Corel MAIL...Und?

Corel has announced that they're taking on Microsoft Outlook by unbundling its MAIL application (apparently in all CAPS to ensure no one thinks the product is boring) from the WordPerfect Suite (remember it?). This is supposed to be news?

Last time I checked, Outlook wasn't the problem. Exchange was. AppTran, Liquid Systems, and CanyonBridge are already tackling the Exchange problem in different but related (and all interesting) ways.

Outlook is just a client-facing application, and one that does its job fairly well. Exchange, on the other hand, is an expensive, difficult to manage beast.

Instead of fixing this problem, Corel completely sidesteps it. Listen to this, from the press release:

Supporting both POP3 and IMAP email protocols, and unlike competing offerings, WordPerfect MAIL is specifically designed for use on individual desktops–without an expensive in-house email server.

This is a feature? Let me translate:

Supporting the same basic protocols that every other email client on the face of the planet has supported since time began, and unlike other competing offerings that actually are useful, MAIL is specifically designed as a communications island that won't allow you to share calendars and other things you've grown used to with Exchange, Domino, GroupWise, etc.

Seriously, this seems like a complete waste to me. Why buy this when I can get a much better product (for free) in Mozilla's Thunderbird? Or the free email program that comes with every system, be it Windows, Mac, or Linux? This is a complete waste.

I also find it funny that Corel bills WordPerfect as the market leader behind Microsoft Office. I come from Utah County, where WordPerfect was born. While a few people there persist in using WordPerfect, the rest of the world has moved on to Microsoft Office. Those that move off Office or simply want an alternative on their desktops (like me), use OpenOffice. Not Corel.

Bees and open source

I was reading through Wikipedia today, trying to learn more about bees. I was watching them in my wildflower garden today, and wanted to better understand why they do what they do.

Wikipedia taught me about my neighborhood bees, as well as a little about open source. Does this sound familiar, or what?

The life cycle of bumblebees [the software industry] begins in the spring when the queen bee [traditional software company, e.g., Microsoft] rises from hibernation. At this time the queen bee is the one who does all the work because there are no worker bees to do the work yet. She searches for a place to build her nest and she builds the honeypots. She also does the foraging to collect nectar and pollen. Bumblebee colonies die off in the autumn, after raising a last generation of queens, which survive individually in found hiding spots. Interestingly bumblebee queens sometimes seek winter safety in honeybee [open source developers] hives, where they are sometimes found dead in the spring by beekeepers, presumably stung to death by the honeybees. It is not known whether any succeed in winter survival in such an environment.

In all this talk about the changing dynamics of the software industry, deference must be shown to the "queen bumbebee." She got the industry started, after all. Just don't look for favors come winter.... :-)

r0ml says...something

I'm never quite sure I grasp the finer points of r0ml's presentations. The guy is wicked-smart, starting at MIT at the ripe old age of 15. And he tends to use numbers and do complex things like add, subtract, multiply, and divide them. The English major in me just can't keep up with all that mathematics stuff.... :-)

This time, however, I know I can't comprehend a word he says, but since it sounds similar to a presentation he delivered at the Open Source Business Conference, and there I nearly grasped the universal oneness of his discussion, I figured I'd give you a teaser of an interesting (read: Matt is an idiot who can't grasp derivatives) article r0ml just published.

Here's a taste:

Those who have suggested that open source and free software is somehow not capitalistic, destroying the value of software and other such assertions, have missed this alternative explanation. It is just as likely that the free and open source software folk have stumbled across the financial engineering insight that a significant portion of the value of software is the embedded "derivatives"--options or warrants--on future maintenance and enhancement. Whether one believes that software has intrinsic value is related mostly to one's view on the correct value to use for volatility in calculating the option value. Larger values of volatility mean the software itself is worth less. Smaller values of volatility reduce the option price, and the software is intrinsically worth more.

Therefore, the major difference in worldview between open source advocates and proprietary software license advocates is explainable as a differing opinion on the correct value of the volatility of maintenance and upgrade pricing. People who believe that the pricing on maintenance is stable and unlikely to change see greater intrinsic value in the software. People who fear that the pricing is subject to large fluctuations see no intrinsic value in the up-front license; stripped of the options, the license value approaches $0.

For the open source movement, perhaps a better way to position the change that OSS is making is this: we're converting warrants on future maintenance and enhancements into options, which means that instead of having a sole supplier (warrants), we have created a third-party market (options) of these derivatives.

How capitalistic is that?

As a rough attempt at understanding, I think r0ml is pointing to a similar point to that which I made immediately below. Namely, open source charges for its value in different ways from traditional software, which was/is anomalous. Open source places a premium on getting software into users' hands, and then derives revenue from support/maintenance that naturally ensues (if it's good software and if the user wants a safety net). It's a sign of a maturing software market.

The natural order of (commercial) things

Every day it strikes me that software is becoming more and more like everything else. Doc Searls was the first to clue me in, with his analogy of the construction industry and how it relates to open source. But daily I see that software isn't only analogous to something fundamental like building houses/bridges/etc. It's not about engineering, at the end of the day. It's about building something and then marketing it, which is what every saleable product on this planet involves.

It used to be that software was different, privileged in many ways. With the rise of open source, it's clear that software really is just like most other saleable goods. The shoes I'm wearing. The couch I'm sitting on. Etc.

SpikeSource and SourceLabs were one of the early indicators of this trend toward normalcy. Whether or not, in the long term, they prove successful is somewhat immaterial. The point is, something like these companies will succeed, if not them.

Why? Because software will increasingly look like everything else. In the clothing world, for example, I generally don't buy my clothes directly from the clothing designer (which, as that name denotes, designs the clothes, but usually doesn't manufacture them). I buy them from a retailer, like Nordstrom or Gap. Now, both of these retailers also do their own labels (Gap, in particular, private labels all of the clothes it sells), making them look more like traditional software companies (Oracle, Microsoft), except that they likely neither design and certainly don't manufacture their clothes. They simply "OEM" them from manufacturers and designers elsewhere (largely in Asia). Their role is to market the clothes and sell them.

Software is following this trend, and will continue to do so. Web 2.0 is all about putting one's polish and brand on a wide body of others' work (mostly open source). Irwin Gross has likened this to what happened some time ago in the chip industry, and I think he's right. We'll continue to see this.

We'll also see the rise of marketing companies (Red Hat, anyone?) that do little "manufacturing"/coding on their own, but instead package and distribute others' work. SpikeSource, SourceLabs, and OpenLogic also generally fall into this categorization, though they write significant software that glues/synthesizes open source projects together. These are, effectively, the Nordstroms of the world. Much depends, therefore, on these companies respective abilities to cast themselves as their retail equivalents: Macys, Nordstrom, Old Navy?

These are the retailers. There will also, of course, be manufacturers (low margin, high volume), distributors, etc. Everything that we're used to seeing in the offline world, we'll see online, with software.

There's nothing wrong with this - it's how most mature industries function.

Some implications:

• Some of the most promising open source companies won't do anything "productive." They'll simply market and sell existing technology. This is one reason that I find many of the new startups interesting, which monetize open source projects that already exist. Some people are good at building things, and others are good at selling those things. Both are needed.
  
• At some point, the industry will mature to the point that open source developers will get paid directly for writing their software...likely by the distributors, and not by end-users. The analogy is contractors buying raw materials, or retailers buying clothing from manufacturers. This isn't a matter of IBM or Novell paying developers to contribute to Apache or the Linux kernel. That will likely continue. Instead, I'm talking about an open source CMS company paying the Mambo project for the right to use the code in some preferred way. Think the source code licenses won't allow this? Give the industry a few more years, and I think such a scenario is not at all farfetched.
  
• The biggest revenue winners will be those, as Tim O'Reilly opines, recognize that real value lies in "data" (customers who like your brand and will therefore pay a premium for shopping at Nordstrom; a network of active users who contribute information to a service (like Amazon), etc.). In other words, think more about marketing, and a (bit) less about the code itself. (Disclaimer: I'm completely incapable of writing a line of code myself, and so can be expected to privilege an art I actually understand. :-)
  
• Those contemplating open source startups should think forward, not backward. Don't try to emulate the software world of today or yesterday. Create the software world of the future. It's coming - the only question is whether or not you'll profit from it.

Motorola's new Crackberry?

So, Motorola today unveiled its newest Razr. It may just be me, but I've grown a bit bored with the Razr already, though I did think it was beautiful the first time I saw it. I bought one for me wife.

The problem, however, is that the software on the current Razr is terrible. Painfully so. Palm OS may be a bit Spartan for Motorola's taste, but its PIM functions are fantastic. Super easy to use and minimalist in all of the right ways.

With this new Razr (dubbed "Q"), things only get worse. Or, at least, not much better. This new phone runs Windows, which is a mistake, IMO. I don't have a knee-jerk antipathy to Microsoft software - quite the contrary, I think Office is an exceptional product. But Windows? It stinks. It's ugly while striving to be sleek and, quite simply, tries to take over whatever device on which it's installed. It's just an operating system, Microsoft - let it do the drudgery it's meant to do in the background, and stop trying to upstage applications, which are the primary reason people buy computers/phones/etc.

Anyway, another potential downside to the new phone is the very thing that makes the original Razr so compelling: size. Motorola is billing the Q as 50% thinner than its top rivals (Treo and Blackberry). Is this a good thing? At some point, it becomes difficult to hold and type on something that is super-thin. I think the Treo is near-perfect in the way it feels when typing. I can't imagine going thinner.

Richard Stallman on Sun

I don't think I've ever made it a secret: I think many of the "old guard" open sourcerors are...outdated. Eric Raymond (as well as Larry Augustin and others at that formative meeting back in 1998, or whenever it was) kicked it off by aggressively promoting the term "open source," and the open source community forever changed.

For the better, in my opinion. When The Movement became about "open" source, and less about "free" source (as in freedom, to repeat that tired cliche), it took on a momentum of its own. At the end of the day, it's just software, and not some fundamental, sacred freedom. No one loses when I write proprietary software (except, perhaps, me, as I've written elsewhere - there are real business advantages to passive-aggressive licenses like the GPL, which liberate as much as they restrict). No one necessarily gains when I write free software.

Richard Stallman, grandfather of the "old guard," still can't (or, rather, won't) get his mind around this fact. That's fine. He has relegated himself to the fringe of the larger open source movement, and probably does some good by keeping the mainstream in the mainstream. Extremes are helpful to marking the boundaries of the maintstream.

All of which makes his commentary on Sun so interesting. Rather than simply lambast Sun, as he is wont to do with commercial entities, generally, Stallman takes a (slightly) more measured approach. He finds that (surprise!) companies, like people, are a mix of good and bad. What he doesn't admit, but which I've found to be true, is that on the whole, people (and companies) are more good than they are bad - we all have faults, but our positive attributes generally far outweigh our negative attributes.

Stallman decries Sun's stance on Java (with which I mostly agree), and grudgingly appreciates Sun's OpenOffice. Why he cares that Sun also has StarOffice, I can't fathom - it seems like we should be grateful for the generous contribution OO.o is, rather than ruining that gratitude with, "But they didn't give everything!!!" That reminds me of my somewhat uncharitable view on charity, which I sometimes allow to cloud my thinking: "It's not what one gives [that determines their generosity], but what one keeps for oneself." This creates an impossibly high, and rigid, metric by which we judge others, including companies. It's simply not a fair measuring tool, because we can never fully appreciate why someone, or Sun, gives this but not that. (At Novell, this is a constant issue. There are simply a range of complex reasons for every contribution we make, as well as those we don't.)

Anyway, enough. Here's Stallman's piece:

How does 'that company' treat free software?

Richard Stallman
Special to ZDNet UK
July 25, 2005, 15:30 BST

People often ask me what I think of some company's attitude towards the free software community. They would like a simple answer, and in a few cases, such as Microsoft, I can give them one. However, in most cases the right answer is complex. Even a small company can have several activities at once, each affecting our community in a different way. A large company can have even more.

There are many ways a company's activity can affect our community. A few companies sue or threaten free software developers, using software patents, the DMCA, or other legal weapons. A larger number of companies help our community by developing free software. Companies can also affect our community by talking about or distributing free software; but unless they do so in a very big way, this will have less long-term effect than their decision to contribute software or not.

Consider the example of Sun. How does Sun treat the free software community? It has not sued developers (though it has software patents with which it potentially could, and has not promised it won't). Sun probably redistributes some free software, but not in a way that attracts a lot of notice. The principal effect of Sun's activities on our community results from its own software development. But even this is not uniform. In one area, it makes a major contribution in a partly problematical way. In another, it cooperates but in a way that can't do much good; in a third, it holds us back. People who judge by one area alone are like the proverbial blind men who touched different parts of an elephant and gave conflicting reports on its shape.

Several years ago Sun bought StarOffice, a proprietary office suite. Sun released a version of this program as free software, under the name OpenOffice.org . This greatly extended the capabilities of free software. However, Sun continues distributing StarOffice as a proprietary program. The two are not the same, so there is a certain tendency for OpenOffice to act like demoware for proprietary software. Nonetheless, it is very useful in and of itself. It is a major contribution with a cloud over part of it.

By contrast, the release of Solaris as free software is a contribution that looks large, but actually helps little. Solaris is a Unix-like operating system; I don't know whether the source code Sun released is the whole system or not. Either way, it doesn't advance our capabilities greatly, because we already have two free software Unix-like operating systems. These include the GNU/Linux system, completed in 1992 when the kernel, Linux, filled the last gap in the GNU system, and the BSD system. (Each of the two has multiple variants.) Having a third one doesn't enable us to do much that we couldn't do before.

A Unix-like operating system is so large that there must surely be parts of Solaris that are better than their counterparts in GNU/Linux. However, the peculiar incompatible licence used to release Solaris as free software mostly prevents us from incorporating those parts. Thus, our community stays with GNU/Linux, and gains little or nothing from this contribution. Now that Solaris is free software, there's nothing unethical about it, but it is not much as a contribution. Fortunately, as GNU/Linux works pretty well, this is no disaster.

The Java situation is much worse. Sun's Java platform is completely proprietary. Because of this refusal to cooperate, we have to implement a free replacement from the ground up, just as we did with Unix starting in 1984. It consists of the GNU Compiler for Java, and GNU Classpath. It works, but it isn't complete: many Java programs run on it, but many others still do not.

Many programmers choose Java because they have heard it is "platform independent", but unless they are careful, they will find that is not so — that they have written programs that run only on Sun's non-free platform. (We call this the Java Trap.) If you like the Java language, please help us liberate it: join the development of GCJ and GNU Classpath.

So what can we say about Sun? Can we add up these three very different comportments and get an overall measure of how a whole company treats the Free World? Maybe we could, but I think we should not try. Any such combined measure would be simplistic. Except for those companies that do something so nasty that it calls for special outrage, such as Microsoft, Siemens, Philips, Ericsson and Alcatel — all reported by newspapers to have threatened to move or cancel operations in various European countries if those didn't support software patents — and when Adobe got Dmitri Sklyarov arrested, we should decline to "add up" all the activities of one company, decline to judge it "as a whole". It is more useful to judge each activity separately, so we can praise or criticize it as it deserves. I wrote this article because when I was asked to comment on Sun, I forgot this point. We all make mistakes — and we can use them as examples to teach others what not to do.

Copyright 2005 Richard Stallman. Verbatim copying and distribution of this entire article are permitted worldwide without royalty in any medium provided this notice is preserved.

Stephe pulls the gloves off....

Though this probably won't win me any friends or influence people, I think Stephe Walli is right by calling into serious question the premise underlying the various open source risk managers out there. Stephe hammers on Palamida in this piece, but he's written before about the general risk management cottage industry.

It's very similar to something Stephe's work colleague (and our mutual friend), r0ml Lefkowitz, will be addressing at this Fall's Open Source Business Conference, in his session "What's Wrong with Open Source?" We hear open source is risky...from vendors peddling open source risk management tools. We hear it's flawless (Secure! Cheap! Solid!)...from vendors peddling open source software. And so on. It's a dilemma, and one on which I'm excited to hear r0ml speak.

Tooting the OSBC horn, we've actually planned a range of sessions that will highlight what CIOs think about open source, and force vendors to justify their claims to those IT buyers. Should be interesting.

The secret of starting an open source startup

In an insightful piece, Dave Rosenberg writes in Release 1.0 about the various business models undergirding open source startups today. He also goes one step farther, pointing out the most successful models. I've written on this before, and Dave's findings seem to be a dead-on hit to me.

Here is the abstract:

If you were to contemplate starting a software company in today's market, you might consider one of many open source models. Clearly the IT marketplace wants open-source applications. Why else would they keep moving up the stack and continue to take market share from proprietary software companies? Databases, ERP, CRM, business intelligence, mobile applications - there is no category that can't spawn an open-source counterpart. There's a low barrier to entry, adoption and business acceptance is on the rise and there is a huge developer community to help build your product. Ah yes, the revered open source community, the invisible mass that supports projects and leads to success! If only you could harness the power and interest of the community, this whole thing would be a cakewalk.

Definitely worth a read. You have to give your email address, but I'm told (by Dave, who is a pathological liar ;-) that they don't do anything with the email addresses.

The secret sync

So, I finally figured out how to get my Mac syncing again with my Treo 650. Apparently, just as formatting one's hard drive is the easiest way to get rid of pesky viruses, so, too, is it often the easiest way to fix a ruined Microsoft Entourage database by simply erasing all of your data. Microsoft should package a self-destruct button with its software. :-)

Seriously, I finally ended up removing every repeating calendar entry in my calendar (figured out that Palm OS can't handle events that repeat 30 years - I cut everything off at 25, just to be safe), including some random events that I found in 2011 - 2014...but not in any of the years previous to them (or after them). The data just got screwed up. But Microsoft had the bright idea of removing all data that I actually wanted to sync - no more data, no more sync problem! :-)

Anyway. Matt's support tip for the day: Try rebooting. Then try erasing everything. If that doesn't work, try strapping a stick of dynamite to your machine.

Finger-pointing support

I'm on the phone right now with Microsoft's Mac support group. I've been having trouble syncing my Treo 650 to Entourage, and finally abandoned Google as my support center. While there have been a range of problems for others with Entourage-Palm sync, none seemed to repair my issue.

So, I called. I have to say, I've found Microsoft's support group to be quite good, overall. Darren, who was helping me (from Canada), was no different. He actually couldn't fix my problem on my first hour-long call, but scheduled a follow up with me to make sure it got resolved. (Pretty impressive service, if you ask me - every time I call the support group at Cingular, they tell me that if by some odd chance my cell phone drops in the middle of the call, I'll need to dial them back to start over. I think there must be a big button on their desks that they push to drop the call when they (as is usually the case) can't resolve the problem.)

Still, my gripe is that we eventually discovered that my Entourage database (over 1GB of data) had become corrupted. This, after checking on my Palm conduits and basically asking about problems with every piece of software on my computer except Entourage. It seemed like Microsoft was trying to look for trouble with everything except its own products.

This is not a Microsoft thing, either - tech support seems like it's set up specifically to look beyond one's own faults to the faults of other products on a given system, pointing the finger elsewhere. (If you've been reading this blog long enough, you may remember my Internet set-up fiasco in London. Only when the hotel, after hours and hours of my telling them that the problem was on their site, looked in their network closet did they see that my room connection was unplugged.)

For those open source support companies out there - those companies whose primary business is service - may I make a humble suggestion? Assume that you're the problem. Always. Physician, heal thyself, before pointing fingers. If we can deliver that kind of support, buyers will come.

VCs funding the wrong guys?

Dave is now 2-for-2 today - what's up? Is he finally getting productive? He has an interesting blog entry today about VCs funding industry executives to start and run open source software concerns. (Examples include Alfresco (Accel) and a new open source ERP company, started by Dave Duffield (PeopleSoft co-founder), currently housed here.)

I list these because the founders are so impressive, that you just want to believe they'll succeed.

But see my last post. I think it's relatively easy for an intelligent person to learn a new industry. Happens all the time. I think it's much more complicated for anyone, intelligent or intellectually-challenged (dumb), to learn a new community. In this case, open source. It's not that it really matters what Miguel de Icaza, Alan Cox, etc. think of you. It's more that open source requires a certain mentality, which doesn't come quickly.

I should know. I spent two years at Lineo trying to ramrod a proprietary strategy onto an open source (GPL-based) business. Open source = 1; Lineo = 0. I learned, but then watched Novell struggle through a massive cultural shift from a proprietary/NIH mindset to where it stands today (amazingly open source minded). It took two years, and was excruciatingly painful.

In short, it's not for the faint of heart, and not something you can just pick up by reading a few books. Open source is a community mentality that simply isn't easily adopted, especially by those whose history lies in proprietary success. So, borrowing from Dave Rosenberg's theory of open source business, a VC's best bet is to invest in a founding technologist - that is, the person who founded a successful project. Look around at the successful commercial open source companies today (JBoss, MySQL) and you'll see that this is what drives them.

Open source marketing and the VP of Community

Dave gets it right in his piece on open source marketing. Dave makes the point (among others) that open source startups (corporate or project) can either spend a ton of money on marketing, or they can build community to accomplish the same end: generating use and contributions back.

Communication is key

Now that you've read this little slice of genius [re the importance of community building], the obvious question is: "What do I do next?" Let's review what we've learned so far:

• Developer marketing is different than other marketing, but it's often easier to achieve, because you are talking to peers with whom you relate more directly.
  
• Support your own efforts by communicating on newsgroups, Web sites and anywhere else anyone is interested in similar subject matter; don't underestimate the value of sharing your expertise with others in the community.
  
• Blogs, t-shirts, and developer events are inexpensive and easy ways to get your message out.
  
• Odds are there is another company or project that you study to obtain ideas.

In the end, it's the dialog that you have with your current and future user base that will drive the success of your project or product. The open source community thrives on the reciprocity between product developers and those who support the efforts. Having consistent, honest communication with your constituents is the first step to launching a marketing effort that will help catapult you to success.

Good points all, which actually leads me to something I've been wanting to blog recently. Every ISV needs an open source (business and developer) community expert. The more companies with which I talk, the more it's clear to me that lots of people have a superficial understanding of open source, without any real clue as to what it means. I could point out examples, but I'm not interested in mud-slinging. Needless to say, any open source startup that hopes to succeed needs a genetic comprehension of open source and street cred within that world. Those that lack these traits will stumble along until they fail - in the end, they'll realize that there really is something more to understanding open source than reading Red Hat's financials and Eric Raymond's The Cathedral and the Bazaar.

This has been as true at Novell as elsewhere. I remember one of our senior Linux executives asking me what I did for the company. I told them the nuts-and-bolts of what I do, but what I should have said was, "I make people take you seriously." They had little clue as to how open source functions, and saw Linux as just another market trend to monetize. Problem is, monetization of open source is wickedly difficult if you haven't a clue about its mechanics and lack relationships within the community. That's why IBM started building its presence in the Linux and Apache communities years before the company tried to leverage them into its business - IBM understood that open source is all about community, and community is something you belong to...or not. If not, don't bother knocking.

2.6 billion mobile phones by 2009

Daily Wireless has a great take on Gartner's recent revelation that 779 million wireless handsets (phones) will ship this year, and over 1 billion by 2009.

Some figures to chew on:

• In 2005, mobile phone sales will reach 779 million units, a 16 percent increase on 2004.
  
• More than 100 million 3G phones will be sold in 2006.
  
• More than 200 million smartphones will be sold in 2008.

North America is used to pretending its a mobile market. Europe and Japan are used to smugly knowing that they're the real markets. But the action going forward? Asia and (gasp!) Latin America:

Asia/Pacific accounts for most sales: one in every four mobile phones sold this year. In 2009 this will increase to one in three. Ann Liang, Gartner's principal analyst for mobile terminals in the region said, "China and India alone will account for nearly 200 million units in 2007, with the Indian market surpassing China in 2009 to reach 139 million units."

North Americans are still buying the latest models, but the bigger story is in Latin America. Hugues De La Vergne, Gartner's principal analyst for mobile terminals in the Americas, said "Sales nearly doubled in 2004 within Latin America and they will reach 100 million mobile phones a year by 2009. Brazil is the powerhouse of the region, accounting for more than a third of sales this year."

So, what does this mean? It means, for one thing, that American VCs (who fund much of the best technology) need to look beyond Silicon Valley for their markets and, most likely, their investments. The best vendors will often be those closest to the action, and not merely the bored entrepreneurs who made money off them the last time.

It also means, as Jonathan Schwartz opined at OSBC West 2005, that "The majority of the world will first experience the Internet through their mobile phones." This web may be the disconnected/connected web that Fabrizio at Funambol sees, or it may be the mini-browsing experience that Volantis envisions. Whatever your preference, I believe that open source will undergird that world, from the framework in which mobile apps are developed and delivered to the client-side apps that make the mobile web fun. The future of mobile is open source. Blessed be the venture capitalist who figures that out.

War of the Worlds

Bryce Roberts and I went to see it last night (after we had put our kids to bed). Highly disturbing. Terrifying, even. Mostly because of what it says about human nature, pushed to extremes. I think it does a good job of showing the negative, and not a good enough job of showing the positive (Different people would react to the extremities of alien extermination in different ways: some noble, some savage.) Worth a watch, though it's not one I'll see twice.

In response, Stephen Walli sent me this link today, which provides an interesting spin on Google and its "Do no evil" policy.

All of which makes me think of some of my favorite literature. Here's one of my favorite passages from any book: Joseph Conrad's Heart of Darkness, which follows the "hunt" for Mr. Kurtz in the heart of the Belgian Congo. It was an anti-colonialist piece, and fantastically absorbing. You should read it. Here's a taste, as the narrator contemplates Kurtz's death and the ignominy of his own life:

Destiny. My destiny! Droll thing life is -- that mysterious arrangement of merciless logic for a futile purpose. The most you can hope from it is some knowledge of yourself -- that comes too late -- a crop of unextinguishable regrets. I have wrestled with death. It is the most unexciting contest you can imagine. It takes place in an impalpable greyness, with nothing underfoot, with nothing around, without spectators, without clamour, without glory, without the great desire of victory, without the great fear of defeat, in a sickly atmosphere of tepid scepticism, without much belief in your own right, and still less in that of your adversary. If such is the form of ultimate wisdom, then life is a greater riddle than some of us think it to be. I was within a hair's-breadth of the last opportunity for pronouncement, and I found with humiliation that probably I would have nothing to say. This is the reason why I affirm that Kurtz was a remarkable man. He had something to say. He said it. Since I had peeped over the edge myself, I understand better the meaning of his stare, that could not see the flame of the candle, but was wide enough to embrace the whole universe, piercing enough to penetrate all the hearts that beat in the darkness. He had summed up -- he had judged. "The horror!" He was a remarkable man. After all, this was the expression of some sort of belief; it had candour, it had conviction, it had a vibrating note of revolt in its whisper, it had the appalling face of a glimpsed truth -- the strange commingling of desire and hate. And it is not my own extremity I remember best -- a vision of greyness without form filled with physical pain, and a careless contempt for the evanescence of all things -- even of this pain itself. No! It is his extremity that I seem to have lived through. True, he had made that last stride, he had stepped over the edge, while I had been permitted to draw back my hesitating foot. And perhaps in this is the whole difference; perhaps all the wisdom, and all truth, and all sincerity, are just compressed into that inappreciable moment of time in which we step over the threshold of the invisible. Perhaps! I like to think my summing-up would not have been a word of careless contempt. Better his cry -- much better. It was an affirmation, a moral victory paid for by innumerable deaths, by abominable terrors, by abominable satisfactions. But it was a victory! That is why I have remained loyal to Kurtz to the last, and even beyond, when a long time after I heard once more, not his own voice, but the echo of his magnificent eloquence throw to me from a soul as translucently pure as a cliff of crystal.

Marketing 101

I'm working on some press releases for Linuxworld San Francisco, which has really forced me to think through who Novell is targeting with its range of products. This is true of any company, of course: you want to interest the market in your product, but you also want to be sure to interest the right market in your products. In many cases, it's a very different market from SKU to SKU.

Linux is Linux is Linux, right? Yes and no. Yes, in the sense that it comes from the same source tree and yes in the sense that it generally follows similar conventions (Available desktops, etc.).

But no in the sense that SUSE Linux 9.3 (current rev) might not be appropriate for a large enterprise, or for the average home user. We have products for those groups, and try to steer the right customers to the right products. The consequence of poor marketing is that buy the "wrong" product, are unhappy with it, and end up flaming you for the crappy product.

As an example, I have a certain glowering friend who constantly harps on me that SUSE Linux crashes, doesn't support X, etc. (Actually, his primary gripe is normally with Evolution and its support for Exchange.) I believe, however, that he continues to use SUSE Linux 9.3, which is intended for the home enthusiast (which he is), within his company (which would normally call for Novell Open Enterprise Server or Novell SUSE Linux Enterprise Server (SLES). These are the products with enterprise-class support and tighter integration - 9.3 is, by its definition, intended to be on the vanguard of technology, which will always entail a bit of pain in supporting it.

In a similar way, Red Hat intends Fedora for a different audience than it does Advanced Server. Marketing.

Having said all this, I blame the vendors (including my employer), and not users, for failing to make the distinctions clearer. To that end, I'm going to do something that most press releases don't do: I'm going to speak plainly and try to avoid hyped verbiage. We'll see if Marcomm will tolerate that. :-)

The fallacy of community

Listening in on the mounting hoopla surrounding Debian and its "wayward son," as Ian terms it, I can't help but think:

Do we really need another community-building exercise?

Ian seems to think so. He's out campaigning for a Debian Core Consortium, as reported here. I've talked briefly about it with Ian, and I'm still trying to determine why we need this or, more to the point, why he needs it.

On a psychological level, I understand it completely. One can claim to have the original Linux tree named after him (Linus), but only Ian can take credit for a distribution named for his wife and him (Debian). Not sure I'd want to see my project forked until I had the separation papers sitting at the attorney's, and I don't think that's in the cards for Deb and Ian. :-)

But beyond the personal, what does Debian gain from this? It seems quite natural to me that distributions fork over time, and no amount of rallying the troops is going to change the fact that a few cadets want to go AWOL. Red Hat went so far as to fork itself (though it's still unclear to me who, if anyone, benefits from Fedora - seems like a half-baked gesture to "community" in an attempt to stop serving that community). But forks, like itches, are ever with us in open source. It's as it should be. Or, if not, it's how it is. We just need to get used to it.

Do I wish that there was just one big united Linux? (Pun intended. :-) Sure. Then we could focus on real customer issues, almost none of which are in the OS layer. But I don't see that happening, and I really don't want any one vendor (including Novell) to control it. I don't like Microsoft's monopoly of desktop operating systems; I'm not sure why I should like Novell's more.

So, human nature being what it is, we just need to get used to fragmentation and forking. In many ways, it's a positive thing - at least in the open source world those AWOL cadets have the freedom to leave. In the proprietary world, they'd be pirates.


P.S. When will we stop talking about "community" as if it means something? What we're really talking about is "customers," I believe - customers that happen not to pay some of the time, and who help us build our products some of the time. They are, in other words, just as self-interested as we are in creating the project/product in the first place. There's no such thing as a large, amorphous community of developers that is sitting by the terminal, waiting for us to release code so that they can start hacking it. There may be self-interested customers who will pitch in when they can, but there's not a pool of developers that is getting siphoned from one project to another.

In other words, I'm not sure the "community" is as zero sum and finite as we sometimes pretend. In my opinion, there's a huge population of would-be "community" members who have yet to join, simply because the extant projects don't appeal to them, or block them through overly technical hurdles to participation, etc. Rather than fighting over this limited pool (of Linux developers), maybe we should be focusing on opening up the "community" to this much wider audience? Just a thought.

So, you've been wanting to try a Linux desktop?

So, it's the wrong one (Red Hat ;-), but it's still a pretty cool way to quickly try out a Linux desktop, without installing anything on your computer. It's a Java app that runs in your browser. Work Spot. Check it out here. I first discovered it 2-3 years ago, but the developer has made some interesting changes to it since then. Pretty cool.

Cool IM client for Mac people

There are other IM clients for the Mac (including the super-solid iChat), but I just discovered a new one this morning that I'm really liking. r0ml et al: Listen up.

Check out Fire, a GPL'd multi-protocol IM client. I like the interface and the security features. No, I don't use the latter (though after reading this morning's WSJ about how identity thieves use IM because it can be more secure than email, I kind of want to). I also really like the "tabbed" approach it takes to windowing, unlike iChat (and other clients) which pops little windows all over my screen.

While I'm on this tangent, I like IM, generally, for a few reasons:

1. It allows me to push an inquiry onto someone's desktop. This is largely the reason I sometimes use SMS instead of email. There are times when I need a quick reply, and this requires interrupting someone's work and not talking with them (which requires much more chit chat before you're allowed to get to the point).
2. Presence. I can tell when someone is available, which mitigates (somewhat) their annoyance at being interrupted (most likely in playing online poker, so I don't feel too bad.... :-). For the life of me, I can't figure out why wireless carriers haven't added presence to their offering. Surely they could clue off whether the phone is set to vibrate to let others know if they're available for a call.
3. It's informal. This goes back to the chit chat comment, but it also goes deeper. While typos in emails are relatively common (because, apparently, people have not discovered the spell check feature), they're absolutely acceptable in IM. This allows for a more free-flowing conversation where it's less about how you say something, and more about the substance of what you say.

I'm sure there are others, but that's my short list.

OSBC: Content over cash

We're going to be having some keynotes at this year's OSBC Boston (November 1-2) that will surprise some people. With each conference, we've become better and better at selecting exceptional speakers, though this has remained difficult at the keynote level. Why? Because top sponsors expect keynotes, and we've had to capitulate at times.

This time around, however, we're reaching out to the best speakers in our sponsors' organizations, and I think we've pulled out some gems. For example, Dirk Hohndel of Intel will be keynoting, a departure from our policy of having VP-level and above speakers, but more in line with our policy for having the best, most intelligent speakers on the planet speak.

I've long wanted someone from Intel to talk about what they did in the chip market (commoditize it into...a much larger market than it ever would have been had it remained a proprietary landgrab), and Dirk offered up a proposal that sounds fantastic:

Growing an industry

Many see commoditization as taking money out of a market segment. But as the past has shown, quite often commoditization simply moves the value line - and in the process can create new opportunities for growth. Intel's move to provide industry standard building blocks in the past has created the multi billion dollar computer industry by establishing a standardized environment upon which a new eco-system could grow - quite different from the vertical computer industry in the past.

The software industry is at an inflection point today, with commoditization and new business models significantly changing the environment in which companies compete. How can we learn from the past and make this transformation successful? Where are the opportunities and where are the risks for the industry?

Sounds awesome, no? Even better when delivered by a great speaker like Dirk. I can't wait.

I'm also still expecting Jason Matusow, who runs Microsoft's Shared Source Initiative, to keynote. Again, he's not a VP. He's a director, like Dirk. But I've never heard Jason give a boring presentation, and he has promised me to deliver something provocative. I'm still hoping for "Windows as an Open Source Platform," but we'll see.

Anyway, we'll also be having Ron Hovsepian (EVP, Worldwide Operations, Novell) keynote. Ron is a nuts-and-bolts kind of guy. I'm looking forward to hearing him deliver a very pragmatic presentation on open source's route to the data center. (Ron has been the primary driver behind Novell's go-to-market strategies for its Linux business.) Plus, someone from Sun will be keynoting again - I just hope they can live up to the standard Jonathan Schwartz set in San Francisco earlier this year. We'll also be hearing from the illuminative and energizing Eben Moglen, general counsel for the Free Software Foundation (and an incredible speaker).

Should be excellent. Hope to see you there.

Morning glory among us

I spent two hours in Utah's 100-degree Fahrenheit heat, pulling weeds in my wildflower garden. I "planted" the garden (Actually, I just tossed a bunch of seeds into a grassless patch in our backyard) so that I could have something by which to remember Stanford's Spring wildflowers. A few months into it, however, 'morning glory' had taken over.

Morning glory is, as nearly as I can tell, the devil incarnate. It looks pretty (little white flowers all over my garden, in my case), but sucks up moisture and winds itself around the other flowers, choking them and pulling them to the ground. And, unfortunately, despite the time I spent yanking it out, it will be back within a week or two. It's nearly impossible to kill.

Some of the open source companies currently getting funded remind me a little of morning glory. They look pretty but ultimately contribute little back to the open source well from which they're deriving sustenance. The best open source strategies are those that include a heavy component of community and code development; the worst are those that siphon off good code, give little back (besides lipservice, like the pretty white flowers on my morning glory), and basically serve as parasites to the industry.

In my experience, most companies aren't like this. I've actually been surprised by how dedicated many new startups (EnterpriseDB, JasperSoft, SugarCRM, and a range of others) are to the open source community. But there are the weeds, as well. I assume they'll get discovered and discarded over time, but it kind of bugs me that they're taking space in the open source garden until then.

A company is what its employees allow it to be

On my flight home from San Jose tonight, I sat next to a guy who works for Linux Networx. We got to talking about various people there (I did some consulting for the company two years ago, and so knew a range of people there), and about the company's corporate direction.

I remember pushing the company to balance their business with the national labs (Big deals, but heavily competitive and low margin) with smaller sized enterprise deals, much as Rackable has built a $100 million business doing. They wouldn't budge. Didn't want the enterprise business. They served the national labs, and that was all they were going to do.

i still hold to my original premise, though the company has done well with its labs-focused strategy. I'm not sure there's a right or wrong in the strategy, but what interests me most is the institutional inability the company had to consider an alternative to its primary target customer. There were channel/business reasons for sticking with what the company already knew how to do, of course, but the primary driver behind the company's decision was its culture. Its people. They knew the labs business, and were uncomfortable shifting gears toward a different business.

Sometimes such stubborn adherence to a strategy is good policy. Perhaps most of the time. But often, as I saw early in Novell's shift to Linux, a rigid adherence to traditional ways of doing things is counter-productive. We like to stick with what is familiar to us, but that is sometimes a recipe for failure as the market matures and moves beyond us.

That said, companies are simply agglomerations of people, and people don't change easily. Management can perform songs and dances to get employees behind new policies or strategies, but the company will generally lumber along a path until the employees slowly, one-by-one, determine that they want to move. I'm not sure there's much a company can do to affect this, either.

I certainly can't do much to change my children when they determine they want to do X or Y. They just do it. I can threaten, but this generally results in Greta reminding me in her two-year old way that I'm an insignificant bug that happened to splatter into her windshield. Next time you want to change your company's focus, just remember that you're dealing with a bunch of Gretas - they'll follow when they're good and ready.

The value of evangelism

I was lamenting to a friend (and Novell executive) the other day that I didn't feel that I had done much for the company in my three years there. His response was flattering, if not accurate: "Most of the folks I talk to credit you with putting Novell on the open source map." This wasn't altogether true - Chris Stone, more than anyone else, turned the Novell ship into open source waters - but it made me feel like the miles I've logged speaking at conferences, the articles I've published, etc. have done Novell and the industry some good.

I can actually remember quite vividly the first open source event I ever attended. Enterprise Linux Forum in Boston, Fall of 2002. This was nearly a year before we acquired Ximian, and open source was probably not on the radar screen of more than a small handful of Novellites. (Funny what three years can do to a company.) I actually had to pay my travel (Air, hotel, car) and admission fee ($795, if I remember right) to the conference, and had to take vacation time. Why? "Novell doesn't do anything with Linux or open source, and so can't reimburse [me]."

From my perspective, it had everything to do with Novell's future - it was the only way that we could become relevant. (I felt this acutely, because my role at Novell was to attract developers to the NetWare platform. After two days with the company, I emailed Chris Stone - then vice chairman - and asked him why we weren't doing Linux, instead. He started asking my opinion on things (like a possible Ximian acquisition), and I then helped to co-found our Linux Business Office (responsible for open source strategy) in early 2003 (January or February - I can't remember exactly).

Three years later (Again - what a difference a few years makes!), I guess it's all been worth it. I would never presume to have moved the needle much on the industry's perception of Novell, but maybe I and the others at Novell who have struggled against the company's internal bureaucracy and inertia and external hecklers and naysayers have actually made a difference.

At least, that's the thought with which I'll console myself tonight. :-)

GPL the right model?

In this Network World piece, I argue that the GPL is a more business friendly license than the BSD, largely because the GPL more closely mimics the traditional intellectual property regime within which software companies have grown up. I had a minimal amount of space to make the argument, so I'm sure there are holes, but enjoy all the same.

I'm not dead yet!

Some of you noticed that my post about leaving Novell and joining a mobile software company (Volantis) was suspiciously yanked from the blog. Most of you know what happened by now. Without going into detail, I'm no longer joining Volantis. I had a disagreement with the CEO - my former boss and still a very good friend of mine - and in the fallout of that disagreement I realized that nothing short of me managing my own business would satisfy me.

I was very depressed about this at first, because I was greatly looking forward to working with the exceptional people managing and employed by the company. Mark Watson, the CTO, for example, is fantastically smart, fun to be with, and will make the company continue to succeed. He glowers a bit too much at times, but everyone has their faults. :-)

But I'd like to create a mobile open source company of my own. That may have to wait, as I'm exploring areas within Novell that I can own and manage but, barring that possibility, I guess I'm just going to have to create the first billion-dollar open source startup. Now if only I could figure out how to do it with two matchsticks and a bottle of hairspray, MacGyver-style, as Joe Kraus recently opined is really all one needs these days.

;-)

Will a G5 inoculate him?

So, while standing in line today at CompUSA (trying to get a copy of Super Smash Bros. Melee for my son's birthday - no luck), I happened to look across the store and see Ralph Yarro (of Canopy/SCO infamy. We made eye contact, recognized each other, and both looked away. There's no love lost between us.

However, before looking away I could help but notice that he had a Apple Power Mac G5 under his arm. (He had brought it in for repair, apparently, though I still firmly disbelieve Macs ever break. :-) My heart suddenly softened toward him. Sure, he's sued companies (IBM, Microsoft, CA, and likely others) to make his money instead of successfully investing it, and sure, the guy has done more to set back Linux (by a few seconds) than anyone else on the planet. But he did have a Mac.

Surely he can't be such a bad guy. Right?

Right?!?

;-)