading. I guess the trading must have been worth more than the $100 million, because Ellison seems to not have put up a fuss.

A San Mateo Superior Court judge on Monday postponed signing off on an $100 million settlement agreement between Oracle founder Larry Ellison and Oracle shareholders over an insider trading lawsuit.

Instead, Judge Jonathan Schwartz [No relation, I'm sure, to the pony-tailed Sun exec ;-) ]will review the matter on Nov. 15 and issue his decision on whether to accept the agreement hammered out by the parties. According to a report in the Wall Street Journal, Ellison, Oracle's chief executive, agreed to pay $100 million over a five-year period to charities and $22.5 million in attorneys fees to the plaintiffs. Ellison, who faced allegations he engaged in insider trading in 2001, agreed to the settlement without admitting guilt.

News: Peru passes law favoring free software

I hope Google's translation is wrong, but the news that Peru has passed a law favoring free/open source software strikes me as completely muddled. I just don't think legislation should enshrine preferences for closed OR open source software.

Sovereignty is about more than the openness of the code or the file formats. Sovereignty is about avoiding lock-in to any one vendor. There are many ways to accomplish this, without writing law that favors open source software over closed source software. r0ml presented a great way to create choice in software at OSBC earlier this year - have a peek.

Commentary: Linux system recovery (Yankee Group)

Reading through the Yankee Group's "2005 North American Linux and Windows TCO Comparison, Part 1" now, and finding it pretty interesting. It seems to be reasonably researched, though something caught my eye:

Windows Systems Recover More Quickly from Security Attacks

In yet another surprise, respondents indicated that network administrators were able to restore Windows servers about 30% more quickly following a security attack than their Linux counterparts.

Yankee Group asked survey respondents to quantify the amount of time it took IT administrators to restore operations among Linux and Windows file servers, application servers, database servers, domain controllers, web servers and user systems. In every instance save one - domain controllers - the restoration process was quicker for Windows servers than Linux servers. The disparity was most apparent in file servers and web servers. It took administrators 18.3 hours to restore a Linux file server following a security incident compared to a recovery time of just more than 12 hours to bring a Windows server back online. On the client side, administrators needed 17.1 hours to restore Linux desktops versus 12.7 hours to fix a Windows desktop system. Domain controllers were the one device in which Linux servers came back more quickly than Windows: an average of 13.2 hours compared to 13.5 hours for Windows.

Time to recover from a security attack is not necessarily reflective of inherent flaws in the core Linux kernel. Corporate customers said that in many instances, the extra time needed to restore Linux servers was attributable to poor documentation of Linux security flaws.[Emphasis mine.] This necessitates that Linux administrators spend more time searching for the appropriate documentation and available security patches to restore a Linux server. By contrast, customers can check Microsoft's knowledge base and regularly updated list of security alerts and available patches.

Think about the implications of this. Following the Yankee Group's findings, Windows recovers faster because...it has traditionally not worked. Because it has not worked, Microsoft has been forced to write excellent documentation to detail why it doesn't work. Linux, on the contrary, has generally worked, and so (among other reasons, mind you) its documentation isn't nearly as complete. Once Linux breaks as often as Windows does, documentation should improve, and administrators will be able to recover a system faster.

I can't wait!

Commentary: Getting the Facts, Forrester-style

Still browsing through Microsoft's Get the Facts campaign website, and I think I found one of the primary reasons Microsoft is hammering on the "You've already bought us once, why not more?" theme. Take a look at this Forrester chart, available here in its original report, posted (and paid for) by Microsoft.



The primary reason that someone would shy away from using Linux, in other words, is fear, uncertainty, and doubt. The primary reason they'd buy Microsoft? Apparently, based on the language used by Microsoft in its case studies and elsewhere on its site, the answer is the same. FUD.

Now, I don't believe that this is all Microsoft offers. I think they actually create some great products. But I am positive, after reviewing the company's website for various products, that it thinks it's primary value is...more. That is, for a company that has bought some Microsoft, it's even better to buy more. Not because it's better. Not because it's really cheaper. But because it's more.

That's a really lame reason. As I've said, with every "more" of any vendor, you then have commensurately "less" freedom (sovereignty, Massachusetts calls it). I'm not talking free as in Stallman, I'm talking free as in "market," as I once heard r0ml discuss eGOVOS. Choice saves money over the long run, though perhaps not in the short term.

Commentary: Of TCO studies and Microsoft

I may owe Stuart Cohen an apology. A few weeks ago I mocked Stuart for not accepting Microsoft's offer to co-invest in Linux-vs-Windows research. Stuart opined that Microsoft would take one negative sentence from a 99-page research report, overwhelmingly favorable to Linux but for the one sentence, and spend $100 million marketing that one line.

After spending a few days poring through the resources on Microsoft's Get the Facts website, I'm starting to see Stuart's point. If anyone bothered to read beyond the headlines in these reports, including Microsoft's own case studies, they would learn the following:

1. Microsoft's primary value, according to its case studies and websites? Since you're already running something Microsoft wrote, somewhere in your organization, it will be cheaper/faster/better to simply run more of the same. [This is all said with a straight face, as if something revolutionary were being stated.]
   
   
2. Windows is cheaper because Windows support personnel have lower salaries. [This stopped being true some time ago. Or stopped being true to the extent that it mattered. Linux system administrator costs have been plummeting as more enter the market. Supply and demand....Regardless, as the Robert Frances Group found, Linux administrators tend to do more (higher utilization rates) with less (higher number of systems managed by one administrator), making that more expensive ONE Linux system administrator cheaper than the 5-10 Windows personnel needed, or whatever the number would be.]
   
   
3. Windows offers stronger intellectual property protection than many Linux vendors. [True, so far as it goes, which isn't very far. The major vendors - i.e., the kind that enterprises would actually buy from - all have comparable IP indemnification offerings.]
   
   
4. Microsoft suggests that it offers 30% lower hardware costs than Linux in RadioShack's case, but when you actually read how they rationalize that number, it really should read: "Microsoft offers 30% hardware cost savings over its existing configuration of UNIX (server powering the POS device in every store) plus Windows (powering the in-store kiosk in every RadioShack site), with little (actually no) comparison made with how Linux hardware costs compare to Windows. Had they made this comparison, they would have likely found what the Robert Frances Group found, namely, that Linux hardware is much cheaper than UNIX hardware, but also significantly cheaper than Windows hardware (Higher utilization rates, more apps/server). In RadioShack's case, their store-by-store configuration would likely not have been helped by these factors, possibly making Windows truly cheaper than Linux for them. But it is misleading in the extreme to try to extrapolate from them to the wider industry, as Microsoft does in this case study.]
   
   
5. Much of what Microsoft bills as its primary benefits (e.g., Windows XP Embedded "is componentized, meaning that RadioShack can choose to install only the software modules within the operating system that the company requires") actually applies far more accurately and honestly to Linux [Componentized? Hasn't that been a hallmark of Linux since its inception???].
   
   
6. Microsoft likes to talk about the long-term cost benefits of Windows, but this is probably its weakest ground. Nearly every case study (on SharePoint, Windows Server 2003, and so on) highlights the primary benefit of going with Microsoft: you've already gone down part of the road, wouldn't it be cheaper and easier to simply keep going down that road? The problem with this line of reasoning is that the further down the road one goes, the higher the costs become, either measured in direct licensing costs (because you are on the vendor's upgrade path and have to continue to stay with Microsoft's adjacent solutions), or in switching costs. The cost of having no choice is extremely high, and gets higher the longer you stay on that path. This isn't isolated to Microsoft - it's an economic fact that kicks in regardless of which vendor you buy into, if the software is closed (and even if it's not, as Jason will (mostly correctly) argue).

Anyway, I'm preparing a much more detailed analysis of TCO between Linux and Windows. I'm just troubled that Microsoft has invested almost nothing in promoting its products, instead choosing to promote the ties (that bind) between its products. If its best argument is, "You've already bought from us, you might as well buy more," the company doesn't deserve to be in business. I understand that it might make for smart, effective marketing. But it makes for poor truth.

News: eWeek's point/counterpoint on Massachusetts' open file format decision

First, David Coursey tears the decision to shreds (using arguments that I've made here):

In choosing OpenDocument, Massachusetts has selected a format that is not supported by a single currently shipping office suite. It is the native format for a release of OpenOffice that is currently in beta. Massachusetts seems to be placing a lot of faith in this open source office suite. I sure hope it ships....

What happens next will largely be determined by what Massachusetts really wants to see happen. It has created an excuse/reason to drive Microsoft Office out of state offices. If that's the goal, Massachusetts is well on the way.

Of course, once Microsoft is gone, nobody but Massachusetts government workers will be able to use the OpenDocument files they create. Even if Massachusetts goes to OpenOffice, the rest of the world will not (at least not anytime soon). Is it really a good idea to require that anyone who wants to read Massachusetts new "open" documents install and learn to use a new office suite, even a free one?

Sure, the OpenDocument files may meet the goal of openness, but will fall down against a goal of "easy to use by the state's citizens and businesses."

Why? Because these people already have Microsoft Office or one of the programs that will read and write Office files (like WordPerfect). They can also download free Microsoft Office "reader" programs or they can download a free copy of OpenOffice. That's more programs that will read/write Microsoft files than the "open" standard Massachusetts has endorsed.

I am not saying Microsoft's formats are perfect but they are popular and likely to remain so. OpenDocument is not popular and isn't likely to become an overnight sensation. Tying the state to such a format will, in the short term, create the same access issues that going to an "open" format are supposed to solve.

Not to be outdone, however, Vaughn-Nichols writes:

I'm a believer in open formats. I always have been. I always will be. And it's good to see that Massachusetts gets it too.

It's really very simple. With an open standard, like OASIS' OpenDocument, anyone at any time will be capable of using the format. That's true of all open standards, from North America's Type A & B electrical plugs to Wi-Fi networking's 802.11g to OpenDocument for office document formats....

Well, the author of the GPL and president of the Free Software Foundation, Richard Stallman, has said that Microsoft's conditions for its "open" XML lack the freedoms necessary to be considered free software. And Stallman should know, don't you think? [No, I, for one, don't agree....]

So, Massachusetts made the smart choice, the only intelligent choice, of going with a standard that is not ultimately under the control of one company.

Look at it this way, would you buy a car that only ran on Exxon gas? Buy a CD that would only play on a Toshiba-brand CD player? Of course not!

Coursey worries that Massachusetts has "selected a format that is not supported by a single currently shipping office suite."

True, but so what? Nothing supports Microsoft Office 12's formats either.

While I can see both sides, I don't think Vaughn-Nichols represents his very well. Waving a flag and dishing out apple pie is a weak argument....

News: Managing Google's idea factory (BusinessWeek)

Interesting BusinessWeek article on the woman (Marissa Mayer) in charge of monitoring innovation at Google, and ensuring it bubbles up to the top. The article is a bit soft, but reading between the lines, you can get a good idea of life inside the Googleplex, and how good ideas find their way into full-time beta land on Google's site. Here are some snippets:

One of the key reasons for Google's success is a belief that good ideas can, and should, come from anywhere. Page and Brin insist that all engineers in the company have one day a week to work on their own pet projects. An ideas mailing list is open to anyone at Google who wants to post a proposal. What [Marissa] Mayer does is help figure out how to make sure good ideas bubble to the surface and get the attention they need. The task is becoming more complex as Google grows, with a workforce of 4,200 now and revenues on track to hit $3.7 billion this year....

Part of Mayer's challenge is realizing when certain formulas falter. For years she ran the company's Top 100 priorities list, which ranked projects by order of importance. But as Google's workforce grew, the list soared to more than 270 projects. Last year Google execs decided it had run its course, and shut it down. "People don't get attached to the processes themselves at Google," says Bret Taylor, product manager for Google Maps. "It's very unusual. Even at small companies, people tend to say: 'This is the way we do X."'...

Office hours are just one way in which Mayer connects with inventive engineers and managers. Another is Google's ideas mailing list, the e-mail thread to which anyone can submit or comment on an idea. At times, the thread more resembles a form of techie Darwinism. Google newcomers who proffer an especially obvious suggestion ("Why don't we search blogs?"), or something off-topic like how to arrange the cafeteria tables, often suffer withering rebukes. "It's about 50% new ideas, 50% indoctrination of new employees," says Mayer....

What Mayer thinks will be essential for continued innovation is for Google to keep its sense of fearlessness. "I like to launch (products) early and often. That has become my mantra," she says. She mentions Apple Computer and Madonna. "Nobody remembers the Sex Book or the Newton. Consumers remember your average over time. That philosophy frees you from fear."

Commentary: Of boring football (soccer) and software monopolies

I'm an Arsenal fan. I love football (soccer, not that rubbish American football saturated with pads, commercials, and steroids),and Arsenal's style of play - attacking, fast paced, elegant - draws me and tens of thousands of others in.

This season, however, Arsenal has had a slow start and will likely not catch up to Chelski (Chelsea, the now dominant London club which dominate due to Russian oil money, not because of their overly smug coach, Jose). Contrary to Arsenal, Chelski plays with a conservative, overly defensive style which is boring to watch.

Because of this dull style of play, which has seeped from Chelsea into English football generally, attendance at matches is down. People are actually turning their attention to cricket, which is mind-numbingly boring by definition. Chelsea should be drawing the biggest numbers, yet can't fill its stadium.

There are two answers to this problem. One is right, and the other is wrong. The wrong response is Arsene Wenger's, Arsenal's otherwise exceptional coach. It is to look for ways to force other teams like Chelski into playing the kind of football that fans enjoy and that brings out the best in players.

But this approach won't work. Force doesn't work.

The better approach is to beat dull football with interesting, dynamic football. As Arsene points out in the article linked to above, Brazil and Real Madrid (which both play a lightning fast, attacking style) have won the most trophies. They beat boredom with flash and style.

What's the software analogue? People have been moaning about Microsoft's dominance for years. Rather than competing with Redmond, industry players keep looking outside the industry, deus ex machina style, for someone to rescue them from having to compete with a currently dominant player. "We can't compete with monopoly," people whine, "We need the government to intervene." Whah!

And so the US Justice Department is called in. Or bureaucrats in the European Union. And now people are looking to Massachusetts to start a government uprising against Microsoft's proprietary file formats in the hopes that this move toward openness will erode Microsoft's dominance.

As with British football, however, the answer is not to try to impose change from above. The answer is to beat the dominant player.

Clay Christensen's Innovator's Dilemma wasn't about how innovators dominate their industries and then are unseated (the dilemma) by government intervention. No such intervention is needed to undermine the innovator. Instead, innovators are unseated through changes to how an industry operates, and what it values.

Surely this change is in motion now. Google from above (services that run over the web) and Linux from below (changing the economics of how we prefer to buy and sell software) are doing more to harm Microsoft than any government body in a free society will. We don't need to wield a government club against Microsoft. It's a foolish way to respond to temporary dominance, creating both bad law and bad precedent.

No, instead we should simply compete. Microsoft will lose to the extent that outsiders can change the rules of the game, and beat Microsoft at this differently played game. Microsoft will survive and thrive to the extent that it can learn these new rules, master them, and twist them to its advantage. This is how the game is played.

So, let's get back to competing, and stop turning to government to solve our problems. I'm a conservative, you see. I believe people are intelligent and ambitious for success - that intelligence and ambition will produce far more change in an industry than government intervention will. Count on it.

Commentary: The desktop is dead

As Peter Rip points out, the desktop is dead. (Thanks for the link, Paul.) While he highlights a few AJAX-based startups who are working on taking the "desktop" out of desktop applications (Writely and Meebo seem the most interesting to me, though I can't remember the last time I willingly used a word processing program - I've got email, after all), and moving them to the web.

As Peter points out, as I've been arguing here for going on two years

The real problem with desktop apps is no one works at their desktop anymore.

Amen. It's not going to happen tomorrow. It actually happened yesterday. It started the day email became a faster, more effective way to communicate than by laboring over documents, attaching them to an email, and then sending them along. Who should Microsoft blame the day its Office cash cow dies?

Itself.

Say what you want about Exchange, but Outlook defined what an email client should do, and how it should look. Once they added HTML/rich text editing, the writing was on the wall. Now as Microsoft speeds its BlackBerry response it's only hastening the beginning of the end for Office.

Which makes me wonder: maybe Microsoft actually has a great way out of the innovator's dilemma....? Maybe, by beefing up its email offerings, transitioning functionality from Excel and Word (PowerPoint probably always makes sense as a separate application) into Outlook it will end up trumping the industry again.

News: GPL 3.0 and how RMS never, ever gets off-message

The man is incredible. How many years has he been saying the exact same thing, over and over and over and...? It's fascinating. I once heard him speak at an MIT event - he was shockingly dull, repeating the exact same things you can read in his writings. But for all that, one has to admire his tenacity. Whatever you may think of his ideology (I, for one, think it's overdone and unhelpful), you have to admire his resolution.

Do you think that GNU/Linux is so famous (more than BSD, for example) because it comes under a license that defends users' freedom?

I don't know--but the question doesn't seem important. Our goal should be to spread freedom and then defend it. That is more important than making our software popular, which would just be catering to our egos.

Take a quick read through this interview, see how many times you can count the words "free" and "freedom," multiply that number by zero, and you have the number of times RMS has said anything new in the past 20 years. But if you liked it the first time, maybe it gets better with age...? :-)

I guess I'd find more merit in this if I felt that software freedom was on par with the majority of human problems we struggle with each day. It may be a means to solve some of those problems, but not many. That's why I've never been able to get excited about hardline approaches to software, either on the "Left" (free source) or on the "Right" (proprietary source). In the end, it's a question of pragmatics - which works better in a given situation?

Actually, the interview does more than wave the flag for software freedom. It also talks through some changes coming in GPL 3.0, which topic will also be addressed at OSBC Boston in this session.

News: Microsoft admits its development model stinks

The Wall Street Journal had a great article on Microsoft's restructuring. Unfortunately, the paper still lives in the Dark Ages and requires a subscription to the print edition (which I have, but you may not). So I'll try to excerpt enough that you won't have to try to dig around for it.

/rant against ignorance in the newspaper industry

Anyway, the article is built around an interview with Jim Allchin of Microsoft. Yes, the same who famously declared open source to be "Anti-American." He comes off better in this article, but strangely understands the problem (Microsoft's traditional development methodology and accompanying architecture stinks) but simply offers more of the same problem, lightly repackaged, as a solution.

Jim Allchin, a senior Microsoft Corp. executive, walked into Bill Gates's office here one day in July last year to deliver a bombshell about the next generation of Microsoft Windows.

"It's not going to work," Mr. Allchin says he told the Microsoft chairman. The new version, code-named Longhorn, was so complex its writers would never be able to make it run properly....

Ultimately, Mr. Allchin's warning proved cathartic and led to what he and others call a transformation in Microsoft's most important product. A key reason: the growing threat from rivals such as Google Inc., Apple Computer Inc. and makers of the free Linux operating system. In recent years these companies have been dashing out some software innovations faster than Microsoft. Google has grown particularly effective at introducing new programs such as email and instant messaging over the Internet, watching how they perform and regularly replacing them with improved versions.

Microsoft's Windows can't entirely replicate that approach, since the software is by its nature a massive program overseeing all of a computer's functions. But Microsoft is now racing to move in that direction: developing a solid core for Windows onto which new features can be added one by one over time.

I'll admit my ignorance here, and suggest that I thought that's what Microsoft has been doing all along (or, at least, has aimed to do, whatever its success rate). But it's true that while the company has become adept at distributing patches for its never-ending supply of bugs, it has failed to provide actual innovation at the same pace. (And it's also true that this Apache-esque code architecture is a significant upgrade for Microsoft, at least in theory. There's still a huge question whether the company can avoid bloating it, but...we'll see.)

But let's give Jim credit for spotting the problem, and for being brave enough to blame everyone at Microsoft for the problem but himself. This takes courage. (Or something. I find it insulting to Microsoft employees that he seems so willing to push off responsibility to everyone else and the market itself.)

Mr. Allchin's reforms address a problem dating to Microsoft's beginnings. Old-school computer science called for methodical coding practices to ensure that the large computers used by banks, governments and scientists wouldn't break. But as personal computers took off in the 1980s, companies like Microsoft didn't have time for that. PC users wanted cool and useful features quickly. They tolerated -- or didn't notice -- the bugs riddling the software. Problems could always be patched over. With each patch and enhancement, it became harder to strap new features onto the software since new code could affect everything else in unpredictable ways.

The 53-year-old Mr. Allchin, who joined Microsoft in 1990 and is now co-head of the Platform Products and Services Division, says he always disdained the fast-and-loose culture of PC software. The holder of a doctorate in computer science, Mr. Allchin craved discipline in code writing. But in the booming 1990s, when it seemed Microsoft could do no wrong, there was little Mr. Allchin could do. As soon as Microsoft was done with one version it pushed on to the next. Mr. Allchin was haunted by what he calls his "little demons."

In 2001 Microsoft made a documentary film celebrating the creation of Windows XP, which remains the latest full update of Windows. When Mr. Allchin previewed the film, it confirmed some of his misgivings about the Windows culture. He saw the eleventh-hour heroics needed to finish the product and get it to customers. Mr. Allchin ordered the film to be burned.

Otherwise stated, "Well, I knew we were doing everything all wrong, but there was just so much money in it that I couldn't help myself!" The mafia has got nothing on Mr. Allchin and, indeed, the scope of the economic damage done by a Byzantine code base, with interlocking products (and security holes wherever they intersect, no matter how pristine the code Microsoft writes), dwarfs anything any one organized crime family ever dreamed of perpetrating. But Jim really didn't mean it. He wanted to use that PhD. It's just that the money was so good.

Not to despair, though, as Mr. Allchin went to work on solving the problem.

Microsoft's culture was facing a new threat. The mass of patches and agglomerations that made up Windows turned it into an easy target for viruses and other Web-based attacks. Mr. Allchin had to divert top engineers into the effort to fix security problems in existing versions of Windows. "The ship was just crashing to the ground," Mr. Allchin says.

In late 2003, Mr. Allchin called on the help of two men. The first was one of Microsoft's best-known "shippers," people known for their ability to turn around troubled software projects. Windows veteran Brian Valentine had a reputation for booming motivational speeches, beer bashes and stunts like showing up to work functions as Elvis, the Easter Bunny or even once a hula girl with a coconut bra.

The second man Mr. Allchin tapped was Amitabh Srivastava, now 49, a fellow purist among computer scientists. A newcomer to the Windows group, Mr. Srivastava had his team draw up a map of how Windows' pieces fit together. It was 8 feet tall and 11 feet wide and looked like a haphazard train map with hundreds of tracks crisscrossing each other.

That was just the opposite of how Microsoft's new rivals worked. Google and others developed test versions of software and shipped them over the Internet. The best of the programs from rivals were like Lego blocks -- they had a single function and were designed to be connected onto a larger whole. Google and even Microsoft's own MSN online unit could quickly respond to changes in the way people used their PCs and the Web by adding incremental improvements.

Careful observers will note that Mr. Allchin's solution was nothing more to start from scratch...with the exact same methodology that got the company into trouble in the first place. I see nothing new here - just the same way of developing products. (Though, again, starting with a much smaller core platform and extending it with plug-ins/code extensions is different for Microsoft. The question is whether they can maintain that initial discipline. Allchin tells us point-blank that they couldn't before - why should now be different?)

This wouldn't be a bad thing if, as Jason Matusow has said, the process works and yields innovative products. But after spending a full page in the WSJ talking about how the system is corrupt to its core, it's hard to give Mr. Allchin the benefit of the doubt that this time, somehow, it will be better.

But wait! There's a light at the end of the tunnel: Mr. Allchin will soon leave his "little demons" behind (where they will haunt us, instead :-).

This week Mr. Allchin announced that as part of the restructuring he will retire next year after Windows Vista is in customers' hands. In a recent interview he said his demons aren't fully exorcised. "There're weaknesses in everything we're doing today," Mr. Allchin says. "But it's such a huge step up from where we were."

Let's hope so.

Now, to be fair to Microsoft, it does a great job - or used to - of creating interesting software that is useful and sometimes innovative. And despite my cavalier dismissal of its "new" architecture (there I go again), it is the case that the company seems to have built a solid, flexible foundation into which it can plug interesting new functionality without roiling the myriad other pieces that depend on the massive block of Windows/Office/etc. code.

The question is, however, how can it continue to do this without relying on a development methodology that has reaped profits for the company perhaps primarily because it has offloaded its responsibility for creating stable, secure products on customers who didn't have any other options? That is, how much of that $25+ billion war chest would still be there if the company had properly invested in securing and stabilizing its code?

Einstein once said that

Problems cannot be solved by the same level of thinking that created them.

Microsoft seems to have properly diagnosed a problem fundamental to its code development. Perhaps it would now be better off jettisoning the "level of thinking" that created this problem in favor of others (Ray Ozzie?) who could help complete the transition to an open source-like development methodology, which possesses all of the attributes Microsoft seems to want to replicate in its new architecture.

Linux may be an enemy to Microsoft, just as Apache is, and other open source projects are. But open source development, itself, is not. The company has much to learn from it. Let's hope Jason and others are able to help Redmond to think different.

Business 2.0: Nine fads to ignore

From the October 2005 issue, Business 2.0 gives us nine fads (technology and otherwise) to ignore. (In no particular order, mind you.)

1. PODCASTING
Yes, it’s nice to get MP3s downloaded to your iPod. But where are the revenues? Podcasting will continue, but as a business, it’s the latest iteration of CB radio.

2. SPACE TOURISM
Travel to the final frontier is riskier than buying a vacation home in Aspen. When the first billionaire perishes in the icy void of space, it’ll take years for this nascent industry to bounce back.

3. OUTSOURCING
It works for rote or clerical tasks, but creative and management functions are still most effective close to home. If your job hasn’t been outsourced by now, it’s unlikely to happen in the years ahead.

4. DOWNTOWN CONDOS
In cities like Chicago, Miami, and San Diego, short-term speculators snapped up many of these “entry-level” homes. With price-to-rent ratios at absurd highs, it’s time to head for the exits.

5. CHINA RULES THE WORLD
Don’t forget your recent history: Japan didn’t surpass the U.S. economy in the 1980s, and it’s unlikely that China will in the next half-century.

6. HIGH-DEFINITION DVDS
Which will it be: HD-DVD or the incompatible Blu-Ray format? Get ready for a rerun of VHS vs. Betamax -- if, that is, consumers want high-definition DVDs at all.

7. HYDROGEN-POWERED VEHICLES
The technical challenges are huge, and they will probably take 10 to 15 years to overcome. Until then, you can safely ignore hydrogen hype.

8. PRIVATE-LABEL MOBILE-PHONE SERVICES
Lucrative niches are few and far between, and the best ones (such as teens or sports fans) are already taken. Whatever’s left is likely to be too small or too hard to reach to turn a substantial profit.

9. CELEBRITY CLOTHING BRANDS
Sean “Diddy” Combs made celeb-branded clothing fashionable. Beyoncé, Jennifer Lopez, and Gwen Stefani followed. Now comes proof that this fad will fade: Jessica Simpson has launched her own line of plus-size jeans.

And just when I was going to buy some Britney jeans....

McAllister: Can open source companies innovate?

Had Neil McAllister asked, instead, can open source innovate, I might have been more gracious to his argument. But he effectively answered his own question simply by adding the word "companies."

McAlllister, a bit dated on his understanding of open source business models (It's not his fault - he hasn't attended OSBC yet. Well, that is his fault, but it's one of which he can repent :-). Open source stopped being solely about "professional open source" well over a year ago. That's no longer the primary business model. It's never been the most interesting, though good companies like Red Hat and JBoss make good money from it.

Having lumped open source together under the professional open source model (i.e., a support model), McAllister then rushes to (somewhat accurately, I think) analogize to the insurance industry:

You can't make money giving away products. You can, however, profit by selling support and services around those products, and that's the way many open source companies, including JBoss, are run. Customers can download the code for nothing, but if they want somebody to call when things start falling apart, they have to pay.

The interesting thing about this model is that when you strip it down to brass tacks it looks an awful lot like an insurance business. Customers pay for a number to call even though they hope to never dial it. The problem is that in practice these support contracts don't work like health insurance; they're more like dental insurance.

With group health insurance, everybody pays but only a few get sick. As long as the majority of group members are young and healthy, the insurance provider still turns a profit, even if a few members file major claims.

Dental plans aren't structured quite the same way, for one main reason: With dental insurance, everybody files. What's the first thing you do when you join a new dental plan? Get your teeth checked. That's why the deductibles are so high and the annual caps so low. Dental plans are structured to let every subscriber get two check-ups per year, in hopes that regular preventative care will ward off bigger problems. Anything extra comes out of the insurer's bottom line.

Pure open source companies are in the preventive maintenance business, too. Providing a Linux (Overview, Articles, Company) distro, for example, means rolling out security patches, providing installation support, running compatibility tests, producing documentation, pressing media. And you bet every customer takes advantage of these things -- if they just wanted the software, they wouldn't have paid.

This is all true, so far as it goes. But it doesn't go very far. Novell, SugarCRM, IBM, Google, MySQL, and a range of others have very different business models for open source, none of which depends detrimentally on the support equation.

But even if we do lump all open source business models together, I'm not sure how this necessarily leaves us without innovation. McAllister doesn't really address this. Even if JBoss makes its money on support, how does this prevent it from investing in cutting-edge, innovative software? They have a way to monetize that innovation, and a license that effectively prevents others from taking it. (Competitors are unlikely to fork GPL'd code, for example.) Again, it's a question of offering "source of code," rather than "source code" (intellectual property).

I'm not arguing that open source companies, in our very young history, have been overwhelmingly innovative. But given that it's only been in the past two or three years that we've begun to believe that open source can be a highly profitable, revenue-generating business model, this isn't surprising.

Give us another (very) few years, and I think McAllister's well-intentioned article will be relegated to a museum, where we stand around and smile at the quaintness of the ideas expressed. "How old fashioned!" we'll say. (That's not to say that every company will be open source. They won't. But we will discover that both closed and open models offer benefits, and both will co-exist.)

Commentary: Learning from Apple's design maestro

Mark Watson sent this link on to me from the Design Museum. It's an interview with Jonathan Ive, Apple product designer responsible for the iPod, iMac, and a range of other Apple products. He offers an interesting take on innovation and design, including the following:

Q. [W]hy are so many new products so bland and derivative?

A. So many companies are competing against each other with similar agendas. Being superficially different is the goal of so many of the products we see. A preoccupation with differentiation is the concern of many corporations rather than trying to innovate and genuinely taking the time, investing the resources and caring enough to try and make something better.

Commentary: Why we need OSBC

George Ou was recently drawn-and-quartered for daring to suggest than an open source project had weaknesses. Matthew Szulik, CEO of Red Hat, got much of the same when he suggestedconsumers should look to Windows, instead of Linux, for home use.

Open source needs to get over this infallibility complex. Someone, somewhere in the open source world decided that absolute perfection was a requirement to be worthwhile. Now, if anyone suggests that open source is less than pristine, they get flamed by a horde of open source bigots.

Question: why is open source bigotry somehow better than closed source bigotry?
Answer: It's not.

It's the same, and perhaps worse, because it inhibits reasonable CIOs (and very few of them are technology zealots who will use a product because it's Right in some religious sense) from adopting open source. Those that do try it out expect it to meet the high standard (perfection) proposed by the zealots, to which it can never live up.

The Open Source Business Conference was set up to address the open source information gap. Microsoft, Novell, and IBM were the three early sponsors for the event, pulling three different vendors with different views of the software/services world. That "Switzerland"-esque mentality continues to pervade our sessions. It's not a place for zealotry - it's a place to get real information on where open source is going, where it's not, who is using it, and how.

If you haven't already registered (and claimed your Sony PSP, which we're giving away free to all who register before the end of the month), I suggest you do so. OSBC is where Optaros, EnterpriseDB, and a range of other startups got at least some of the inspiration for their growing businesses. (When I go back to review the attendee lists from our first two events, it's amazing how many of today's startups were there, still in stealth or pre-birth. Very cool.) OSBC helps to create, in other words, the future of open source business. No bigots need apply.

Kedrosky: It's official - Gmail stinks

Paul actually says it much nicer than that, and leaves out hope that Google's Gmail will get better. I'm sure it will. I'm just not sure why pundits and the Valley have made such a fetish of the company (as I complained back on Halloween 2004). In the meantime, it takes someone like Walt Mossberg to finally get people to stop foraging at the Google trough - the company does search exceptionally well (and Google Maps and Google Mobile are also cool). But most of the rest of the company's work? Rubbish.

Lessig: Google sued by angry authors everywhere....

Well, not really. I think the latest lawsuit to be an incredibly naive, myopic suit, led not by authors but rather by the oafs that represent them. But then, I've come to expect that from the organizations that supposedly represent the rights of the creative class. (Also see Tim's thoughts on the lawsuit.)

Content is thrives in ubiquity. John Steinbeck (or JK Rowling) loses nothing, and gains everything, when his work is read, cited, and discussed by more and more people. (Or, in Steinbeck's case, his estate benefits.) It's not about milking a copyright - that "right" is only as valuable as the volume of people that actually care about one's work.

It's no different from a startup: the IP that company generates is worthless until someone buys the product/solution it offers. Startups have a problem with visibility and distribution - protecting IP should not be their biggest priority. Guarding intellectual property is a way to guard an incumbent position, but is not a good way to grow that position. Growth is about ubiquity, and ubiquity thrives in open access.

I often hit on this point when presenting at conferences. Below is one of the slides I use. Speaking to the slide, I generally point out that both the music and movie industries have completely failed to leverage new technologies, fighting them every step of the way (as they seek to hoard the cash position they already have, not appreciating how much bigger that position would grow if they'd simply adopt new business models/technologies).

Fortunately, as with the music industry's resistance to the compact disc, they generally lose these battles. In the case of the CD, they kept royalties to artists flat (arguing that it would cost them a great deal to market the new medium), raised prices ($8.95 to $15.95), and consumers bought a lot more of them, effectively ballooning their market by BILLIONS of dollars. They were fortunate that technology ran roughshod over them in this case - they'll hopefully be fortunate again to have P2P run over their wimpy business models and force them to grow up, as I've argued elsewhere.

Here's the slide:

Berlind: Linux desktop? How passe

David Berlind has an exceptional blog post on the Vista versus Linux desktop debate. His verdict (one that I've been championing for some time. In response to a question at the Swarthmore presentation I gave last night, as to whether I personally saw a place for Linux on the desktop, I responded that while Linux will play an increasing role on the desktop, primarily outside the US and Europe, the "desktop" as such is already in decline.

Mobile and software-as-a-service will increasingly pillage the application territory we've come to associate with the desktop, as the notion of productivity (communication via email versus communication via Word documents) changes. Who cares what OS is running on my laptop (unless, of course, it's a Mac, and then I care deeply, because I like pretty things :-)? I care about the services my device - very mobile (phone) or somewhat mobile (laptop) - enables me to access, not the OS that bridges the divide between hardware and application.

Anyway, Berlind makes this argument and more, first dismantling the idea that Linux is a disruptive force on the desktop:

To really motivate people to change, the technology typically must be so disruptive to the status quo and the benefits so obviously undeniable, that people and organizations can't not consider the change. On the server front, Linux has fit that bill. But on the desktop front, about the best we can argue is that Linux is keeping Microsoft on its toes (as desktop OSes go, Vista is going to be pretty darn good, if you ask me). At the end of the day, they're both operating systems. But, at the end of the next major disruption, the fact that they are operating systems -— thick ones at that (by the time the end-user is fully empowered) -— will have been their downfall.

That's because desktop operating systems are the equivalent of an on-premises backoffice solution in what is increasingly becoming an off-premises, on-demand world. If the meteoric success of software as a service (SaaS), application service provider (ASP)-delivered solutions like salesforce.com and RightNow Technologies teaches us anything, it's that to really get people to do the double-take that's a pre-requisite to switching, a solution has to turn the status quo on its ear. And that's exactly what salesforce, Rightnow, and others like NetSuite and Authoria are doing.

He then calls out the likely "desktop" players of the future: Google and IBM. Have a read to see why.

News: CIO's negative on Microsoft's subscription-based pricing

It's hard these days to imagine anything more pre-historic than enterprise licensing. At least, that's what the momentum behind Salesforce.com, SugarCRM, MySQL (i.e., Network), etc. would have us think.

Microsoft arguably brought this model to enterprise IT, and yet it continues to fail to participate. Why?

According to a group of UK-based CIOs, as reported Silicon.com, it appears that while Microsoft is bullish on getting paid constantly, it's not as bullish on the other trait that normally comes with subscription-based pricing: "cheaper."

Kirk Downey, CTO at Centrica, said Microsoft has failed to address the root problem organisations have with SA - the cost - and that it compares unfavourably to programmes run by the likes of Oracle and SAP.

He said: "If Microsoft is to position its SA programme alongside annual spend for maintenance on other critical technologies, it needs to demonstrate the economic value of the investment and introduce more flexibility in the commercial model to allow enterprise customers to negotiate this expense into their planning cycles."

Sean Powley, head of IS strategy at the London Borough of Barnet, said: "It still looks like Microsoft wants to have its cake, eat it and charge users to watch. Where's the flexibility in the new arrangements for complex organisations like local authorities?"

Dave Rosenberg: The British Computing Society gets open source wrong...

Stroll over to Dave's blog to read about the British Computing Society's completely misguided understanding of open source. It's hard to believe that they could get it so wrong. As an American, I've come to expect the British accent to connote intelligence or evil (a la Star Wars). Guess not the BCS in this case...

A few lowlights that Stephen [quoted by Dave] pointed out:

Intellectual Property: A major flaw at the heart of the open source movement is the misconception that most individuals actually have the legal right to contribute their intellectual efforts to OSS projects

Huh? Distribution vendors, system vendors, and open source foundations all give explicit permission (or even require) that open source work fall under an existing open source IP policy.

Conceptual Integrity: The process of creating software is more akin to an engineering discipline than an artistic endeavour, and this raises another point of concern with OSS. Like any engineering design project, good software needs a designer (or software architect in the current industry jargon) with a clear design concept which must be adhered to rigorously otherwise the software becomes progressively messier as it is developed in a piecemeal manner.

Most commercial software is driven solely by schedule. Nothing ever gets worked on unless it is a bug, or a new feature for money. Any other work is a drain to the bottom line. Open Source is continually getting re-examined and cleaned up. The internals of many parts of Linux change over time to keep a clean framework and keep maintainability. Commercial software rots over time and becomes a maintenance burden "sorry we can't add that new feature since it means changing the ISAM layer, and the guy who understood that left 2 years ago for Google."

News: IBD on open source

Good article in a recent Investors Business Daily on open source. A few gems:

It's easy to see why Marc Fleury's numerous enemies accuse the 37-year-old CEO of having a god complex.

By his account, his 100-person software firm is responsible for a 40% drop in shares of BEA Systems (BEAS) over the last two years. It has sparked at least one IBM (IBM) acquisition. And it forced Sun Microsystems (SUNW) last spring to give away the software recipe to its Web application software.

His software competes with much larger players, promising most of the same features for free, or pretty close. In May, IBM struck back. It bought Gluecode, a company founded by some of Fleury's former employees.

"I'm not afraid of IBM," he declared over noodles one day last spring in San Francisco. "IBM should be afraid of me."
It's hard to tell whether Fleury, head of open-source middleware maker JBoss, is bluffing, delusional...or right.

And, importantly, the article draws out the customer-vendor intimacy that John Powell, CEO and co-founder of Alfresco, talks about (and which I blogged about recently):

Still, open source does have its advantages. Because the code is open, customers can pinpoint problems and provide clear feedback.

"Going out and doing focus groups to find out that your customer wants blue buttons on the screen instead of red buttons is useful," said Novell (NOVL) executive Matt Asay. "But it's much more useful to have your customers creating those buttons beside you." Novell, which started as a conventional software maker, bought SuSE Linux last year.

In addition, most open-source firms rely on word-of-mouth referrals and distribute over the Internet. By the time users contact the company, they're "qualified leads," ready to pay for support and add-ons.

In theory, it all adds up to faster development and lower costs -— a sales pitch that never gets old.

Amen.

Talk about flattening....

Amazingly, I've somehow become the #3 (measured by Google) spot on the web to find the humorous (in a bitter sort of sense) photo of President Bush fishing the aftermath of Katrina. You can find it here, though thousands have already found it without this second pointer.

How?

I can't describe it except to point to Friedman's "flattening" concept. It's not like it's much of an honor to be host to a picture that belies my own general feelings on President Bush. The thing that fascinates me is that somehow it became me, with a few simple HTML tags and a few stray people linking to it, who linked to others, who linked to others, and so on.

I spoke to a group of Swarthmore students tonight on free culture and open source. (This was easily one of the most intelligent, well-informed groups I've ever addressed. It made presenting a pleasure. I was genuinely surprised by how perceptive they were. Not like law school, where people were plenty smart but ruined it by pretending to be much smarter. These students were awesome.) Anyway, as part of my presentation, I waxed a bit Lessig-esque, telling them that if their idealism wasn't fueling the open source/uploading companies of tomorrow, someone else's would. It might as well be theirs.

Maybe that's the secret behind this silly Bush photo example and of open source, generally. Sure, someone else can do it. But in our flattened world, why not me? Why not you?

Talk about flattening....

Amazingly, I've somehow become the #3 (measured by Google) spot on the web to find the humorous (in a bitter sort of sense) photo of President Bush fishing the aftermath of Katrina. You can find it here, though thousands have already found it without this second pointer.

How?

I can't describe it except to point to Friedman's "flattening" concept. It's not like it's much of an honor to be host to a picture that belies my own general feelings on President Bush. The thing that fascinates me is that somehow it became me, with a few simple HTML tags and a few stray people linking to it, who linked to others, who linked to others, and so on.

I spoke to a group of Swarthmore students tonight on free culture and open source. (This was easily one of the most intelligent, well-informed groups I've ever addressed. It made presenting a pleasure. I was genuinely surprised by how perceptive they were. Not like law school, where people were plenty smart but ruined it by pretending to be much smarter. These students were awesome.) Anyway, as part of my presentation, I waxed a bit Lessig-esque, telling them that if their idealism wasn't fueling the open source/uploading companies of tomorrow, someone else's would. It might as well be theirs.

Maybe that's the secret behind this silly Bush photo example and of open source, generally. Sure, someone else can do it. But in our flattened world, why not me? Why not you?

News: Yahoo! moving beyond content aggregation to creation

So, this is old (i.e., a week) news, but I didn't see it until this morning in The Financial Times (Subscription required, for some asinine reason, given that they make almost no money from online subscriptions and much more on advertising - got that directly from them while in the UK) before boarding a flight to leave London. Yahoo! has hired a veteran reporter, Kevin Sites, to be a one-man war correspondent. Smacks of corporate "uploading" to me. Yahoo!, as a way to boost youth readers and circumvent the news gathering system, is going direct and uploading its own content in a more informal medium. Good for them.

Open source = uploading (?)

I liked his book, but Tom Friedman is turning out to be equally interesting over email. Perhaps more so, because it's nice to see that he's human and has the odd typo now and then. :-)

Inspired in part by Kevin Kelly's article in Wired, Tom is seeing open source less about downloading source code and more about the ability to upload. In the "uploading" world, open source is just part of a larger trend. Blogging is another part of this trend, flattening out avenues to publication. I no longer need to be a New York Times columnist to be heard (though, of course, I'd still need to deliver similar-quality product to get my name out such that an equal number of people would follow my blog. In open source and blogging, crap is still crap. Cream (rising to the top) is still cream. The only difference is in the lower barriers to try.).

I like the uploading metaphor. Though many (most?) open source users never upload, as it were, the point is that many more do now than did before. It's a trend toward bottom-up development, just as blogs are.

What open source needs to do is replicate what blogging has done: my mom blogs (we have an Asay family blog). She doesn't write open source code. But she, and I, likely could if someone would find a way to lower the bar (cartoon-object-oriented programming?) to develop code or by simply finding roles in open source development communities that non-hackers can fill (like usability testing).

Do this, and open source will become a true upload phenomenon, rather than the (mostly) cheap download phenomenon it is today.

Commentary: Tom Friedman inspires a long rant on (open source) developing nations

In trying to wrangle Tom Friedman, author of The World Is Flat and I have been exchanging emails in (my) hope to have him keynote OSBC San Francisco. (My review of his book is here.)

Unfortunately, he's traveling then (don't ask where - it's a sign of how flat the world really is that I don't even know which planet the city he's visiting is on :-). In his last email, he said something that really hit me, touching as it did on what Nicholas wrote recently:

And how are we going to help those poor people so they are not going to be poor anymore? Through Linux or Bono? Through open source or Live Aid?

I had a long train ride today, and I thought about this statement the entire way.

We, the "developed" world, tend to think of poverty, AIDS, etc. as things that a rock star benefit concert can solve. In other words, we believe they're somewhat finite issues: Country A needs $100 billion. Find that money, and all of their problems are solved.

What we do, instead, is fund a small fraction of that money, which does no more good than providing the entire lump sum would. Not to be trite, but it's the old "give a man a fish" cliche. Unfortunately, giving a man Microsoft Office is not the same thing.

Why? Because Office is owned by someone else. It doesn't matter so much that Office is owned by Microsoft - I'm not sure that there is anything necessarily benevolent or malevolent about that particular company.

No, the point is that it's owned by someone else, over there. It's true that in a flattened world, developing economies have greater and greater opportunities to participate in the more advanced, developed economies. And that's a good thing. But the greater benefit seems to be that with open source software, they can own their customers from the beginning. Not as indentured servants through a consulting/off-shoring contract to some BigCo in a developed nation, but rather as owners of the solutions they sell locally.

"But," you respond, "You're an idiot, Matt! With open source, they own NOTHING. Therefore, anything they develop is necessarily communal property, leaving them no better off than before."

Not true. For one thing, not all open source software is licensed under a GPL-type license. Vendors can build IP around open source, if they wish, in a BSD-styled world. Even where the license is the GPL, I've argued elsewhere that the GPL is actually a robust capitalist tool, for a variety of reasons, not the least of which is the "source of code" argument that I have advanced at OSCON 2004 and elsewhere. (In short, open source companies can charge a premium to the extent that they "own" the development team, even if the code is open. Hence, MySQL can charge a premium for its database, even though other companies could fork its code.)

But most importantly, open source software enables the leveling of IT to match an economy's ability to pay. Nigeria no longer needs to export naira (its currency) to US, European, Ghanian, Japanese, etc. firms. It can build up its IT from the ground up, with advanced open source-based solutions, paying as much or as little as is reasonable for its particular market.

So, the question is not how can these Nigeria-based companies become multi-billion dollar corporations. Instead, it's how can they become multi-billion naira corporations, one naira at a time.

And, in these places, in this context, it really doesn't matter if OpenOffice is as good as Microsoft Office (or insert your open source versus proprietary matchup of choice). That's comparing apples to oranges. It only needs to be good enough to earn naira in Nigeria. (Remember how bad Microsoft Office and WordPerfect were in their first incarnations? OpenOffice is a Rolls Royce compared to the beat-up Ford Pintos those products were....)

Open source, in short, can allow an economy to develop at its own pace, self-contained. Once those local companies have excelled at serving local markets, they're ready to expand internationally.

Just as the US, Europe, Japan, etc. have done.

Announcement: Dirty little secret (OSBC)

Most people don't know this, but we will generally comp a free pass to qualified IT executives at OSBC. It's normally $1,495.00/person, but we feel that our CIO attendees pay for themselves many times over by enriching the discussion at the event. We have a strong roster already (with CIOs/senior IT people from Fidelity Investments, Priceline, M1 Global, Fannie Mae, JP Morgan, and a range of others already confirmed), but if you would prefer to spend your IT budget on half of a new Apple PowerBook (and who could blame you?), send me an email and let's get you registered. masay a t osbc.com.

Sorry. All others pay cash. :-)

News: Gates on Ellison

Ah, some blood in the water. Actually, this CNET interview with Bill Gates is funnier than it is acerbic.

In an interview last week with CNET News.com, Microsoft Chairman Bill Gates reflected on Oracle's planned acquisition of Siebel Systems.

Oracle CEO Larry Ellison "forecast big consolidation, and he wanted to see that come true, so he's making it come true. It's a brilliant forecast. If the next three people under you don't write code but they do deals, what do you get? You get deals. They will probably do more deals than anybody, and we'll write more code than anybody. We've been very serious about our CRM (customer relationship management software) plans, including scaling that up to very demanding cases," he said.

I think what Gates says is true, at least so far as it relates to corporate actions reflecting management backgrounds. It's very hard for a company to be other than what its most senior people allow it to be. In Oracle's case, it means the company will acquire its future. In Microsoft's case, it will write code more than it will buy companies.

I'm not predicting that Microsoft will be the more successful as a result (still lots of debate on the quality of all that code Microsoft writes), but I do personally prefer the internal building route (having been burned by acquisitions in a former life at Lineo). I think it's a better model.

Commentary: Jonathan was credible until...

...he started using words like "momentous," "phenomenal," "blockbuster," and other hyperbolic words to describe Sun in his blog. Too much, Jonathan, too much, even if it were true. (The truth, of course, is that however super-wow-amazing your hardware becomes, customers are fine with 'good enough' Intel servers. Sorry about that.)

News: Microsoft's downward spiral?

Interesting article from BusinessWeek on Microsoft's declining culture. The gripes?

Employees' complaints are rooted in a number of factors. They resent cuts in compensation and benefits as profits soar. They're disappointed with the stock price, which has barely budged for three years, rendering many of their stock options out of the money. They're frustrated with what they see as swelling bureaucracy, including the many procedures and meetings Chief Executive Steven A. Ballmer has put in place to motivate them. And they're feeling trapped in an organization whose past successes seem to stifle current creativity. "There's a distinct lack of passion," says one engineer, who would talk only on condition of anonymity. "We're missing some spunk."

Of course, it's hard to put too much credence in this when the company's revenues (albeit not its share price) continue to boom, with net income keeping pace:

o question, most companies would kill to have Microsoft's problems. It's comfortably the most profitable player in the tech industry. And it's making more money than ever, with net income of $12.3 billion on revenues of $39.8 billion for the past fiscal year. Its twin monopolies, the Windows PC operating system and the Office suite of desktop applications, give it important advantages when it thrusts into adjacent markets, such as server software for corporations and instant messaging for both businesses and consumers.

Ballmer maintains that the company is in terrific shape. In an interview in a Las Vegas hotel, he says one of Microsoft's strengths has always been its culture of self-criticism. What's different now, he says, is that the internal debate is spilling out into public view because of blogs and e-mail. He says internal surveys show that 85% of the company's employees are satisfied with their jobs, about the same level as in past years. "We have as excited and engaged a team of folks at Microsoft as I can possibly imagine," says Ballmer. "(Employees) love their work. They're passionate about the impact they're having on customers and society. (The 85% number) is a real, real powerful statement about where our people are."

Still, one of the company's biggest weaknesses is also its biggest competitive strength: integrated innovation.

While Microsoft's internal reformers don't directly criticize Gates, they're frustrated with the sluggish pace of product development. As the company's chief software architect, Gates bears that responsibility. He's the author of a strategy called "integrated innovation." The idea is to get Microsoft's vast product groups to work closely together to take advantage of the Windows and Office monopolies and bolster them at the same time. But with so much more effort placed on cross-group collaboration, workers spend an immense amount of time in meetings making sure products are in sync. It "translates to more dependencies among shipping products, less control of one's product destiny, and longer ship cycles," writes Dare Obasanjo, a program manager in Microsoft's MSN division, on his blog.

Maybe. Time will tell.

News: Indonesia standardizes on Linux

Maybe Microsoft should focus less on writing software, and more on swaying elections....

Indonesia has announced that it's standardizing on Sun's Java Desktop System, customized for Indonesia's culture. It's part of the country's IGOS (Indonesia Goes Open Source) program. One more domino....

Commentary: Need more blood in the water

It seems to me that the IT industry has become a little too polite, a bit too staid. It used to be that you could count on McNealy ripping on Gates, ripping on Ellison, ripping on...his shipmates. Now we all try to be diplomatic, whatever we might say about each other behind others' backs.

Politesse is nice (as Morrissey might sing), but politesse can stop us from having an open, candid (Habermas-style) debate. I'd rather see some blood in the water than to have everyone hug each other into oblivion.

But then, maybe I'm just following the league of my football club's coach, Arsene Wenger, who feels that British football (namely, Chelski, who manage a goal per game for the hundreds of millions they've put into the team). As Arsene noted here:

The studious Frenchman has observed a shift in English football from an emphasis on attack to defence, in what he believes is a negative stance taken by too many coaches.

Statistics show that top flight games this season yield an average of just two goals, which is one of the lowest in Europe.

''I'm very concerned,'' admitted Wenger. ''I feel the attraction of English football has always been positive but the trend is to attack less.

''If this is to get into the brain of football it will go downhill quickly. The cautious approach exists everywhere in The Premiership.

''If everyone refuses to play we will get nowhere. Teams are trying not to lose and there is less risk in English football.

Agreed. IT isn't just a game, true enough. But a little honest contention would be good for the industry.

Commentary: Open source in the developing world (Nicholas/Yoda)

Nicholas left this (see below) as a comment to a posting I'd made. I think it's worth reposting here, for those (probably most of you) who missed it. A passionate, reasonable argument as to why open source matters in the developing world. You can find his comment, as well as my original post, here.

(Sorry I took so long to reply to this. I've been unwell the past couple of days)

I unfortunately do not have [m]any immediate references that I can refer you to. In truth it does appear that no one has done a significant amount of research regarding 3rd world development in this space. This led me to realize that I (and those with similar thoughts) may have to be the one who will to write the papers and do the research in this field. (of course, it may be that I'm not as widely read as I thought!)

This lack of reference work has led me to take up the task of writing a number of papers in this regard. I am therefore, in the process of getting in touch with Prof. Celestous Juma of Harvard who is a leading thinker and has done significant research on the technology, science and development mix. In addition, I'll be getting in tough with a number of Chinese IT luminaries (including Dr. Dato Lee-Cheong) involved in Open Source so that I can get a real glimpse into the effect that Open Source has on development.

Open Source is not a sufficient condition and will not, as you said, raise 3rd world countries from the dust. I believe that it is, however, a necessary condition. Recently, the talk by Andrew Zolli, has convinced me even more.

I feel that only open source can facilitate the type of grassroots development that is required to 'evolve' entire geographic regions. The gratis aspect is important but I believe that it is less important than code visibility.

The reason for this is that code is a description of a solution to a problem. I forget who it is who said that "the chief purpose of code is for other humans to read. It is a side effect that it runs on a machine...". In this regard, code takes on the characteristics of great literature or scientific papers. The more our world becomes virtual and we rely on software to solve even more problems, the more we describe solutions through code.

The logical end to this is that we end up having an even greater body of 'literature' and scientific work in the form of code. If this is the case, then we now find that freely sharing this code and allowing it's amendment, annotation, modification and\or redistribution is key to development of a society.

This, in turn, has a very real implications for the development of 3rd world countries. The positive changes that Brazil and China are making are fuelled, largely, by open source and "open innovation". When a society begins to examine its problems and is then free to discuss (through Code) the solutions to these problems, a very real network effect begins to occur. This effect, that begins with the geeks, is the very genesis of the grassroots development that then burgeons.

This isn't as utopian as it sounds. It is actually happening in Brazil! Open Source gives a society an opportunity to explore their problems at a grass roots level that the proprietary way simply cannot.

I realize that I seem to be theorizing quite a bit but rest assured that the academic work will follow shortly and I shall keep you updated every step of the way.

P.S.. I had also been informed about the $100 laptop project. It is a noble idea and it's sustainability is an interesting topic for debate. I'm not yet convinced that we have the connectivity to facilitate the success of such a project. However, something is better than nothing and any effort is welcome. What I'd love to see, is the freedom to tinker being promoted through this project.

I'm reminded of stories I was told of 12 year old Chinese students making cheap digital watches. This is the kind of bottom up learning that must be facilitated for a project of this nature to succeed.

Commentary: Of British football and enterprise software

Reading through The Times today, I chanced across this article on the decline in fan attendance. Apparently, British football has something in common with enterprise software:

High ticket prices, kick-off times changed for television, increasing numbers of live televised matches and less competition for the top three places in the Premiership have been blamed as attendances drop significantly for the first time in nearly 20 years.

Being slightly liberal with analogies, it looks like:

• High ticket prices = High license costs
  
• Schedule changes for TV = A misplaced view of who one's primary customer should be (The fan willing to pay to attend, not the media conglomerate who wants to advertise to the casual fan who sits at home (OK, I was stretching that one);
  
• Live televised matches = Open source software eating proprietary software's free lunch;
  
• Less competition for the top-three slots = Consolidation of the biggest few players, leaving customers (and would-be startups) little choice

Something has to give (in football), just as it has in software.

Talking with John Powell (CEO, Alfresco) today, I had my faith in open source reaffirmed. I hadn't expected that. You may remember that I once offered John a backhanded compliment, calling his (and co-founder John Newton's) credentials fantastic, except when it came to open source. The implication was that he and Newton were simply latching onto the open source train because it appeared to be an easy way to raise money.

I was wrong.

He and Newton didn't start the company as an open source startup. History essentially forced them that direction, as it will continue to push others to their same conclusion. In their post-Documentum and Business Objects startup, post-Bubble, they found it increasingly hard to close big-ticket license sales, despite having a fantastic product with cool, innovative technology. (Be quiet, Stephen.) Some of that innovation may have even been a bit too bleeding edge, and the entry cost was simply too high to allow for would-be buyers to take on a multimillion-dollar trial.

They discovered, as they looked at open source, a number of things:

• Subscription-style pricing enables a perpetual beta (a la Google) mentality, for both buyers and sellers. This is good for vendors, because it means that they have some leeway to put enough functionality on the market such that prospective customers can download it, try it out, and give feedback on what the "finished" product should look like. For the buyer, it allows them to help guide development before the product is fully baked and, hence, less susceptible to change in that particular buyer's favor.
  
• For both buyer and seller, open source builds intimacy. This is a G-rated blog, so let me explain (as John explained to me). The intimacy John noticed was a co-development, co-trial and error, between vendors and buyers. While we say that buyers simply want to, well, buy, this isn't true of every buyer. There are the alpha-geeks in many organizations who, if the core technology is good enough, are interested and available to help develop a product into something they'd want to buy. This makes for a huge competitive advantage over the strictly proprietary vendor, who is forced to develop to completion, hope that someone buys it, and then tweak it if they don't. This process wastes time and money. Open source is more efficient.
  
• Large, enterprise software packages generally require cross-organizational buy-in within an organization. Open source provides a way for products to get adopted in more of a bottoms-up manner, without massive groupthink.
  
• Related to all of the above, open source breaks down the roadblocks vendors put up to keep customers away from their core developers. Think about this. If a customer has a problem with a vendor's software, our first tendency is to route it through their sales rep, who is perhaps the most empathetic but least competent to fix the problem. If they call tech support, they don't meet anyone competent to resolve their problem until they get past levels one and two support. Finally, at level three, they might actually be talking with a developer. In the open source model, contrarily, the customer can reach the developer from the first. It is a good thing to let customers participate in the development process. It creates better code.

Anyway, a fascinating meeting. Alfresco may be new to the open source world, but they arrived here because the proprietary world they thrived in was no longer working for them. Welcome.

Now could you drop those prices on my Arsenal tickets?

News: Yahoo! Mail doesn't stink (as much) anymore

Dan Farber has a post today on the emerging Yahoo! Mail. The fruits of the Oddpost acquisition are finally paying off.

Unfortunately, Oddpost was a year behind CanyonBridge from Day 1, and Zimbra has passed them, too. Will Zimbra or CB get acquired? Maybe. But I can't see either team (or their VCs) selling out as quickly as Oddpost did. Oddpost had angel backing - Zimbra and CB have Benchmark, North Bridge, and Venrock behind them, respectively, none of which are going to be happy with the $33 million base hit that Oddpost scored.

Anyway, Dan writes:

Yahoo is beta testing a new version of its free Web-based e-mail, and it's looking more like Microsoft Outlook and adding similar features. Think of it as a richer Web client that performs almost like a desktop client. Lots of people are familiar with the Outlook interface, and a free service with similar basic functionality and user controls makes the WebOS more real. Gmail, Hotmail and others will follow suit or come up with something better, but it's difficult to get away from the conventions (folders, windows, icons, etc.) that have been accepted since the Macintosh was introduced in 1984.

Best venture firm in Europe?

I think that title has to go to Index Ventures, home of Danny Rimer and a portfolio that includes Skype, MySQL, Zend, Trolltech, Jaluna, and others. Index has perhaps the most interesting portfolio of any venture firm I know.

And great people. I met with Danny this morning to learn about interesting new startups in Europe. (SecureWave is one, btw. Permissions applications and devices that can touch a corporate network or one of its nodes (laptop, desktop).) It became clear that a great deal of cool companies are founded here - fewer, unfortunately, get funded, as Matthew Langham has lamented before.

I was especially intrigued by Danny's interest (and investment) in Trolltech. I've known the company ever since Lineo partnered with it on the Sharp Zaurus deal we did together, but never managed to keep up with it to see how the company has done.

Very well, it seems. I had no idea, for example, that the Skype client and Google Earth were built using Trolltech's Qtopia. Nor did I know how much of a competitive force it has become in the Chinese mobile phone market (increasingly Linux-based). I plan to take a closer look at the company when I get back to the US. I'll let you know what I find.

Anyway, I've got to head to head to my meeting with Alfresco, another cool EMEA-based startup.

Worshiping at the Apple shrine

So, I couldn't keep away from the Apple Store in London. Isaac (my son, who is traveling with me, watching videos or with a sitter while I have meetings) got to go to Hamleys, so I made him come to my toy store.


Btw, the Nano (which Isaac is listening to here) is more incredible in person than I had thought possible. It is amazingly cool. Looking at pictures of it online, you don't get a good sense of just how small and thin it is. It's like carrying around a scrap of paper, except that it cranks your tunes. Now if I could just get a phone that did that....

News: Microsoft learns how to anger friends and influence people (to buy elsewhere)

I can understand many things. Fighting with customers to browbeat them into using your software isn't one of them, as the Boston Globe appears to report Microsoft is doing in Massachusetts. I think there are a lot of good reasons for Massachusetts not to mandate its "open policy" in the way that it has done. Reasons I've expressed here.

(My main problems with Massachusetts stem from its thinly veiled finger-in-the-Microsoft eye and its choice of "open" formats. I think it could have done better than what it came up with, if its goal is to ensure maximum access AND longetivity. On this latter point, it's funny to hear anyone talking about longetivity in technology. The hardware and media changes (3.5" floppies, anyone?), the software standards change (even Microsoft Office file formats have changed from version to version. I have almost as much hope of having my Office 97 documents to show up well in Office 2003 as I do OpenOffice's saved .doc/.xls/.ppt files :-). There are no long-term standards in technology. As Dave Dargo has been saying lately, technology only cares about NOW.)

So, I agree with much in this Microsoft statement:

''Were this proposal to be adopted, the significant costs incurred by the Commonwealth, its citizens, and the private sector would be matched only by the levels of confusion and incompatibility that would result from the fact that the OpenDocument format is such a nascent and immature format," Microsoft general manager Alan Yates wrote in the company's comments. He suggested a move to OpenDocument would flout practices requiring the state to seek ''best value" in procurement.

But I disagree with the underlying sentiment in these excerpts from Microsoft's 15-page complaint:

• No other government entity in the U.S. has made similar policy moves.
  
• Beyond the immature and parochial status of the OpenDocument format [Agreed, but might have been said in a nicer way....]
  
• It is also possible that the proposed policy violates applicable Commonwealth statutes.
  
• Moreover, there is no principled basis for the Commonwealth to adopt these unprecedented revisions

Microsoft could ease some of Massachusetts' concerns by creating some sort of baseline open version of its file formats. Maybe I'm being naive, but it seems like this would get the sides closer to a compromise. Instead, Microsoft appears to be browbeating its own customer - one that seems more determined to flout Microsoft than to worry about open standards. If this is the case, arguing with Peter Quinn and his team is only going to accelerate its move away from Redmond. A softer approach seems warranted.

At any rate, the Boston Globe indicates we're having a session on this at OSBC Boston in November. Hmmm....Maybe we will. We hadn't planned on it, but I think a session just came open.

'The debate is likely to intensify in November at the Open Source Business Conference at the Newton Marriott sponsored by International Data Group, where Microsoft representatives are expected to face off against Massachusetts government officials at a session on open standards.

Why not?

The relative badness of bigness

As I read the news about Oracle acquiring Siebel, I had to wonder: why is it that Microsoft is the only company that gets criticism for "integrated innovation"? It seems to me that all of the major vendors (my employer, included) are pursuing integration strategies, or suites of services. Even Red Hat, that bastion of choice. (Tongue firmly in cheek.)

If we dislike bigness (and US antitrust law is founded on that distaste), why not these other big, acquisitive, integrating players? Shouldn't we spread our animus more evenly?

If we did, there would be many more companies to hate. And wouldn't that make us all a little happier? ;-)

Commentary: More on customer service

I made it to my hotel in London. Finally. I say "finally" because Delta prolonged my trip by several hours. The difference between how I was treated by Delta and Marriott (where I'm staying) says a great deal about customer service, and offers a lesson for open source vendors (and others).

Delta

By a weird twist of fate, I had a five-hour layover in Atlanta last night. It was my fault, but once I arrived, I asked the reasonable thing: "Could I please standby for the earlier 7:40 flight?" (Mine left at 10:00 PM, and I had arrived in Atlanta at 5:00 PM, so I was there a good two hours and 40 minutes early....) The answer?

No.

When pressed as to why, they said, "Because there are no other 'L' fare seats on the plane."

"Are there free seats that aren't 'L' fares?'

"Yes. Quite a bit."

"So, I can't take one of those free seats, which are no better or worse than the 'L' fare seats [Note: They're all the same. The only difference is how much I paid for the seat; in this case, a "cheap" $1000+.], simply because? And instead I'll need to wait five hours for the 10:00 flight, and sit next to plenty of non-'L' fare seats, simply because Delta has an unreasonable policy?"

"No, sir."

"No?"

"No, there aren't any seats left on the 10:00 flight. It's full."

"So, wouldn't it be to your advantage to switch me to the earlier flight, which has lots of empty seats, so that you could possibly sell my seat on the 10:00 flight?"

"We can't, sir. It's Delta policy...."

...to be idiots. I'm a Platinum (100,000+ miles/year) on Delta. Somehow, you'd think they'd try to take care of me before they collapse into bankruptcy.

Now contrast this with the service I received upon checking into my hotel in London. Marriott is restoring the Grosvenor House Hotel, where I'm staying. Perhaps because of this, but mostly because I'm also Platinum with Marriott (too much time on the road), they went out of their way to take care of me, upgrading me to a suite (They did this for me the last time I stayed here, as well, despite my having probably the lowest rate of any guest in the hotel), walking me to my room, pressing my clothes (when they saw I was annoyed they didn't have a steam iron), and generally bending over backwards for me.

Similar situation. Delta nearing bankruptcy; Grosvenor House in the middle of a major restoration (noise, clutter, etc.). But one is seeking to build customer loyalty through service (Marriott), and the other is pinching pennies and sticking to The Policy (which no one understands) to try to save its way to ... what?

You don't scrimp your way to customer satisfaction. By all means, companies should economize, but not on their customers. Companies should lavish attention and service on their customers.

As Dave Rosenberg recently wrote, IT generally doesn't work, is too expensive, and stinks. But people don't, whether they're your employees or your vendors/contractors. People can provide service. I think the best open source companies will be those who deliver exceptional customer service, whatever their product, business model, etc.

Commentary: Why leave Microsoft? (Stephen)

A great post from Stephe on his reasons for leaving Microsoft. OK, Stephe, you've convinced me: you don't hate Redmond as much as I said. :-)

Here's a snippet:

Microsoft is a Very Large Company. It is wrestling (as all large companies do) with the very processes that enabled it to succeed and grow. It is at a point in it's life where it needs to be defined by operational excellence as much as product development. The single scarcest resource is (as one friend pointed out five years ago) executive bandwidth. It takes forever to get a meeting with the very execs that can open the doors for success through their sponsorship. And once one of those meetings is scheduled, you enter data preparation and PowerPoint hell. For me, it had become not very fun. I simply don't have the patience of people like Danese Cooper, Simon Phipps, Bill Hilf or Bob Sutor. When interviewing with companies like Microsoft, you will often hear about the incredible reach you'll have in the world. That is of course predicated on the ability to be heard inside the company first, and on the ability to actually ship something.

Commentary: The world is not so flat?

Richard Florida pens a nice riposte (PDF version) to Thomas Friedman's excellent The World Is Flat. Florida argues, instead, that our world is "spiky." Says Florida:

By almost any measure the international economic landscape is not at all flat. On the contrary, our world is amazingly "spiky." In terms of both sheer economic horsepower and cutting-edge innovation, surprisingly few regions truly matter in today's global economy. What's more, the tallest peaks - the cities and regions that drive the world's economy - are growing ever higher, while the valleys mostly languish.

Sobering and disquieting, if true.

After working through some measures of "spikiness" (The world's population is clustering into cities; patents [Don't say it, Stephen!] are overwhelmingly concentrated in the US and Europe]; scientific citations are almost completely in the US and Europe; etc), Florida makes this observation:

This is not to say that Indians and Chinese are not innovative. On the contrary...Indian and Chinese entrepreneurs founded or co-founded roughly 30 percent of all Silicon Valley startups in the late 1990s. But these fundamentally creative people had to travel to Silicon Valley and be absorbed into its innovative ecosystem before their ideas became economically viable. Such ecosystems matter, and there aren't many of them.

Again, disquieting, for those of us who either cannot afford to move or simply do not want to. But Florida gives no quarter:

...[A]lthough one might not have to emigrate to innovate, it certainly appears that innovation, economic growth, and prosperity occur in those places that attract a critical mass of top creative talent. Because globalization has increased the returns to innovation, by allowing innovative products and services to quickly reach consumers worldwide, it has strengthened the lure that innovation centers hold for our planet's best and brightest, reinforcing the spikiness of wealth and economic production.

Unfortunately, we see this even within the open source business community, where supposedly location doesn't matter. Zend (Israel), MySQL, and others have all emigrated to the US, to be close to capital (and not so much customers - Silicon Valley is on the wrong coast for that) and development talent. EMEA's loss, a loss that doesn't get replenished fast enough, as Matthew Langham has written. If we have a choice in the matter, do we really want our economy so tightly bunched in a few economic centers?

This leads him to a troubling conclusion:

We are thus confronted with a difficult predicament. Economic progress requires that the peaks grow stronger and taller. But such growth will exacerbate economic and social disparities, fomenting political reactions that could threaten further innovation and economic progress. Managing the disparities between peaks and valleys worldwide - raising the valleys without shearing off the peaks - will be among the top political challenges of the coming decades.

Let's hope the world is flatter than he believes....

Speaking truthfully (Habermas)

I had forgotten Habermas, whom I first encountered in my Masters program, in this whole Microsoft vs. Linux debate. What I've been arguing is, essentially, this: The truth behind which is better for users is something that will only be arrived at through open, candid debate. This, of course, is the essence of Habermas' communicative rationality.

Utopian for him, utopian for me. The problem in this technology debate is that the two sides don't seem to want to argue about technology (i.e., to have an objective, forthright debate), but rather to mythologize themselves and demonize the other side. Neither is constructive.

I'm willing to bet that users are smart enough to continue to buy both Linux and Microsoft, where each is best suited. But I guess this isn't really what either side wants, is it? That is, since when is salesmanship about getting the customer what she actually needs?

/naive and idealistic sigh

Announcement: O'Reilly Emerging Telephony Conference

eBay buying Skype? Small potatoes to the Really Big telecom news of the day:

Tim thinks telephony is interesting, too, and is hosting its inaugural "eTel" (Emerging Telephony Conference) January 24-26, 2006, in San Francisco.

I say that tongue-in-cheek, but there's a grain of truth to it. Tim signifies approval of an emerging space by starting a conference around it. (Of course, I think part of the reason O'Reilly is spinning out conferences at a torrid pace - Web 2.0, eTel, Where 2.0 - is that they've figured out a formula for making them mint money. He's a capitalist, after all.

Tim's not always right, but I'd bet with him more often than I'd bet against him.

See you there.

News flash: Eric Raymond is god (he tells us)

Hubris, hubris, everywhere, but not a drop to...think.

It was dumb for a Microsoft recruiter to invite Eric Raymond to apply for a job at Microsoft. But it was worse to read ESR's response:

If you had bothered to do five seconds of background checking, you might have discovered that I am the guy who responded to Craig Mundie's "Who are you?"” with "“I'’m your worst nightmare"”, and that I'’ve in fact been something pretty close to your company'’s worst nightmare since about 1997. You've maybe heard about this "“open source"” thing? You get one guess who wrote most of the theory and propaganda for it and talked IBM and Wall Street and the Fortune 500 into buying in. But don'’t think I'm trying to destroy your company. Oh, no; I'd be just as determined to do in any other proprietary-software monopoly, and the community I helped found is well on its way to accomplishing that goal.

Oh, my. It's rare that someone will parade their ego so rashly, and so openly. Eric Raymond was a pivotal figure in the early rise of open source. He did not, however, "talk[] IBM and Wall Street and the Fortune 500 into buying in." He has not been Microsoft's worst enemy since 1997 (or even since last week). He also did not write "most of the theory and propaganda for it."

The community did, Eric. And, in the case of convincing real users to use it, that honor goes to IBM and real vendors.

Individuals should never overestimate their importance to a movement, a community (which Bruce Perens also did a few weeks ago in his letter complaining of OSI's rejection of his services). As it stands, the community has largely moved on to grow well beyond a few core open source innovators, in which group I'd happily include both Perens and Raymond. All of which begs the question,

"What have you done for the community lately?"

P.S. Stephen quit Microsoft to join Optaros. He was not fired, as you suggest in your blog, and was not in any danger of being fired. He was working alongside Jason to help the company experiment with and grow less frigid toward open source. He simply moved on because he felt his time was better spent with an open source-friendly employer.

Picture: George Bush fishing...Katrina

I'm a Bush supporter, but I can't help it. This is too funny.

Commentary: eBay Redux

The more I think about the eBay-Skype deal, the worse the idea gets. I hinted this morning that I thought eBay is mistaking what it needs to improve customer acquisition/retention/value. Trust is what eBay needs. Talk is what they bought.

Think about it. The history of retail is toward removing communication from the sales process. Most recently, the auto industry recognized this with its no haggle policies". When I go to the mall (which happens very infrequently, fortunately) I don't expect to have a conversation with the maker of a given product. I might be willing to talk with the salesperson, but mostly I just want to walk in, find an appropriate product (which is the only reason I'd ever talk to the salesperson - to get more detail on a given item), buy it, and walk out.

Now, even if you agree with this, you might believe that the world of used goods is different. In this alternative universe, we need to talk with the seller to find out its deficiencies, repair history, etc.

I disagree.

It is true that we may do this, but I would submit that we'd strongly prefer not to do this. We'd rather trust the seller and simply buy the product, without having an extended conversation with a complete stranger, asking uncomfortable questions of them. Some people can do this, but I'm betting most people would strongly prefer to have the used product buying experience feel much more like the neutral retail buying experience.

How do we get there? Not by overlaying an already worry-prone buying experience on eBay (My one near-purchase on eBay was someone in Canada who attempted to defraud me) with a geeky communication medium (Skype).

No, the answer is to find ways to overlay the eBay buying experience with trust. eBay has done this moderately well in the past, giving users ratings based on their purchasing/selling history on the site. It was a great start. Now they need to find ways to bake in the trust built outside eBay. As I suggested in my previous blog entry, I believe this is best done by tying users together, six degrees of separation style. Establish the relationship between me and the seller, so that he/she becomes a trusted source (or so that I can see the link is somewhat tenuous to the buyer/seller, and make an informed decision).

Again, I think human nature tends toward sociability, but that this sociability is generally constrained by bonds of family, friendship, and community. We don't want to be able to call a complete stranger to ask about the figurine they're selling. What we want is some neutral way to trust them without ever talking to them. Skype doesn't get us there.

Commentary: More on the (official) eBay acquisition of Skype

It's official. $2.6 billion dollars later, eBay is buying Skype.

Now what?

Well, according to the BBC (and the press release/conference), Meg Whitman (eBay CEO) said:

"Communications is at the heart of e-commerce and community," said eBay chief executive Meg Whitman.

"By combining the two leading e-commerce franchises, eBay and PayPal, with the leader in internet voice communications, we will create an extraordinarily powerful environment for business on the net."

This is absurd. Don't believe me? Well, consider, for a moment, eBay's demographics, and try to imagine them going out to buy a headset so that they can talk to the person from which they're going to buy. Last time I checked, eBay's demographics don't mesh well with the typical Skype user:

* The most visited categories are:
1. Collectibles
2. eBay motors
3. Computers
4. Electronics
5 + 6. Clothes/Shoes + Toys

* Demographics for eBay users are:

1. Age: 45-54 (28%); 35-44 (26%); 55+ (22%); 25-34 (17%); 18-24 (7%) [i.e., 75% are over 35....Does your dad use Skype?]

2. Income: $35-75K (31%); $25-50K (23%); $75-100K (19%); $100-150K (14%); $150K+ (8%); $0-25K (5%).

I think eBay must have inhaled at some point, because this deal is on crack. It smacks of Time Warner - AOL: marginally great in theory, absolutely poor in practice. Let the pureplays kick their tails in while they try to figure out the synergy in this deal.

A much better bet for the typical eBay user would be for someone to network address books, i.e., build a trust community so that they know that Phil, from whom they want to buy, is a brother of Cousin Charlie's. THAT is the sort of communication that eBay users want - the kind that they can trust.

News: The Access acquisition

By now you've seen that Access is acquiring PalmSource. I'm really glad to see it, as PalmSource hasn't been doing very well this past year, and needed a boost from a major player with more resources. That said, I'm not a big believer in acquisitions, having struggled to incorporate the seven (or was it eight?) acquisitions we made at Lineo. Novell has done better with its acquisitions, but it's no secret that there is always difficulty merging two corporate cultures, roadmaps, etc.

All that said, I remain a believer that

"When I think of the Linux desktop of the future, I think of it on my phone; on the embedded devices I spend most of my time with. The only thing that has been missing is the applications...."

I said this when PalmSource acquired China MobileSoft (Madeline Duva, CMS' former CEO, incidentally, is up to something....), and still believe it. We may be a long way off from that vision, but the fact remains that mobile devices (especially phones) are increasingly the one computing device everyone has, whether in China, the United States, Nigeria, or Hungary. The "game" eventually goes to the company that can serve that market best.

Speaking of momentum (and innovation)...Apple

Thanks to Matthew for pointing me to this great analysis of Jobs' iPod Nano announcement. Summary: "We win."

Just a reminder, Stephen, that yes, we do buy innovation. Lots of it. In fact, OSBC is buying a boatload because we're going to be giving away the Nano to early OSBC San Francisco attendees, just as we're currently giving away (until the end of the month) Sony PSPs to full registrants.

It's not that we wake up one morning and say, "My, I sure need some innovation today! Where can I buy some?" Rather, we buy products that catch our eye, that deliver results, that give us a performance or feature edge. All of this stems from innovation.

As a reminder, innovation comes in different forms (process versus product), but it attracts customers, whatever its form. Again, it's not that customers seek out innovation, per se, but they/we do seek out its effects.

[Just giving Stephen a hard time. He can dish it out as well as he takes it. :-) ]

Commentary: r0ml asks, "What do you do after you win?

I can think of very few people I'd rather listen to or read than r0ml. He has an exceptional post today on open source's momentum (well, Linux's). Take a look.

...[F]ollowing the ignoring phase, we got Linux (which arguably began the laughing phase) and the thousands of "What is Open Source"” slides (1. Run 2. Study 3. Distribute 4. Modify) -— followed by the hybrids in the fighting phase. And then, one sees the Microsoft Windows XP release notes (KB306819). From which I quote:

Copyright 1985, 1988 Regents of the University of California. All rights reserved.

And also

Copyright 1998-1999 Greg Roelofs. All rights reserved.

which presumably refers to an open source implementation of PNG. There are more such copyright statements embedded therein. So, for some definition of "“open source"”, Microsoft Windows XP is open source. That is to say, it contains open source components - it is "hybrid open source". I guess that marks the "“then you win" phase.

Then what happens? Well, in Gandhi'’s case, one needs politicians, and the work of governing must go on, but the work of the revolutionary is done. In the object-oriented (and open source) case, there is implementation work related to specific technologies, products, and industries -— but the work of evangelism is done. You can still attend OOPSLA - but the "What is Object-Orientation" presentations are gone.

Different people will define victory differently. But, given that you think youÂ’ve won, then what do you do?

You do something else.

Exactly. But what?

The importance of momentum

My team, London-based Arsenal, lost today to a middling team, Middlesborough. 'Boro hasn't been worth watching in years; Arsenal went undefeated two seasons ago (and ended up second in the English Premiership last year). The talent is heavily weighted toward Arsenal, but today's match was heavily "goaled" toward Boro: 2-1.

What happened? Middlesborough scored five minutes before the end of the first half. They went into the locker room jubilant. Arsenal went into their locker room wondering how to pull even.

If you're into sports (any sport), you know the importance of momentum. In basketball, a team can go on a 10-15 point scoring spree. In soccer, one goal can completely alter the mindset of the players on both teams. One goes into defensive, why me? mode. The other into attack, we can beat these losers mode. When that switch goes on/off, the talent on the pitch (field) gives way to momentum.

In technology, it's no different. At any given moment, momentum governs who wins (revenues, market share) and who loses (share price falling, customers defecting) most of the time. The equivalent of talent - superior technology, customer service, etc. - matters at the start of the game, and perhaps in lulls in the game. In these down moments where each side is trying to build momentum, talent still matters. But it becomes much less important once the mind game is won (with employees, customers, and competitors), momentum develops, and that side is off to the races.

In the operating system wars, Linux currently has winning momentum. IBM's billion-dollar pledge helped to overcome initial friction to build momentum, but other things (including conferences like OSBC and OSCON) have helped it along. Most of Linux's momentum was built beating out Unix: Linux was dramatically cheaper (mostly in hardware costs) and leveraged existing skillsets. That Unix-to-Linux momentum is now carrying Linux into the much more important, and difficult, battle: Linux vs. Microsoft (where the hardware is just as cheap, the skillset equation probably favors Microsoft, etc.).

Many things contributed to build Linux's momentum. Now this project-specific momentum is bleeding over into the development model - open source - that fuels it. Momentum is on open source's side.

But it won't necessarily remain as such. Had Arsenal scored (by fluke or skill) quickly in the second half, momentum would have evened out and talent could have played a larger role. In basketball, coaches will call a time out to try to dampen the enthusiasm and adrenaline that fuels the other side's momentum. Or teams will take out a particularly momentum-inspiring member of the opposing team. Etc.

In IT, something like SCO could have stalled open source's momentum. It did, for about 0.3 seconds. But a more credible lawsuit from a more credible litigant could do the same. As could, more likely, "rolling thunder" announcements from the Microsoft camp of CIOs who were switching back from Linux to Microsoft. Or [insert scenario of your choice here].

The point is, momentum is fickle unless it's constantly fed with news and actions that reinforce it. Those who grow arrogant or reckless with their momentum will often find doing so leaves them open to attack, quickly reversing the tide of momentum.

All of which relates to the posts I've made over the past few weeks. VCs seem to be funding an open source bubble, fueled by over-confidence in the momentum and its natural staying power. Pundits have grown careless, believing the fight is largely won already. Even open source vendors seem to be resting on the laurels of open source's momentum, rather than actual value that they deliver to customers.

I sat through a presentation the other day which basically said "open source's domination is inevitable." Of course it's not. Nothing (except death, taxes, and highly verbose emails and posts from me) is inevitable.)

Two years ago Arsenal was invincible. Literally. This year, they're going to really have to work just to keep in the top three in the Premiership. Nothing is inevitable.

More on Massachusetts (Stephen)

So, ever the ardent anti-Microsoft crusader (I can say that about him - he's on a plane right now and won't be able to respond for hours :-), Stephen Walli comes to the rescue of Massachusetts' decision to opt for "open" standards (no matter that no one uses the technology in question, or that PDF isn't as open as Peter Quinn would like to believe). Stephen makes some good points:

I find it ironic that when Microsoft has been confronted in the past with governments trying to develop open source software preference policies, they (rightly in my opinion) argue that the government shouldn't care how the software was developed and licensed, but rather focus procurement policy on software acquisition against standards. Now that they have a government behaving exactly as they have requested, they are claiming that governments should focus procurement policy only on standards that Microsoft already supports, or de facto technologies that they sell, i.e. the status quo.

Not strictly accurate, but mostly so. Microsoft's argument (and mine) is that governments and people should buy software that works. Period. It doesn't matter if it's open or closed. It matters if it solves a business problem.

Which leads into a point Stephen makes with which I have some difficulty:

Microsoft is still confused about why customers buy things. Customers don't buy innovation. (And customers couldn't care less about nor do they buy patents.) They buy solutions to problems. The government has a problem, and determined a solution, and Microsoft could support that solution, but would rather try to tell the customer that they don't understand their own problem space, and should keep giving Microsoft money for innovations, regardless of the problems the customer wants to solve. It's the "we know best what's right for you" defense, and Microsoft still doesn't appreciate how arrogant it sounds.

Stephen is here confusing patents with innovation - I don't think Microsoft is making the same mistake. When, for instance, Jason talks about innovation (e.g., here and here), he's talking about cool products that people buy, and not the patents undergirding them. He's describing the tight integration of disparate pieces of IT that Microsoft pulls together to remove complexity for customers. You can argue whether this integration is a good or bad thing (I think it's good, but can lead to bad consequences; the inverse is also true - sometimes decoupling things leads to inordinate complexity and, hence, cost), but let's not misconstrue Microsoft's argument.

That argument? That Microsoft is an innovator (sometimes measured in patents and copyright), and customers should buy from it because it will consistently deliver cool, useful (i.e., innovative) new things worth buying.

Again, you don't have to agree that Microsoft actually does these things. But let's at least understand the company's argument, and work accordingly.

As for Stephen's parting shot...

I commend the Commonwealth on its move. It has some short term pain, but should show a lot of benefits in the long term. This is how I would want my government to spend my tax dollars -— take the long view, and do the hard work to serve my needs best.

...I think he both understates the short-term pain and overstates the long-term benefit. For myself, I prefer my state documents to be in HTML (i.e., download nothing, fill it out online) or PDF, because I'm not always at a machine that supports either Microsoft or OpenOffice formats (more of the former can be found than the latter, though).

My beef in all this? That we state our assumptions, biases, etc. clearly. I don't like ideologies. Peter Quinn is on the record as all-but-saying he's doing this as a kick-in-the-face to Microsoft, which won't keep up with the times. That's not good policy or good use of tax dollars. IT is just a tool. We shouldn't invent a religion around it.

Commentary: Jason says Linux is doing its own "embrace and extend"

Hating as I do hubris and ideology of any flavor (open or closed), I liked what Jason had to say today on his blog.

Responding to an online debate as to whether Linux should look to Unix or Windows for inspiration on how to most effectively compete, Jason says:

It is not good enough for Linux to simply emulate Windows, it is going to have to extend beyond and that is why innovation is the whole ball of wax. Windows Vista is not just a rehash of Windows - it is value-driven and reaches beyond anything we have done before. Otherwise - why buy it vs. running the high-quality OS already available? So Linux is not going to have to match Windows XP (which is going to take a long time and a lot of hard work to do), it is going to have to match the pace and quality of innovation coming out of our dev teams. This is not just about user interfaces, it is about extensibility, security, manageability, quality, performance, etc. etc. etc. Each having it own elements of compulsory requirements and innovation value-add.

So next time you feel like pointing an accusatory finger at Microsoft and repeating the nasty "embrace and extend" - think about the engineering, marketing, sales, implementation, and support challenges being faced by the Linux vendors. That commercial community is completely focused on "embrace and extend" today.

Startup: OpenI

A new open source BI (Business Intelligence) company has hit the street: OpenI ("OpenEye"). I found out about it at my son's soccer practice, of all places. I haven't taken a deep enough look yet to see how it compares with JasperSoft (for which I serve as an advisory board member), Greenplum, Pentaho, etc. It does seem that, so far, most of these companies tend to complement each other, rather than compete with each other.

If you know of a good analysis of the emerging open source BI market (and its players), please let me know. I'd like to link to it.

Commentary: Speaking of eBay

Speaking of eBay (Salesforce saying it wants to be the eBay of applications), what's up with thenews reported by the WSJ that eBay is thinking of buying Skype? I don't care what the prognosticators say, it's weird (if it's true). You can invent tenuous links/rationales for the merger, but they all come down to this:

eBay's growth has stalled, Skype's star continues to rise, and so despite being very different, somehow a mashup between the two makes sense.

I'm not buying it. I hope it's a silly rumor. However, for Index Ventures (and the Skype guys), it would be an amazing coup. Skype has only raised two rounds of funding, making the deal immensely valuable for the investors of cash (Index, DFJ) and toil (Skype).

What are Skype's revenues, anyway? Next to nil, I suspect. Or $72 million, following Om's reasoning. I'd put $100 million as "next to nil" when we're talking a $3 billion cash offer that is allegedly on the table. Inflated, in my view, given Skype's increasing competition (Vonage, for one, which I love) and its still marginal runway to real world phone use (Vonage and others have them licked there).

News: Salesforce creates an "AppForge" of sorts

Phil Wainewright has an interesting post today on Salesforce.com's new AppExchange site. As CEO Benioff explains:

"We have built an eBay for enterprise applications....We want to make it as easy as buying music on iTunes and playing it on your iPod. We can have customers download these applications using the same type of technology. That was the breakthrough for what we call the AppExchange. You can share apps, try apps, buy apps, test apps, you can search for apps, and publish reviews of them. You can even sell apps. You can look at the applications by company, industry, size of company, language, by price.

"The power of that is you can reach this long tail of applications. SAP and Oracle may deliver 10% of the applications you need to run your business, but there's this large percentage of your business that won't be managed by Oracle or SAP. This is the long tail of applications."

Aside from the dreaded overuse of "long tail," as well as the fact that I think SugarCRM is working on something very similar, this is interesting news with cool potential. It's not dissimilar from what I've been arguing Microsoft should be doing with its various platforms, instead of trying to build everything itself.

Commentary: Tim on why Microsoft can't best Google

Tim posted this on the O'Reilly Radar yesterday. He cites James Governor (Redmonk), who summarizes it all as such:

Microsoft's business model depends on everyone upgrading their computing environment every two to three years. Google's depends on everyone exploring what's new in their computing environment every day.

And so (as the argument goes), Google will win and Microsoft is doomed. I disagree with this premise, whatever I may think of his suggested outcome. I disagree not so much because of anything Microsoft has done (though License 6.0 comes to mind), but because of my Mac.

Apple chooses to sell me an operating system - they sell me a moment in time in that OS' lifecycle. Or, rather, they sell me that snapshot with the right to receive updates, to a point.

Guess what? I'm constantly checking for new updates, and get excited (OK, I'm a loser) when Apple delivers new functionality to me through their auto-update feature.

Would I pay for this as a subscription, most likely initiated at the time of the purchase of new hardware? Sure. I already do (paying $250+ for the AppleCare Protection Program). Google hasn't stumbled onto anything particularly novel. Quite the opposite. They found a way to let me get their service without paying anything. I have never once, in my life on this earth, clicked on an ad that Google has served up. I'm a free rider.

Back to License 6.0. As I discussed with Steve Mutkoski (Microsoft Legal) and Dave Dargo (Olliance Group) at OSBC Legal, it's highly ironic that the buzz in tech is "subscription" when Microsoft arguably broke this new ground with License 6.0, which everyone (at the time) whined and moaned about. Much of that complaining probably stemmed from the Software Assurance Program aspect, but some also likely emerged from the lack of guarantees that Microsoft would actually deliver timely upgrades.

Think about that. Five years later (or however long), who really thinks about this when tapping into Novell's Red Carpet, the Red Hat/MySQL/etc. networks, or Google? Microsoft was ahead of its time, and could now say, "We really think those Linux guys are SUPER-innovative in the way they price. We've decided to follow suit with License 7.0, our Linux-like pricing mechanism."

"Brilliant!" The pundits would gush. "Microsoft is finally understanding that it needs to join the 21st Century!"

Whatever. Reminds of a venture firm with whom I met recently. They don't like funding companies that haven't bought into this "new" subscription-style pricing. Funny how our thoughts change (as just five years ago they wouldn't have touched a subscription-priced offering, and didn't).

In concocting theories around buzzwords and associated phrases (like "Web 2.0," "architecture of participation," "data is the Intel inside," etc.), which are more descriptive than they are prescriptive (see my rant on the FSB mailing list several months ago on this topic), we need to be careful not to throw the baby out with the bathwater, as it were. I'm not sure enterprise customers care much about buzzwords or catchphrases. They care about value, however delivered.

It may be the case that some prefer subscription services. Many likely don't. It may be that in five years, everything will be paid for in this way. But maybe not. There's nothing sacrosanct about it. It certainly doesn't reflect the normal commodity world outside tech, where you can rent-to-own, but the general theme is to pay full-freight upfront, and those who can't use credit.

Microsoft and Google will ultimately win or lose customers according to the value they deliver to customers. It's really not any more complex than that.

Startup: Zimbra

The startup formerly known as Liquid Systems (funded by Benchmark) has emerged as Zimbra and looks very cool. I think the world of Scott Dietzen (President & CTO, and former CTO of BEA), and am willing to give the company the benefit of the doubt based solely on his affiliation with it. But they don't need to rely solely on good credentials.

You should check them out. IDG has a nice write-up on them here. As I understand it, Zimbra's Exchange clone is much like what AppTran has done. The difference lies in the "client." (Is a client a client if it's not really there, but instead runs in a browser?)

That difference, incidentally, is something I still wonder about. I don't like web-based email, no matter how slick. I'll use it if, like CanyonBridge, it integrates into other systems (CRM, ERP, etc.) that I already use in a browser, but I'm not sure I'd choose to go to the browser without that integration incentive. But maybe we're getting to the point that browser-based apps are slick enough that we will no longer think about whether it's a fat or thin client?

TBD....

Commentary: Dave Dargo's OSBC Legal presentation

Dave Dargo delivered a great kickoff presentation to yesterday's inaugural OSBC Legal Seminar. Entitled "Understanding the Business Imperative: Open Source Development and Business Models," Dave (formerly VP and Founder of Oracle's Linux Program Office) walked through the industry's evolution, and offered some interesting insights into why we're seeing the Windows vs. Linux battlefield today.

A few gems:

"Linux is a great operating system because it gets out of the way."

Dave's point was that Linux is a minimalist OS, allowing vendors and customers to focus on the value up the stack (applications, middleware, etc.). Microsoft, of course, has both the belief that the OS is more than "just a foundation to more interesting things" as well as the financial incentive to continue driving value into the OS (to justify upgrades, etc.).

This tension between Linux and Windows came out in Dave's next slide, which reads:

O/S License Value
Established Price Point: $0.00

Whatever Novell and others may be charging for support for and around Linux, Dave's (technically accurate) point is that the license cost for SUSE Linux is $0.00. He therefore questioned Microsoft's continued ability to charge for Windows.

The problem with this line of reasoning (as I suggested during his presentation) is the differing views of value. Linux has zero license costs, but arguably those costs are made up in integration (giving rise to the SpikeSources of the world), administration, installation, etc. Microsoft's goal is to pre-integrate all of that, to take as much complexity out of running Windows (and associated applications) as possible. Same end, but different means to get there. Either way, you pay. No free lunch. So, really, the license cost of Linux is zero only in the narrowest sense, and not a very useful or accurate one.

Interestingly, on the integration question, the jury is still out on whether Microsoft will be able to sustain it. While I was in Seattle I spent time with several current and former Microsoft employees. The former employees all said that the company went downhill once it started trying to coordinate divisions more tightly, forcing innovation to coincide with Windows development, for example. Most of the existing employees acknowledged a certain loss of flexibility in the company, and some slowing in their ability to execute quickly. But they said the tradeoff is tighter integration of products and, hence, a better customer experience.

I guess my question from this is whether or not customers will want to buy into the complete Microsoft package. There's something very alluring and compelling about having a vendor bake out the complexity of technology so that you can get to work. But there's also something compelling about being able to mix and match the best applications, without worrying about which browser you're using, which operating system, etc.

Does the Linux versus Microsoft decision ultimately come down to this? A choice between complexity and maximum freedom, or simplicity in exchange for more limited freedom? If so, I can see pros and cons in both.

My right to English-only spam

Forget CAN SPAM. Forget the English Only amendment to the US Constitution. I want someone to defend my right to receiving spam in my own language. I can't even read the Russian, Chinese, and Pig Latin spam that I get on a daily basis. How am I supposed to exercise my right to unsubscribe? To see if I'm missing out on a major deal on V1@gr@?

The US is burning while my congressman fiddles....

:-)

News: Lexmark ruling and 'box-wrap' licensing

Every so often, US courts send clear reminders as to just how far they've gone in overreaching their authority, as well as how poorly they're using that authority. The Ninth Circuit's recent Lexmark decision is about as bad a decision as I've seen. It's amazing to me how far that particular circuit will go in expanding social "rights," while also doing its utmost to harm consumers (i.e., giving businesses a stranglehold on customers).

Notes Donna Wentworth on so-called 'box-wrap patent infringement':

What'’s that, you ask? Evidently, it'’s when you ignore the terms written on the side of Lexmark printer cartridge box, refilling the cartridge with ink even when the company has designated it "“single use only."” According to the Ninth Circuit ruling [PDF] this week in ACRA v. Lexmark, opening the package means you agree to Lexmark'’s wishes. And if you break that agreement, you could face claims under contract and patent law.

As Fred von Lohmann explains it, it'’s sort of like when you buy those fancy Gillette Sensor razors, then purchase cheap replacement razor heads -— except that a court has ruled that if the package says "“single use,"” then by opening it you'’ve agreed you can'’t have any cheap replacements (but you can buy another Gillette "single use"” razor). And that means the company that makes the replacement heads is out of luck, too.

Writes Fred:

[The strategy here is] a variant on the "“shrinkwrap license"” that used to appear plastered on software. Lexmark is bringing this practice to the world of patented goods. If you step outside the bounds of the "“contract"” (by giving your spent cartridge to a remanufacturer), you'’re suddenly a patent infringer. More importantly, Lexmark can sue cartridge remanufacturers for "“inducing"” patent infringement by making and selling refills.

So, I buy a Lexmark printer, not knowing (or, at that point, caring) about the cartridge inside, until that first cartridge runs out. At that point, I'm forced to buy another Lexmark cartridge. Or, at least, I can't refill my existing cartridge. And if Lexmark can deny me that right, then what's to say they can't also restrain my downstream actions by, for example, specifying that I must upgrade to a more expensive Lexmark cartridge on my next purchase?

The Ninth Circuit writes:

We agree with the district court that Lexmark has presented sufficient unrebutted evidence to show that it has a facially valid contract with the consumers who buy and open its cartridges. Specifically, the language on the outside of the cartridge package specifies the terms under which a consumer may use the purchased item. The consumer can read the terms and conditions on the box before deciding whether to accept them or whether to opt for the non-Prebate cartridges that are sold without any restrictions.

The district court found that the ultimate purchasers of the cartridge — consumers — had notice of the restrictions on use and had a chance to reject the condition before opening the clearly marked cartridge container. Arizona Cartridge, 290 F. Supp. 2d at 1044-45. These findings support the conclusion that the consumer accepts the terms placed on usage of the Prebate cartridge by opening the box....

We hold that the contract on its face appears to be able based on the district court’s findings that consumers (1) have notice of the condition, (2) have a chance to reject the contract on that basis and (3) receive consideration in the form of a reduced price in exchange for the limits placed on reuse of the cartridge.8 The contract permits Lexmark to restrict the use of its patented item and gives Lexmark a legal basis for asserting its ability to enforce its restriction.

Obviously, the "prebate" discount that consumers got in exchange for giving up the ability to refill their cartridges plays a role in this particular case. Fair enough. What is troubling to me is the court's expanded view of what constitutes a contract. "Single use only." In those three words the court found offer, acceptance, consideration (exchange of value for a burden/restraint), and privity of contract. Amazing.

No doubt this is great for manufacturers. The next time you buy a car, you might well see a ribbon around it telling you that by opening the door and driving away, you agree to have all service performed at the dealership (at a 5% discount off their heavily inflated prices to cover the "consideration" requirement). Ridiculous, you say. Not in the face of the Ninth Circuit's reasoning, and especially not if a majority of manufacturers in a given industry adopt similar tactics such that consumers have no real alternatives.

News: On OSBC, CIO needs, etc.

Here's an interview I recently did with IT Business Edge. It was theoretically about OSBC, but the meat of the interview has much more to do with where the open source world is going, who the key emerging players are, etc. Here's a snippet:

Question: What are the burning issues or problems that people are worried about? Is it migration costs, C-level buy-in, or is it spread across a lot of questions about bringing open source into the IT department or increasing it?

Asay: There are a few different things. I've been involved with open source since 1998, so this is an issue that I got over a long time ago, but I've been surprised in going out and talking with IT executives to find out that one of the biggest issues is licensing. There's still a lot of, not so much a fear over SCO, the company that brought a lawsuit against IBM, as just confusion as to what all these different open source licenses mean. It's just a general misunderstanding or lack of information, so that's another focus of OSBC. We have an entire track devoted to legal issues, which in the normal software world would be death to a conference; if you spent a third of your time talking about the law, you'd put everybody to sleep. But in open source, actually, you just have to know at least the fundamentals of open source licensing to feel a certain level of comfort.

Matt finally gets around to responding to Luis

So, Luis took me to task for pulling down this post. Not sure why it caused even a moment's confusion, as neither my original post nor Luis' rejoinder were ever at odds. (Unfortunately, Google has not cached Luis' riposte, so I can't link to it here. My bad.) I was on my way back from a difficult three-day camping (mostly hiking) trip through the Wind River Mountains, was dead tired on my drive home (where I read Luis' post on my phone), and just deleted the post, not thinking it worth the trouble to respond (or to defend myself). I didn't understand Luis' quibble in light of what I actually wrote, but was too worn out to be bothered.

However, several days later I just stumbled across Luis' riposte again, and figure I'll respond. It's clear he misread my blog, though I don't think he did so disrespectfully or intentionally. He's just an active open source developer that is probably fed up with people telling him that he owes the world code + all of his spare time. Fair enough. Luis said:

Matt Asay wrote this post about 'the open source ethos', which I thought made a fairly common mistake- assuming that if an open source volunteer is not being generous to you, that he is not generous at all.

I agree with what Luis says; I do not agree that it responds to what I actually said. I was arguing that open source is blessed (or plagued, if you will) with an ethos of giving back. I was not suggesting that one should spend all of one's time giving back, or that one should be indiscriminate in how we give back. I was just saying we should give back. That's a normative statement. But then, open source is a somewhat normative world.

In my original post, I took issue with my friend's statement that he didn't get anything out of helping others (in this case, an academic), and so he wouldn't help. I was challenging that egocentricism, Luis, and not one's right (and, probably, duty) to choose the best way to give back. John Stuart Mill, we love ye.

So, Luis basically argues: "Just because I choose to give back with code or in some other preferred way, doesn't mean I'm not giving back."

I have no counter to that, both because I believe it and because I never said otherwise. I just said that I thought a completely self-centered way of giving back (Calculating "gifts" based solely on the value they bring to ourselves) didn't strike me as being in keeping with open source. Are we disagreeing? I don't think so. Which is why it was just as easy to hit "Delete" as to take the 10 minutes typing this. :-)

(I do appreciate the comment, though. Thanks for keeping me honest.)

Product: iPod Nano

I was about to turn off my Mac and read some Trollope before going to sleep, when an email from Apple announcing iPod Nano hit my inbox. Oh, my. It is one of the most beautiful industrial designs I've ever seen. This will make a great giveaway for OSBC West in February 2006. (We're giving away the Sony PSP this fall in Boston, but only to those who register before October 1.)

There's a review of the Nano in PC World, but I'm not sure anything more is needed than two fully functional eyes. It's thin, has the same storage capacity of the original iPod mini, and has a color screen. And it's under $200. What more do you need to know?

This is the sort of technology I'm waiting to see from open source. Candidly, I'm not sure this is what open source does best. A very smart/wise IBMer (who shall remain nameless) once observed that, by definition, open source is not an innovator. By his reckoning, open source relies on communities, which can only form around established, understood technologies. Open source, in his mind, would always follow (though sometimes quite closely) on the heels of proprietary software. This view is similar to that expressed by my friend (a few posts down).

Perhaps a bit overstated. Perhaps. And perhaps less persuasive in our growing world of commercial open source. Perhaps. I'm actually not yet convinced. This is not to say that open source has less value than proprietary software - not at all. Who would argue that Dell is worth less than IBM, for example?

No, the question is not whether open source is valuable, but rather where that value lies. I believe software companies will always have the ability to choose how they wish to innovate - in the code itself, or in the process that delivers that code to market. Some will try to succeed in both, but I think Dell (and Wal-Mart) shows that you have to pick your path. Neither is right (or wrong) - it's a personal preference. But you have to choose what kind of innovation you're going to deliver. (This is actually consistent with Jason Maynard's (CSFB) criticism of Novell, as reported in his open letter to Novell's board: be a product innovator or a process (open source) innovator. Not both.

Agree?

News: Consumers to lead the way to desktop Linux

But then, what do I know? (SearchEnterpriseLinux interview here. Me at my most inarticulate.)

Russell Beattie on mobile Linux

Those of us who have spent time in the embedded world will yawn at Beattie's belated discovery (that Linux will dominate mobile). But for the rest, here's an interesting take on why Linux will beat out Symbian and Microsoft in the embedded OS war (for phones).

News: Ballmer threatens to kill Google...about time

In a sign that Microsoft is not abandoning ship in the face of a Google threat from above (no similar response to Linux from beneath just yet), Steve Ballmer has threatened to kill/bury Google, according to this CNET story. All I can say is, "About time." Here's a bit more:

In a sworn statement made public Friday, Mark Lucovsky, another Microsoft senior engineer who left for Google in November 2004, recounted Ballmer's angry reaction when Lucovsky told Ballmer he was going to work for the search engine company.

"At some point in the conversation, Mr. Ballmer said: 'Just tell me it's not Google,'" Lucovosky said in his statement. Lucovosky replied that he was joining Google.

"At that point, Mr. Ballmer picked up a chair and threw it across the room hitting a table in his office," Lucovosky recounted, adding that Ballmer then launched into a tirade about Google CEO Eric Schmidt. "I'm going to f***ing bury that guy, I have done it before, and I will do it again. I'm going to f***ing kill Google." Schmidt previously worked for Sun Microsystems and was the CEO of Novell.

It's nice to see some life in Redmond. I was worried the company was going soft. :-)

Startup: Centeris

I was just about to post on Centeris (very cool open source startup - fills a deep need for mixed Linux and Windows IT environments), but then saw that they're still in stealth mode. I spent time with them today in Seattle, and met with their CEO, Barry Crist, last week. Great team, great idea, and great technology.

Anyway, I'll forbear. But keep an eye on this one. I think they're formally launching (the company or a beta of their product) at OSBC East. This is one to bet on. (Sorry I can't say more.)

In Europe...care to meet?

Btw, I'm in London (mostly) from September 13-20 on business, with the intention to drop in on Brussels and Paris while I'm there. If you happen to be based over there, please send me an email. I'd enjoy getting together. I'm mostly interested to learn about new open source startups (Matthew and I have been trying to figure out why Europe lags the US in startups, especially open source, given that several of the best ones have been founded over there), but would also appreciate candid appraisals of the market (including Novell's place in it).

Given the amount of spam I get, I assume my email addresses are plastered all over the 'Net. Just in case, though, you can find me at either masay _at_ osbc.com or masay _at_ novell.com.

News: MySQL cozies up with SCO

OK, in true journalistic fashion, I've tried to put a sexy, attention-grabbing title on an otherwise not very interesting piece of news. MySQL (surprise!) is in the business of making money. Lots of money. As I've posted before, the company has basically doubled revenues each year for the past several years. Not bad.

I guess if we needed any confirmation that, in fact, MySQL cares about making money, check out this news from eWeek. MySQL is partnering with SCO to bring its database to SCO's OpenServer 6. Ugh. This must be one of those plug your nose and hold your breath sort of moves. I can't imagine that MySQL was happy to be associated with SCO, though the company is right to want to serve its customers. (More on this below.)

So, my personal preference would be to avoid the partnership simply based on principle. (My distaste for SCO runs deeper than most others', because I've dealt with the Canopy Group well before SCO ever thought of suing anyone. You know, back when Canopy was suing everyone but IBM. :-) But Brian Akers and others (see Arjen's comment below) have pointed out that MySQL's first consideration is serving customers and, as Horton the Elephant might say, "A customer is a customer...no matter how unfortunate they are to have bought into SCO."

This is probably the correct, principled path. I know (both from public and private comments by MySQL employees) that they don't have any personal love for the SCO litigation. Quite the opposite. But they're in the business of serving customers, and SCO's OpenServer crowd offers a significant base of customers needing a quality database...

...and SUSE Linux Enterprise Server. Dial 1-800-529-3400 for more information. :-)


[Note: This post has been expanded from its original form. My perspective wasn't as clear as I wanted it to be in the original form.]

Where open source fits

I received a thoughtful email from a good friend, responding to a range of my recent posts. In it, he took the trouble to lay down a general framework for understanding where open source fits, and how it grows. He said:

I'm a firm believer in open source where a problem has been comprehensively solved in theory, the theory is well known, the practice is easy to reverse engineer and the next stage is to sit around thinking of better and more efficient ways to apply that theory. The job of open source companies is either to package, or to innovate around the edges. The job of closed source companies is to innovate beyond the edges, to effectively create new edges. At some point, the frontier will move, and everybody shuffles on to higher ground (maybe too topical a reference though).

I'd say that contemporary operating systems are well solved. As are XHTML web servers. And PC browsers. In the blurry region (and probably about to sink) lie the desktop and office productivity suites, probably most application servers, and advanced email (in the sense they do calendaring and directory) servers.

Of course, it's not a free dynamic, the pool is small enough that big companies can directly affect the system through intervention, dragging new areas into the commoditized pool (viz IBM's support of Linux for high end server apps has ironically holed UNIX below the waterline, and Sun's open sourcing of StarOffice).

Anyhow, it's the job of entrepreneurs to find new edges. The open source bunch might find them as well, but there are various issues with commercial validation (of approach as much as of market) which mitigate against it).

Great thoughts, and I generally agree. Not that open source development communities can't find/create those "edges," but that historically they haven't, for whatever reason.

However, I do believe the rise of commercial open source tends to mitigate against the validity of some of my friend's assertions. For instance, SugarCRM has an explicit policy of developing intellectual property and then, over an accelerated schedule, injecting it into its open source product. Much of these developed innovations go directly into open source, without a layover in proprietary land. Alfresco, JasperSoft, and a range of other emerging open source vendors do the same in their respective markets. Each has an explicit strategy of undercutting established, proprietary competitors by freely releasing innovative source code, yet still charging for it in different ways.

I still don't think this completely obviates the points my friend was making, but it does seem to dampen them. Or does it? Thoughts?

More on the Massachusetts decision

Dave seems to be a bit punchy today. He doesn't even pretend to respect Microsoft (except for Bill Gates) in this post on the Massachusetts' decision and Microsoft's ungainly response. It's hard for people like me to defend Microsoft's right to exist when they say things like this (as quoted by Dave):

Alan Yates, GM of Microsoft's Information Worker Business Strategy: "As we look to the future, and all of these data types become increasingly intertwined, locked-in formats like OpenDocument are not well suited to address these varying data types - as the proposed policy itself acknowledges. It's this need for choice and flexibility that led Microsoft to design Office in a way that supports any XML schemas that a customer chooses, a capability lacking in less functional formats."

This put Dave to sleep, but I have to admit it gave me the giggles. Microsoft is many things, but earnestly seeking "choice and flexibility?" Even I, the resident Microsoft apologist, have a hard time taking that line seriously. Microsoft competes hard, and hasn't yet figured out that it can do so while not completely locking in customers. That's fine, but don't make the company look silly by pretending otherwise.

More to the point, Dave is right that Microsoft miscalculated in attacking the very customer that continues (by Massachusetts' CIO, Peter Quinn's own admission) to run Microsoft Office nearly everywhere. I think, generally speaking, Microsoft is starting to learn that it is steadily losing the ability to browbeat customers and partners into obeisance. But this is evidence that it's not happening fast enough.

Another example is Microsoft's refusal to allow a Mono BOF (birds of a feather) at the company's Professional Developer Conference. As Stephe points out, Microsoft's stance ends up doing it more harm than good. Perhaps, as Stephe argues, Microsoft needs to lend its "standard" credibility by having another company beyond Microsoft actually use it. More blandly, what would have come and gone without much of a ripple now has legs as a news story that doesn't seem to be going away. Last year Miguel set up shop at the PDC, had a great turnout and...well, not much happened. What is Microsoft afraid of, then?

A word of advice to Microsoft: Take a page out of Jason Matusow's book. That is, be a little less hyper-sensitive, a bit less defensive. Cooperate and collaborate more. It won't hurt. And it may well help.

News: Open source integration

Thanks to Dave for rubbing it in and pointing out this CNET article on open source enterprise application integration. He sent it via private email, so as not to embarrass me too terribly much. But I'm such a ball-of-integrity, I just couldn't keep it secret. ;-)

Back in June, I suggested that EAI might be the "next big thing" for open source. I still feel that way, and generally feel that open source (as a mostly neutral intermediary) has the right attributes to serve as "glue" to bring together both open source and proprietary software.

Dave, who knows a lot more about the technology, looked seriously at starting a company around it. Looks like he should have (and still could, I suppose, though often the first credible alternative out the gate tends to dominate). Sorry, Dave. You were right. Again. It happens. :-)

More on that integrity "thing"

I've been in deep soul-searching mode these past few weeks, trying to figure out where this open source ship is going, and what sort of passenger I want to be onboard. It started with my failed move to Volantis, if not a little before (causing the rift there which, I'm happy to say, did not injure my friendships with Mark and Matt).

Rereading some of my posts over the past few weeks, this questioning/indecision shows through. I spend more time taking open source to task than its alleged enemies (Microsoft, principally, but only because as people we like to single labels to describe all that is wrong in the world, however inaccurate or misguided these labels may be at times).

Au fond (there's no good English substitute, but it basically means, "at its foundation/core"), it's about integrity. Here are a range of definitions for the word, but it essentially means "wholeness" and, in particular, "living according to one's beliefs." I'm sometimes not a very good example of the word, and don't mean to apply the word to myself too strongly. But with regard to open source, it's had a strong effect on my decisions in the past few weeks.

For me, open source is about opening up development to a wider group of people than one's own employees, to tap into their expertise. It's also a great mechanism for those who want to take on incumbents in given markets (i.e., a sales/marketing/distribution tool). Done right, it makes great software at lower price points than traditionally developed software.

But there are many things it's not. It's not the end of Microsoft as we know it (or shouldn't be - Microsoft's biggest problem in all this seems to be that it has confused Linux with open source, conflating a specific project with a general development methodology). It's not the solution to world hunger. It won't raise third-world economies from the dust (any more than it will raise third-tier vendors from the dust). [Having said that, please read Nicholas' (aka "Yoda") take on one of my posts. You can find it here. I'll follow up with him on this, as I think he has insight that I don't understand well enough.] It's just a (great) development/distribution model.

Hence, I've grown weary with the constant hubris of the open source business and development communities. Some within these crowds probably feel that such exaggerated truth (which, last time I checked, translates to "lying") is warranted given the aggressive marketing Microsoft has done over the years, or given the value of the goal they have in mind (benevolent leadership in the industry, rather than lock-in). But I'm not an ends justifies means sort of guy. It conflicts with my integrity. (That word again!) I am somewhat naive in believing that honest, worthwhile ends are achieved through honest, worthwhile means.

This naive approach has not always served me well. I can remember at Lineo (embedded Linux startup acquired by Motorola/Metrowerks in 2002) when I was over our NetComm/Residential Gateway business. When meeting with Netgear or Linksys, I felt compelled to reveal our weaknesses relative to Monta Vista as much as our strengths. I assume they appreciated the candor, though they didn't always buy from us (perhaps as a result?). Same with Novell. There are things that Novell does well, perhaps better than any, but also things we don't. I can't lie about the latter (and probably don't trumpet the former enough).

Maybe this is why I never practiced law....

Anyway, sorry for the personal nature of the post. I just thought I owed some form of explanation for the past few weeks' of blogging. It's unlikely to change, but at least you now know why I'm wringing my hands in angst. :-)

The problem with Massachusetts

eWeek and others have reported the Commonwealth of Massachusetts recent decision to only publish Commonwealth documents in "open formats." This means that

...[S]tarting on Jan. 1, 2007, all electronic documents created by state employees [will] be saved in only two format types: OpenDocument, which is used in open source applications such as OpenOffice.org, and the Adobe-created Portable Document File (PDF). OpenDocument can be used for saving documents such as letters, spreadsheets, tables, and graphical presentations. It is the default file format for OpenOffice 2.0.

This seems like great news and, I think, on the whole, it is.

My one quibble? Not so much with the PDF requirement (except...see below), but with the OpenOffice file format requirement. 99.9% of Massachusetts residents who try to open such files are not going to be able to. Massachusetts may think its doing citizens a favor by publishing "openly," but how open is it if its citizens can't read them? If a tree falls in the forest and no one is there to hear it....

It may be that this is a short-term problem. (And, given that PDF is widely used, perhaps not a problem at all, as most people will simply choose that option rather than the confusion of OpenOffice file formats.) Or it may be that this is a short-term solution - aside from its openness, what reasons do we have for believing that OpenOffice will be around 10 years from now, instead of Microsoft Office? Do we have data that support this belief? Is it better to have an open but unused format, or a closed and widely used format? It's a policy preference, ultimately.

Again, I prefer open file formats and open standards. But I think Massachusetts' decision solves nothing and contributes little to open source's success. It may do the opposite by introducing confusion, leading to retrenchment in Microsoft products. (And, as I've opined before, why do we continue to think about the desktop with a very last century perspective?)

Another thought, since when is PDF open? It's an open standard, yes, but owned by Adobe (however benevolently). From the PDF guidelines, available here:

1.5 Intellectual Property
The general idea of using an interchange format for electronic documents is in the public domain. Anyone is free to devise a set of unique data structures and operators that define an interchange format for electronic documents. However, Adobe Systems Incorporated owns the copyright for the particular data structures and operators and the written specification constituting the interchange format called the Portable Document Format. Thus, these elements of the Portable Document Format may not be copied without Adobe’s permission.

Adobe will enforce its copyright. Adobe’s intention is to maintain the integrity of the Portable Document Format standard. This enables the public to distinguish between the Portable Document Format and other interchange formats for electronic documents. However, Adobe desires to promote the use of the Portable Document Format for information interchange among diverse products and applications. Accordingly, Adobe gives anyone copyright permission, subject to the conditions stated below, to:

• Prepare files whose content conforms to the Portable Document Format
  
• Write drivers and applications that produce output represented in the Portable Document Format
  
  
• Write software that accepts input in the form of the Portable Document Format and displays, prints, or otherwise interprets the contents
  
  
• Copy Adobe’s copyrighted list of data structures and operators, as well as the example code and PostScript language function definitions in the written specification, to the extent necessary to use the Portable Document Format for the purposes above.

The conditions of such copyright permission are:

• •Authors of software that accepts input in the form of the Portable Document Format must make reasonable efforts to ensure that the software they create respects the access permissions and permissions controls listed in Table 3.20 of this specification, to the extent that they are used in any particular document.
  
  These access permissions express the rights that the document’s author has granted to users of the document. It is the responsibility of Portable Document Format consumer software to respect the author’s intent.
  
  
• Anyone who uses the copyrighted list of data structures and operators, as stated above, must include an appropriate copyright notice.

I'm very willing to accept Adobe's best intentions, and find the license broadly written. However, I find it ironic that we're willing to trust them but not Microsoft. Sure, we have a few court cases that warn us off giving full credence to Microsoft's best intentions, but why would we be willing to place ourselves at the behest of any one company, if we're worried about openness? Maybe Adobe becomes the next monopolist, 10 years from now, on the back of the PDF format. Stranger things have happened. If we're worried about openness, then, why not just default to .txt or .rtf? File formats are not the only route to lock-in, even if we accept PDF as an open standard forever and ever.

I think the reason is that Massachusetts isn't worried about openness as a general principle, but rather as a specific defense to Microsoft. Peter Quinn says as much:

"Microsoft has remade the desktop world," Quinn said. "But if you've watched history, there's a slag heap of proprietary companies who have fallen by the wayside because they were stuck in their ways. Just look at the minicomputer business, for example. The world is about open standards and open source. I can't understand why anybody would want to continue making closed-format documents anymore."

I'm not a Microsoft apologist - far from it. But I don't have any particular mistrust of Redmond, as compared to any other company out there that is driving value for shareholders. Building on a quasi-open format like PDF may well be a better, clean policy bet than that of forcing citizens to buy Microsoft Office (because government documents are published in Microsoft Office's closed formats).

But let's call a spade a spade, and not euphemize it with "openness." The only reason he's defending his document file formats against Microsoft is because Adobe never figured out how to monetize PDF to the tune of $11 billion a year, like Microsoft has. We can begrudge them, but let's do so fairly and...openly.

Microsoft needs to accept Linux, says OSDL's Cohen

So, now it's not a matter of which is better, but rather of accepting the fact that both Windows and Linux need to cohabitate (who isn't these days?). Or so says Stuart Cohen, CEO of OSDL, in this SearchEnterpriseLinux piece. All true, but I wonder why Stuart expects Microsoft to accept these facts of life, when I'm sure he's fully hoping to see Windows obliterated from the face of the earth.

Not going to happen in the next five years, so let's stick with just living together for now.

If we're really going to be stuck in mixed-OS land for some time (and we are), companies like Centeris become incredibly important. I had breakfast with Barry, Centeris' CEO, this morning, and think highly of the company (and him). I like the fact that Centeris seems to view open source as a practical necessity, and not a religious necessity.

Same with Peter Yared at ActiveGrid (one of the most exciting open source companies on this planet, in my opinion, having spent time with them this week and having heard Irwin Gross of Worldview, who led the Series B round, proselytize them for the past several weeks). As the world begins its shift to the "light web," away from J2EE (like the shift from UNIX to Linux, it will take time, but seems to be underway) and toward LAMP, ActiveGrid is well positioned, and has a very pragmatic vision of how to get there. Not religious

I'm from Utah. I know all about religion. It doesn't belong in technology.

Open source's integration quandary

Jason, who earned the distinction of near blog extinction this summer (I think he posted negative-103 times), finally came out of hibernation today. About time. In doing so, he reminded me of something I heard yesterday in a quick trip through the Bay Area. The integration problem in open source.

I had been talking with an active LAMP user, who said that all he needed in terms of integration/certification/etc. he found in Bitrock's Installer product. Point and click the various components you want, and BitRock puts it all together in a nice, synthesized package. No testing required - they've done all that work upfront. OpenLogic does something similar.

It was great to hear such a solid testimonial for BitRock's technology. Daniel, president & CTO of the company, has been an active, early supporter of the Open Source Business Conference, and is a genuinely great person. Always nice to see kudos thrown his way.

But what does this mean for SpikeSource? And SourceLabs? Given that BitRock (and OpenLogic) don't have the marketing budgets to blast their stories to the world, maybe nothing. But let's assume they did: Would the world need SpikeSource and SourceLabs? I'm not suggesting an answer - I'm not qualified to do so. But it does seem that if BitRock and OpenLogic can do the heavy lifting upfront, then perhaps there is less need for someone to certify stacks down the line?

Jason, for his part, makes the valid point that Microsoft has invested heavily in taking the integration load off customers:

If OSS projects are going to be competitive with offerings from companies like IBM, BEA, Oracle and WebMethods (as stated in this particular article), they are going to have to solve the integration problem or get killed on the long-term cost comparisons with their competitors. Religion of licensing, purity of coding practices, rebelling against Redmond all fall to the wayside when a customer starts asking the hard questions about controlling costs. Software licensing is traditionally only 3%-5% of the overall solution costs, so where are the big dollars spent? Integration. Microsoft has been working on transferring the integration workload onto the technology's shoulders away from manual integration for years. This was not done by accident - it has been done to offer value to our solutions. I'm not saying we are there yet, but this is the right place to be making technology investments.

Think about it. The company that can take those integration costs off the table (be the software proprietary or open) a) is an attractive vendor, b) locks in a competitive advantage against others who force CIOs to pay heavily for someone to declutter the complexity of their IT systems, and c) becomes the platform upon which other, competing vendors must build if they want a shot at the enterprise. It's tough work, but genius if you can make it.

Book Review: Democratizing Innovation (B-)

Another day, another book. This one was Eric von Hippel's Democratizing Innovation. You can download and read the entire thing here, though I'd encourage you to buy it if you think it would be interesting.

The idea behind the book is interesting - innovation is rapidly returning to the world of users instead of being relegated to manufacturers/vendors. This basic premise is actually what made me realize that the United States, at least, seems to be retreating a little from its infatuation with homogeneity in what we wear, drive, and eat. (Blogged it here.

The first paragraph of his book reads:

When I say that innovation is being democratized, I mean that users of products and services - both firms and individual consumers - are increasingly able to innovate for themselves. User-centered innovation processes offer great advantages over the manufacturer-centric innovation development systems that have been the mainstay of commerce for hundreds of years. Users that innovate can develop exactly what they want, rather than relying on manufacturers to act as their (often very imperfect) agents. Moreover, individual users do not have to develop everything they need on their own: they can benefit from innovations developed and freely shared by others. (1)

The cynic in me already sees red flags waving all over the place, but von Hippel spends a few hundred pages trying to data-away those concerns. He doesn't succeed, but for those already convinced by the argument, they'll find plenty of citations to support their bias.

Still, the book does offer some interesting observations (whether or not they're true is another matter). Von Hippel posits that "empirical studies show that many users - from 10 percent to nearly 40 percent - engage in developing or modifying products" (2). This number seems implausibly high to me and, indeed, von Hippel keeps using the crutches of open source developers (to which his argument certainly applies) and kitesurfing (???) and other "extreme" sports for empirical support of his argument. Unfortunately, however, I would tend to view both of these as outliers - do they really reflect the kinds of activities that are typical to the majority of the population?

Even within these outliers, von Hippel is careful to point out that

Studies of innovating users (both individuals and firms) show them to have the characteristics of "lead users." That is, they are ahead of the majority of users in their populations with respect to an important market trend, and they expect to gain relatively high benefits from a solution to the needs they have encountered there. (4)

This makes sense. The ubergeeks of [Insert industry name here] will tend to be those dissatisfied with current products, with the aptitude to actually do something about it. (See also 22.)

Von Hippel celebrates such lead-user innovation, but he doesn't give us much we can do with it. He chides manufacturers for being somewhat inwardly focused (innovating incremental advances on existing manufacturing methods or feature sets), as noted here:

Users, having an investment only in a need specification and not in a solution type [i.e., they have a naked need and don't care how it happens to fit into the product set or roadmap of a given vendor - it may well lie completely outside that vendor's ability, intentions, and desires], independent of solution type used. Manufacturers, in contrast, want to supply custom solutions to users that utilize their existing expertise and production capabilities. (48)

Consequently, even outsourcing to a third-party custom-development shop will not give users the same degree of customization, because those third parties have an incentive to try to develop something that meets their needs as much as the customers (i.e., something generalizable that they can then sell to other clients).

Von Hippel eventually suggests that manufacturers can participate in this user innovation age by farming innovations from lead users. No problem. Isn't this what successful manufacturers already do? Did I miss a news flash? Yes, many navel-gazing organizations struggle to do this and, even when they attempt to do so, often fail because it's hard to translate user needs into a product (given that the vendor has only felt the need/pain secondhand, and so might not fully understand how to build the product to suit user needs - see 70). But these are process issues, not fundamental flaws in a manufacturer's world.

More to the point, isn't this the very lifeblood of entrepreneurship? Industry X is not meeting my needs; I have the ability and insight to develop something better; I get funding and create a startup to tackle the opportunity; I become BigCo through an IPO or am acquired by a BigCo; and other entrepreneurs arise to fill in the voids my formerly innovative product leaves. Sure, this system doesn't always work efficiently (tried to raise seed stage money recently? Ugh.), but it's operational, and understandable. It's not new, whatever von Hippel may think.

Von Hippel makes other arguments (e.g., Users demonstrate real incentives to freely share their IP-protected ideas - see Chapter 6), but the crux of his argument is already noted above. I think von Hippel has allowed open source to cloud his vision somewhat, and takes a somewhat too rosy view of it, as well.

We need to stop lionizing open source and simply appreciate the (many) benefits it provides. It is not a panacea. It is not the Messiah. It's just a great way to develop and distribute software. Isn't that enough?

News flash: Unilever doesn't worship Linux

You probably saw this already, but here it is again. I talked with Neil Cameron (Unilever's CIO) about speaking at OSBC a few months back, which he declined due to internal discussions around open source. Now I get a bit more context. I wish he would have agreed, anyway, as it's precisely this sort of open, honest debate OSBC is designed to foment.

Here are Neil's reasons for not worshipping Linux:

The consumer goods giant, which manufactures brands such as Domestos, Dove and Flora, announced plans back in 2003 to cut £66m from its IT budget by switching from a Unix server platform to Linux running on Itanium. The aim was to eventually migrate the company's massive SAP systems onto the Linux platform.

But...Unilever CIO Neil Cameron said the cost benefits of migrating en masse to an open source platform are no longer as clear cut as they were two years ago because of security and support issues.

"I was a bit concerned that we started to talk in terms of religion about this [Linux]," he said. "The world moves on. That was run on a financial model and the reality of a financial model is the landscape changes and the model doesn't work anymore."

But he said the emergence of Linux as a cheaper and viable enterprise option has been good for competition because it forced proprietary vendors to raise their game.

"It drives a bit of competition into the marketplace and stops suppliers being complacent. I think suppliers through open source have become more responsive. Suddenly I can do things with more proprietary products at a price performance that says actually the gap between that and open source isn't as wide as it was two years ago," he said.

Translation: We were able to opt out of switching to Linux because the very threat of doing so forced our proprietary vendors to give us fair pricing and deferential treatment."

I hope we're soon beyond the whole "Linux is cheaper" argument. I don't care. I don't buy things because they're cheap - I buy them because they provide value. If Linux does that, it won't matter if it costs the same as Windows. Its stability and flexibility will be reasons enough to buy it, whatever its relative costs to Windows. Unilever's experience simply indicates that Linux vendors are still telling a somewhat outdated story. Upgrade that story to the true value of Linux and this sort of backsliding won't happen. That's not to say that Unilever and others, at that day, will automatically bow down to Linux, but it is to say that facile excuses ("It's no longer as cheap as we thought") will no longer cut it, as CIOs will be making decisions based on Linux's true merits.

Open source: Back to an "agrarian" economy?

I finished Von Hippel's book last night. The book review will follow, but in the meantime, it spurred some thoughts.

At one time, the world's economies were essentially "flat": people essentially fed, clothed, and innovated for themselves. Eventually, people (like me) realized that they were terrible blacksmiths but could raise grain/food relatively well, while others became tailors, silversmiths, innkeepers, etc.

When we got serious about this division of labor and decided to push it into hyper-growth, we started the Industrial Revolution. Next, Henry Ford and his ilk fomented the mass production revolution. Eventually, we ended up where we are today, a homogeneous consumer society that eats the same mass-produced Twinkies (and sushi), wears the same mass-produced clothing, and drives the same mass-produced cars.

In the United States (and increasingly, unfortunately, elsewhere) we buy this homogeneity in the bowels of mass-produced suburbia at mass-produced malls.

But I've noticed something. At the higher end of the market, consumers have begun to revolt. Neighborhood markets, corner drug stores, and local dry cleaners are coming back. People seem to be getting tired of the dreary sameness of mass production, are willing to pay to escape it, and small businesses (or large chains heavily disguised as small businesses :-) are cropping up to make my corner of Salt Lake City feel like...mine, and your corner of Boston feel like...yours.

IT, it seems to me, is undergoing a similar change. We started out with custom-developed software, matured into mass-produced software, and are now maturing further into...custom-developed software. Open source is driving this trend.

Yes, many (perhaps most?) CIOs still demand commercial software. 'Who wants the headache of ever-changing, user-supported open source?' they say, blithely ignoring the army of system integrators, consultants, and others hired in to...modify and customize the proprietary, commercial software they've just bought.

It's not a question of whether CIOs want customization. They do. The real question is whether they want to start with flexibility and harden it to meet their requirements (open source), or whether they want to start with inflexibility and (try to) soften it to meet their requirements (traditional software). There are arguments for both, but cost, stability, and supportability don't necessarily cut one way or the other.

Dresdner Kleinwort Wasserstein (DrKW) is an example of a successful IT shop that has worked with both systems, and is increasingly opting for open source. In this article, its CIO (JP Rangaswami) indicates his reasons for accepting the relative fluidity of open source:

One who has invested time in open source is JP Rangaswami, global CIO at investment bank Dresdner Kleinwort Wasserstein, who said open source will continue to be attractive not because of the price but because of the flexibility, adaptability and mobility it provides.

"Free as in freedom, not free as in gratis. Democratised innovation is not an option. It is the only option and it is the only way," he said.

Open source, again, is not a panacea. It's just a return to the past where users, not vendors, controlled CIO's destinies.