Proprietary software serves the vendor, not the customer

I just posted an entry over on my InfoWorld blog about open source's successes, and how they "prove" the rightness of the model. I say something in that blog, however, that I want to emphasize here:

Let us be clear. Whatever the merits of proprietary software, they are purely vendor-favoring. There is no customer reason to make software proprietary. None. There is no customer benefit that attaches to proprietary software. There is only a vendor's ability to temporarily monopolize a piece of software and thereby profit from it.

For some odd reason, the customer is lost in all the debate about open source versus proprietary software. There is no reason for closed source beyond vendor insecurity. That is, vendors close their source code because they worry that customers won't buy it otherwise.

Fine. Nothing wrong with making a profit from one's creations.

But the point is that proprietary software is not the only way, nor even necessarily the best way, to make money from software. And if CIOs show themselves happy to pay for open source software, what, then, is the purpose in keeping proprietary software around?

The mindshift in the industry is shifting. When I started at Lineo, CIOs saw very little reason to pay for open source software. That was in 2000. Seven years later, however, the market is very different. My company will quadruple sales in 2007 (we're already well on our way) over excellent results in 2006. That's partly because we're doing a decent job, but it's also because it costs much less to sell the market on open source software.

Two years from now, I suspect there will be almost no reason to lock up software (and customers) with proprietary software. The winners will be those that make the shift now, rather than waiting for the market to discipline them.

Fleury's finest moment?

I have to admit, I think Marc Fleury is awesome. He's a difficult personality at times, but a straight shooter and did a fantastic job building one of the world's most successful open source companies.

But Marc, this presentation [Streaming Video] from JavaPolis is a bit over the top. Not the message, which I agree with (and think Marc is refreshing with how he calls "foul" on some of the practices open source companies use). But the medium....

Dressing up like Flavor Flav? My CEO has a super-embarrassing ring tone ("Wipe Out!" at 3 billion decibels), but I'd take that any day to "the Flav" showing up for work. :-)

All that said, try to get beyond Marc's stagemanship in this presentation to listen to what he's saying. He knows what he's talking about. And he made enough money putting his talk into practice to not need to care one bit whether I like how he dresses.

Off-topic: The new Shins album rocks

Listen to "Sleeping Lessons" and you'll be hooked. Awesome tunes.

But why are they coming to Salt Lake City the night I need to be 400 miles away at my daughter's soccer tournament? Grrr....

You can listen to the entire album here.

More on the Pentaho move

If you took a look at my InfoWorld blog, you noticed that Pentaho has moved to a 100% open source model. Good for them. I think they'll find that abandoning the proprietary nipple helps a company grow up in 21st Century software capitalism. Pentaho just made it that much easier to club competitors, help customers, and spread its software.

Open source, however, is not a panacea, as Andrew Aitken points out on one of my recent posts:

Matt, I wholeheartedly agree with your current assessment of the revenue ramp of open source companies but have a bit of conern for the future of that prediction. As someone who is also very familiar with the revenue numbers of many open source companies and having completed over 100 strategy engagements in the open source space to date we can attest to the fact that it is definitely becoming easier to achieve a rapid revenue growth rate. Our concern comes from the fact that with all the numerous proprietary companies open sourcing something, no matter how small or insignificant, in order to call themselves an open source company, (almost one per day for the two months we tracked this in mid-2006) the market is going to become confused and could potentially feel misled. Resulting in skepticism of a different kind, not about open source technologies per se, but about the business entities themselves, potentially slowing commercial open source sales. Additionally, there will be the failure of a number of open source companies over the next couple of years, the inevitable result of over funding of particular industry segment or technology.

I couldn't agree more. There are scads of companies who open source their software to try to remain relevant in a market that long ago abandoned them. Open source will not save them. Garbage in, garbage out.

But Pentaho is at the top of its game, so this move should really help them. I applaud the strategic shift and wish them well.

Of kings and competitors

Reading William James still, and came across this great quotation from Ralph Waldo Emerson, one of my favorite philosophers/essayists:

When you strike at a king, you must kill him.

In other words, in some acts, "halfway" is worse than not at all.

And such it is with open source, on many levels. Halfway won't do if you're an open source vendor. Straddling the fence leaves you...in a not-so-pleasant position. And open source, itself, cannot afford to go halfway in competing with the Proprietary Bloc. The proprietary world has everything to lose from open source. Keep this in mind when you see the big vendors chumming up to open source. It is always at least a little disingenuous, because there are billions of dollars at stake.

No halfway. Kill the king. Burn the boats.

Why I never worked for a law firm

I did interview with a few firms while I was at law school, but was fortuitously saved from legal practice by Lineo (which is where I had my open source baptism by fire). Reading this article [Subscription req'd] in the WSJ makes me realize just how lucky I am not to have ended up at a big firm.

(I was tempted to go to Simpson Thacher, I will admit - a $10,000/stipend to live in London with a $160K starting salary sounded great. But then I did the math, and with the amount of hours I'd have to work, it came out to be below minimum wage. ;-)

Anyway, there are many problems at big law firms, but I'm not sure having partners treat people with basic human dignity will be the panacea:

Faced with a surge in turnover of its associates, the prestigious law firm Sullivan & Cromwell LLP has been putting on a charm offensive to hold onto junior lawyers.

The crash course in etiquette went into high gear at a partners meeting last February. To deal with low associate morale and high attrition, a confidential slide presentation reviewed by The Wall Street Journal urged partners to say things like "thank you" and "good work" to associates they supervise....

The presentation showed that the New York firm, now with about 625 lawyers, lost 31% of its associates in 2004 and 30% in 2005. The average associate attrition rate for law firms of about that size or bigger for 2004 was 21%, up from 16% in 2002, according to a study by the National Association for Law Placement.

Another slide showed that in American Lawyer magazine associate-satisfaction surveys, Sullivan compared unfavorably with peers like Davis Polk & Wardwell, Cleary Gottleib Steen & Hamilton LLP and Debevoise & Plimpton LLP. In 2005, Sullivan ranked 155 out of 160 law firms in a survey of midlevel associates.

There is real, useful work to be done by JDs, but the best work (and most fulfilling) won't be within the four opulent walls for the white-shoe law firms. It will be at open source software companies.

The freedom to act

I was reading this morning, and came across a passage that really struck me: we are free to act, not to be acted upon. This jibes perfectly with what I've been reading in William James. Freedom is in action, not in waiting for others to act upon us.

In the open source world, many companies treat open source as a ball-and-chain to be endured (for the marketing benefits of saying they're "open source"), rather than as a tool that gives them essential liberties. Open source is a tool, even a weapon, for MySQL, Red Hat, and others. It's a cross to be borne by others. This is reflected in their licensing: open source, but not too much.

(Incidentally, this same phenomenon shows up in all aspects of software. There are true Web 2.0 companies, and then there are those that try to hide their true business models in AJAX. There are true SaaS companies, and then there are those who are hosted and little more. There is a fundamental difference between acting a part and being that part.)

It is my strong conviction that the best open source companies are those that embrace open source the most fully. That act with open source, rather than allow themselves to be acted upon by open source. I'm not alone in this belief: the companies that attract the most developer community interest are those that thrive as communities, just as much as they thrive as companies.

Perhaps there's a correlation?

Does the Linux Foundation matter?

I wasn't quite emphatic enough about the Linux Foundation over on my InfoWorld blog. Let's face it: this is positive for both FSG and OSDL, but will it matter for the industry? I'm doubtful. I'm not a big fan of committees deciding anything of real value, because they move too slowly. Red Hat certainly isn't going to slow down long enough for the LF to be relevant to it.

So, the big question is whether the Linux Foundation can lead Linux, rather than play catch-up/"please slow down!" I'm not holding my breath, but think Jim is a good person to give it a try.

Weigh in on attribution (OSI)

I know I promised no more on the topic, but the OSI has just posted something highly useful to closing the debate, one way or the other. You can comment at license-discuss@opensource.org (or you can follow the debate by subscribing here.)

From Michael Tiemann on the license-discuss list:

Last December the SocialText folks made the decision to submit their licnese for review, which we appreciate. The license-discuss list has been full of discussion, but that discussion has not yet been reduced to a coherent argument either for or against. Rather, we have heard many many opinions as to what one person does or doesn't like about the SocialText license, attribution in general, or positions that others have advanced for or against either topic.

As I see it right now, either the OSI Board can attempt to pick up all these disparate pieces, try to place them together (where they fit) separate them (where they conflict) and then judge whether one position or another is more compelling in light of the OSD. That's an easy task when all the pieces fit together and all land strongly to one side. In the case of the SocialText license, I feel there's significant risk that if we take on the responsibility of making the arguments, we may create a bias that is not faithful to the real arguments you want to make. Therefore, we'd like to invite those who think we should not approve the SocialText license to work out a common position on *why* we should not approve it, which could inform how SocialText could remedy your concerns. And we'd like to invite those who think we should approve it (or should approve it with some minor change) to work out a common position on why we *should* approve it. If one or both sides are willing to do this, I think that the Board's decision process will appear much more transparent.

One way or another, the Board owes SocialText and the open source community a ruling, and we'd like to do as good a job as we can. If the challenge to organize is taken up, we'll set a timetable based on input from the position leaders. If no organization effort is apparent, the board will take it upon itself to make the decision by the end of next month (which gives time for one meeting to discuss and one meeting to decide). Thanks!

M

Help build a consensus one way or another, and let's get this resolved.

My very last attribution post

Ca suffit! I have hit my limit on the attribution debate. I'm a little weary of people haranguing companies for using attribution licenses, primarily because it is the OSI, not the companies, that should be under pressure right now. A generic attribution license has been submitted to the OSI. It's been sitting there for months now, going through endless theorizing on license-discuss, with no concrete decisions coming out of the OSI. (Note: I'm an OSI board member and also work for a company that uses attribution, as such, I have recused myself from voting on the issue within OSI.)

As I've written, I'm no friend of attribution. I think it's a weak method to achieve aims that the GPL is better constructed to do. There isn't a day that goes by that I don't advise a company (including my own) to adopt the GPL.

But this doesn't mean that attribution isn't open source. There are plenty of open source licenses that I don't like, don't use, and would be happy to see go away. But they're still open source licenses.

Here's why I'm not buying the anti-attribution movement's open source argument:

• Attribution does not burden distribution (any more than many other open source licenses, including the GPL). Some of the criticism has claimed that attribution prevents redistribution. This is quantitatively not true. There are tens of thousands (perhaps more) of sites that use attribution-"burdened" software, many of which include composites of different open source projects. Whatever theory might predict, reality definitively shows that attribution has not unduly burdened redistribution.
  
  
• Attribution will consume the UI as different attribution licenses are combined. I grok this point (though it's more theory than reality). It is, however, not really any different from the existing fact that some open source licenses simply aren't compatible with others. (Having said that, I'll state for the record that I don't believe that Apache licensing truly is incompatible with the GPL, but there is a strong perception that this is the case, which prevents a lot of interweaving of the two licenses that might otherwise happen.) Perhaps you won't want to integrate Zimbra with SugarCRM with SocialText...and? This is not an open source problem. It's a practical problem.
  
  
• Attribution doesn't work in headless environments. This is actually a good point (i.e., can you use the software if you're using it headless, or in an environment where you don't have a UI, as it were?), and goes to #10 of the Open Source Definition. Perhaps it would require modification of these MPL+attribution licenses, but I'm not sure the baby should go out with the bath water.
  
  
• Attribution is consistent with OSD #4 (Integrity of an author's work), droit morale, etc.. While some in the development community don't buy this argument, I can tell you from actually sitting in on these discussions that it is a big deal for a company to see its software poorly supported. The OSI provides this rationale for OSD #4:

  Encouraging lots of improvement is a good thing, but users have a right to know who is responsible for the software they are using. Authors and maintainers have reciprocal right to know what they're being asked to support and protect their reputations.
  
  Accordingly, an open-source license must guarantee that source be readily available, but may require that it be distributed as pristine base sources plus patches. In this way, "unofficial" changes can be made available but readily distinguished from the base source.

Rather than spend our time counting how many attribution licenses would fit on the head of a pin, we should be focusing on the practical effects of attribution. I can understand that some developers prefer not to use software that requires attribution. I totally get it. But this is no different from me not wanting to use BSD-style licenses, Apache not wanting to use GPL-style licenses, etc. This isn't about preferences. It's about what is open source, and what is not.

I am a pragmatist, in the William James slant on the word. I believe in what works. If it works, it is right. Attribution has blessed us with mountains of exceptional code. Code from which tens of thousands of groups are deriving daily, tangible value (with similar numbers of developers involved in modifying, forking, and promoting). I know, for example, 6,000 active users of attribution-"burdened" software at Alfresco alone. Add that to the Zimbra, SugarCRM, SocialText, MuleSource, Intalio, etc. etc. users and you quickly get to a massive population of users and developers that don't seem to share the same distaste for attribution.

I guess this makes me an unwilling apologist for open source attribution. It's not something I prefer to do with my time - I don't like attribution, but also don't think that personal preferences should get in the way of pragmatism. Pragmatism in the Jamesian sense.

/Matt talking about attribution (hopefully forever)

On the average intelligence of CEOs

I loved this blurb from BusinessWeek, quoting Carl Icahn:

"I have my anti-Darwinian metaphor: The CEO is the fraternity brother type who is great to have a drink with. He's a survivor and maybe not all that smart, but he works his way up the ladder in the corporation. And if you're a survivor you never have someone beneath you who's smarter than you. So you eventually work your way to CEO. You have someone a little dumber than you underneath, and eventually we'll have morons running everything...which we're getting closer to."

I knew I had something to look forward to. :-)

And so the community rips itself apart...again

You know, the one thing worse than proprietary software is watching the open source community rip itself apart. The link I'm referencing started with a fantastic case study from a Wal-Mart developer about how they're (happily) using MuleSource, the leading open source ESB.

Unfortunately, a select few have taken it upon themselves to cry foul on MuleSource, arguing that it's not actually an open source company because...it uses the Mozilla Public License? No, actually, because it uses the MPL plus attribution (the MPL allows addenda, and attribution has been used by a number of companies as one such addendum).

I am personally not a big fan of attribution, though I also feel that they are well within the bounds of the Open Source Definition. Opponents claim that they burden redistribution. Apparently these same opponents have never read the GPL, which is massively burdensome (in good ways, in my opinion, but not in many others') on redistribution.

So much so, in fact, that very few companies will touch the GPL. Talk with most Fortune 500 companies and they'll pull out the garlic and crucifix if they hear your product is GPL'd. (Though this inexplicably doesn't apply to Linux, which they happily use. Reason, thy name is not enterprise IT.)

Rather than sniping at those evil capitalists (which pay the wages of 99.999999999999999% of the world's open source developers), inflicting attribution on the world (which attribution, btw, was approved in various guises by the OSI several times in the past), why don't you try using the software? Or getting back to your day jobs?

Just what is open source

So, I think I've found some things to do with this blog. For instance, I'd like to experiment here with some ideas, and have it be more of a discussion forum. InfoWorld can be more "authoritative." AC/OS can be more speculative.

On that note, I've been torn lately on what constitutes an open source company. Nearly everyone claims the honor of being open source, yet there are very few that really make open source the crux of their business. It's a tool, but not the model.

Some models rely on a percentage of the being open source, with a significant percentage (30% or more) remaining proprietary. Is that open source? I'm of two minds on it. I mean, I wouldn't call myself a Christian just because I obeyed 70% of the commandments (or tried). Nor would I consider myself a Jew if I obeyed seven of the 10 commandments (but decided killing someone now and again to be fair game), or if I obeyed all of them...70% of the time.

Or what about those that fund open source projects, but have the overwhelming majority of their software proprietary? Waxing religious again, am I a Christian just because I pay tithing? Isn't adherence to the commandments more critical than cash when it comes to defining one's religion? Or would I be a Democrat just because I funded Planned Parenthood, even though I was a registered Republican?

The problem with taking these sorts of definitional hardlines is that they tend to exclude...everyone. In someone's mind, every open source company out there is not open source enough. I view Red Hat and MySQL as the quintessential open source companies (bordering on "free source" companies, in the FSF-type perspective), but there are many who feel differently.

We do need guidelines for using the term "open source" so that it doesn't become hopelessly diluted, and I still think the OSI is the best group to manage the Open Source Definition. But you know the best policing mechanism I've found? The community. It's what helped to drive Alfresco away from a hybrid model toward the MPL. I think it's probably the most potent weapon we have against fakes.

I may go live on Open Sources for a while...

When I started blogging for InfoWorld a year or so ago, I opted to keep AC/OS active because I wasn't sure how permanent the move was going to be. I wanted to have a consistent blog-space that would remain my own, regardless of where I blogged temporarily. Now that InfoWorld has consistently treated me well (and is even taking over management of OSBC, bless their hearts), I've decided to stop co-locating my blog.

That's right, if you want to follow my open source commentary (the exact same kind of thing you're used to reading here), you're going to need to do it from

http://weblog.infoworld.com/openresource/rss.xml.

That's the RSS address. The normal URL is http://weblog.infoworld.com/openresource or just click here.

Again, same content. Just a different place. And, frankly, I often add detail to the InfoWorld blog.

Rather than leave AC/OS vacant, however, I think I'm going to use it for my snarkier commentary that InfoWorld prefers that I not write. I think I may also offer more personal commentary (though I hate blogs that are more travelogues and "aren't I cool?" vanity presses than they are informational) on things outside open source. Like why this is the year you should start listening to The Essex Green, Clap Your Hands Say Yeah, and The Shins. Or maybe the occasional Arsenal commentary.

Or maybe it will be so dull here that I'll just leave it hollow until InfoWorld boots me out.

Anyway, please feel free to keep this feed active if you want to see if my experiment in being interesting beyond open source is worthwhile. It will probably fail (I'm married with four kids, after all - I became boring long ago, but very happily so). So, if you want the best source for open source news/commentary on the 'Net, please be sure to subscribe to Open Sources here.

C++ from the founder's mouth

Krishna Rangarajan (AllianceBernstein) and I had breakfast yesterday in New York. I love talking with Krishna because he has a fascinating vantage point on the IT industry, and always has interesting things to say. In the course of our conversation he pointed me to this fascinating article in MIT's Technology Review. It's an interview with Bjarne Stroustrup, founder of the C++ programming language. The interview follows the development methodology behind C++, and talks through its pros and cons.

My favorite line from the interview was Stroustrup's response to complaints that C++ is hard to use (despite being widely used)

There are just two kinds of languages: the ones everybody complains about and the ones nobody uses.

Exactly.

Free markets and prosperity

The WSJ has an interesting OpEd piece today on the rising tide of economic freedom across the globe, and how that contributes to rising levels of prosperity.

Here's bad news for those who oppose global free trade: Not only did the world-wide trend toward greater economic liberty hold steady over the past year, but the incomes of poor individuals across the globe are rising as result. The world isn't only growing richer. The gap between the per-capita income of have-not populations and that of the developed world is narrowing....

[E]conomically free countries enjoy significantly greater prosperity than those burdened by heavy government intervention. The per capita GDP of the top quintile of countries, ranked according to economic freedom, is now almost $28,000 while the bottom quintile is less than $5,000. The associated higher GDP rates that come with economic freedom "seem to create a virtuous cycle, triggering further improvements in economic freedom. Our 13 years of Index data strongly suggest that countries that increase their levels of freedom experience faster growth rates," says the report.

While almost a platitude here in the US, history does seem to bear out the idea that greater economic freedom trends toward greater economic prosperity.

You can guess, therefore, one reason I am so emphatically pro-open source. Open source is all about free markets. Yes, we have a free market in the software world, but I'd argue that opening up the code gives customers greater degrees of freedom in two ways: lower costs (so that they can spend their IT dollars more broadly) and no (or lower) lock-in (so that they can spend their IT dollars more discriminately). All of this makes for a true free market in software. "Free as in free market," as I heard r0ml present back in his 2003 eGovOS presentation.

The open source market opportunity(ies)

More content will be created in the next two years than in the entire previous history of humankind, and over 93% of it will be electronic, according to a report Accenture released in 2006. Think about that.

Now compare it to the growth in the ECM market:



Looks good, right? Wrong. When you have massive amounts of new content, you'd think the market would follow it with dollars. But we're not seeing that. In fact, the only vendor (besides Alfresco, I mean :-) to be making incrementally large piles of cash from this exploding market need is Microsoft (with Sharepoint).

Why?

I think the problem largely stems from the way the big ECM vendors have attacked the problem: in big, monolithic ways. Because they're structured for the enterprise sale, they sell almost exclusively to large enterprises, who long ago bought all the ECM software that they'd need for some time. "Back office ECM" is how I like to term it. It's something that enterprises use, not individuals. As such, the big, proprietary vendors only managed to achieve 5% penetration in the vast majority of their target customers.

Why?

Well, also because their products are complex, "heavy," and cumbersome to deploy and to use. They're also obnoxiously expensive. Between hard to use and hard to afford, it's not difficult to see why ECM has seen its growth limp along behind the market need for content management.

Microsoft changed the game by focusing on "front office ECM" needs. Not even true ECM, really, but more like a collaboration portal. Sharepoint (Services, a stripped down version) comes free on every copy of Windows Server 2003, allowing it to grow like a weed (with not-so-weedy financial results - $0 to $500 million in just three years).

It's plagued by lock-in, sure, but given the choice between Microsoft's lock-in (with the "offsetting" benefits that it's easier to use, integrates well with much of their existing "productivity infrastructure" (read: Microsoft Office), and is much cheaper than FileNet, Documentum, etc.) and Documentum's/IBM's/etc. (with the compounding demerits that they are expensive, hard to use, and complex to administer), you can see why enterprises are voting with their feet (and wallets).

Microsoft has long done this to markets: improve ease of use and lower costs. At times it goes about this in inappropriate (read: illegal) ways, but there is nothing illegal about making the customer experience better by making it easier and cheaper. Such behavior tends to hurt incumbents but benefit the larger market (by making a much bigger market/universe of users, as Larry Augustin illustrated in his Open Source Business Conference 2004 presentation):



The one thing missing in all this, of course, is the freedom to leave Microsoft when you're tired of buying all of the other pieces to make your initial purchase work. (Sharepoint, for example, requires you to use IE, buy SQL Server, Windows Server, and, depending on which version you run, you may need to buy Office 2007 while you're at it.)

Enter open source.

Microsoft is doing a great job of locking customers into a cheaper, easier-to-use, collaborative way to manage content. Think of how much better that picture would be if you made it even cheaper, even easier-to-use, even easier to deploy, even easier to collaborate, and removed the lock-in (not only at the repository level but also at the component level, meaning that you could use the best of breed CMS with your choice of operating system (including Windows, if you want), your choice of database (including SQL Server, if you want), your choice of application server, your choice of authentication mechanism, your choice of productivity suite, etc.

Your choice, in short.

It would be better. And, arguably, it would grow that ECM market you see above much more effectively than Documentum, Vignette, and their ilk ever dreamed of doing.

This is what I do every day at Alfresco. But the real point of this blog entry is to say, "So can you." It's not about competing with Microsoft or this or that company. It's about delivering a superior customer experience for woefully underserved markets, and nearly every enterprise software market is underserved because of the proprietary incumbent products' cost, complexity, and distribution models.

This is open source's opportunity. It might as well be yours, too. Just be sure to burn the boats before you get started. Your customers will be happier, as will you. Your competitors? Not so much.

Who contributes the most open source code?

I've long said that Sun contributes the most open source code...and gets the least credit for it. Well, now the European Commission is supporting the claim with facts, not anecdotes. As Matthew Aslett details on his blog, Sun tops the corporate open source list with Red Hat coming in third to IBM's second. Sun, by the way, wins by a landslide. The others on the list may surprise you....

Don't try to upsell your community (Fabrizio)

I had dinner with Fabrizio, a good friend and CEO of Funambol, the leading mobile open source company. He was in Salt Lake to ski and was kind enough to call me so that we could hang out.

Fabrizio said some things about open source that rang true with me, which I had not considered before. I'll list two principles he mentioned, and will discuss each in turn:

1. Don't upsell your community, and
   
   
2. Sell open source to those who don't like/trust open source.

At first glance, Fabrizio's principles fly in the face of most open source businesses out there. But when you scratch the surface of his thinking you see that it actually undergirds the most successful open source businesses. Let me explain.

Fabrizio's first principle - "Don't upsell your community" - basically means that there are multiple markets for open source, and the developers who download your product will not be an important source of revenues. They yield many other benefits - product extensions, product feedback, bug fixes, etc. - but don't expect them to fund your development.

You can see companies making mistakes in this regard all the time. They set up models that are designed to goad "free-riding" developers (or, "the community") to pay. Hence, the now ubiquitous "Community" vs. "Professional" or "Enterprise" versions of products, with the hope that Community users will become Professional or Enterprise buyers.

They won't. Give it up.

This is not to say that product segmentation is wrong, but it means that we need to be clear about how and why we segment an open source project.

In Fabrizio's case, he recognized that consumers aren't going to pay him money, and the enterprise market might not be fertile ground yet either. So he's focused on mobile operators, and finding a great market there (especially in developing geographies). He gets a huge amount of value from his user/development community (fast approaching 1 million downloads), and doesn't confuse his goals with the operators with those of his user community.

This has meant two things for Funambol:

1. They don't write code that enterprises may want. They write code that their mobile operators will want, and let the community extend Funambol to meet enterprise-y needs (like Exchange connectors - operators and consumers don't care about Exchange, so why should Funambol spend its development dollars there?).
   
   
2. Funambol's licensing is designed to maximize code reuse and utility for its developers (i.e., GPL), while not harming the mobile operators (who simply want to buy their way out an open source license, anyway).

This brings me to Fabrizio's second principle - "Sell open source to those who don't like/trust open source." These are my words, not Fabrizio's, but they're my best rendering of what we discussed. Fabrizio keeps finding that the customers willing to pay the most money tend to be those that appreciate the benefits of open source, but don't want open source, itself. In his mobile world, this means that they like the fact that the code is effectively in permanent escrow with a massive user base and so it doesn't have to bet on Funambol, per se, but rather on its community. They also don't have to fret about being locked into their vendor.

So, they like the benefits of open source. But their lawyers don't want to be bothered with thinking through the implications of possibly (though this possibility is remote) having to share their code, or figuring out support, etc. So, they buy a commercial license to Fabrizio's code to get the benefits of open source without the so-called risks of open source. They want commercial open source. They want a company behind the community, but they definitely want the community, too.

I think Fabrizio has it right. In my content management/collaboration world, my primary customers are enterprises. They want open source, but they don't want to lose the commercial relationship with a vendor. Enter Alfresco.

But they also don't want a lopsided entity that is mostly commercial, and very little community, open source. I personally feel that the GPL (and other free software licenses) is the best way to ensure maximum community input while retaining the ability to give enterprises a way out. It won't always be like this - I assume in a few years enterprises won't be so keen to buy their way out of the obligations of open source (because they'll recognize that the obligations lead to greater and greater benefits), but while the world is as it is...it's a great model.

Look at what MySQL has done with its Enterprise offering, coupled with their Network. MySQL took nothing away from its community, but added to what companies wanted (better support, more QA, etc.). Its development/user community gets the freedom of GPL (v2) so that they don't really have to care that there is a company behind the project. Enterprise customers, for their parts, get the commercial license so that they don't really have to care that there is a community behind the product.

Everyone wins.

This is as close to "The Right Model" as we currently have in open source, in my opinion. It's community-maximizing without being overly reliant on support dollars. (If your business depends on selling support exclusively, you're going to find that you may successfully pull in five-figure deals, but you'll always strain and struggle to get the six- or seven-figure deals. Those require something beyond vanilla support - not proprietary software, but rather "proprietary" service (meaning the code is free, the services around it are not).

Thoughts?

Delta Airlines entertains with Red Hat

Not sure if this is public (quick scan with Google says "no") but on my flight to NYC tonight Delta's entertainment system was rebooted and I noticed that it runs Red Hat. It's one of the old Song planes, so maybe Delta's standard system runs something else (though Delta is reportedly expanding its use of the Song in-flight entertainment system, so....)

Anyway, given how hard it is to get success stories/press releases/etc. out of enterprises, I figured I'd announce this one for Delta. Delta runs Linux. I knew there was a reason I liked Delta so much. :-)

Roy Russo enters the Attribution Debate

Roy Russo, a prominent JBoss developer (Portal, in particular) has waded into the open source attribution debate with an interesting take on the badgeware phenomenon.

Roy points out some reasons companies use badgeware (Cash and credit), though he doesn't give companies much credit for doing it for other, less devious motives (since it's provided without support, for instance, in some cases attribution is a good way to ensure users know that they could get support from the developers of the product, if they got the code directly from these developers). But this isn't Roy's concern. He's not worried about whether MPL+attribution is open source or not. For him, it is.

No, his real concern seems to be that a) the OSI (full disclosure: I'm on the OSI board) is not taking an active position on the issue and that and b) people need to stop whining about open source purity and instead focus on the tangible benefits of having more code (attributable or no). He writes:

Working for JBoss, I've had to listen to the pinheads at BEA launch public crybaby fits over "JBoss is not as open source as BEA, as open source as IBM, as open source as Apache, as open source as your mom". From them, I expect the ridiculous and incoherent ramblings of a lunatic... from the OSI I would simply expect a coherent and reasonable bulletpoint in their definition of open source.

Badgeware is certainly not going away, and my fear is that if the OSI chooses to ignore it, or worse, refuses to approve the licenses, they will find themselves as irrelevant as the UN in short time.

Putting my OSI hat on for a moment, I can tell you, Roy, that the OSI spends a lot of time working through the issue. I've actually been pleasantly surprised with the rigor that fellow board members like Michael Tiemann and Russ Nelson have put into the matter.

But it's a thorny issue, and one that OSI prefers to move slowly on. That said, I agree with you: the OSI needs to act or risk irrelevance. I, personally, can see both sides of the attribution debate, and I think the market is largely deciding the issue for us. The market seems to be quite happy with SugarCRM, Zimbra, Alfresco, SocialText, and other companies that use attribution. Perhaps the OSI's vote won't matter much in the end?

Open source: Europe vs. USA

From Rishab's excellent report came a few graphics that I thought worth calling out separately from the report itself. It's interesting to see how Europe's adoption and usage of open source compares with the US'. Here are a few slides (from IDC and Optaros - its report is here [Registration req'd]) that depict the differences.

First, here are the open source applications most prevalently used in Europe:



Compared to the US, first large ($1B+) enterprises:



And then mid-sized companies ($50M - $1B):



It's fascinating to see how much more open source database uptake there is in Europe, compared to the US (whether in large or mid-sized enterprises). Most of the other categories compare well, but database usage in Europe is impressive. MySQL has been knocking the ball out of the park with Web 2.0 companies (Google, Yahoo!, etc.), so maybe it's primarily the most innovative users of IT in the US that favor open source databases, while traditional enterprises favor geriatric software? :-)

On the topic of which industries are the widest adopters of open source, the tables show many commonalities between Europe...



...and the US:



Shockingly, the US financial sector seems to be WAY behind other industries in using open source software. This is shocking because a) I'm personally aware of significant open source use (at every level, from OS to application) within every brand-name financial services company you can think of, and b) if the report is accurate it means that this use pales in comparison to Europe.

Regardless, the numbers look very good for open source. As for the Proprietary Bloc...? Not so good.

The economic impact of open source (EU study)

Rishab Aiyer Ghosh and UNU-Merit put together this insightful report [PDF] on the impact of FLOSS (an unfortunate acronym for free/libre open source software) on the European economy. It was financed by the European Commission’s Directorate General for Enterprise and Industry.

Why does it matter? The paper starts by suggesting that the information economy accounts for ~10% of the GDP of most developed nations, and more than 50% of their economic growth. Think about that. If even remotely true, it means that the kind of IT we espouse matters a great deal. It's fundamental to our future economic growth. Do we want closed or open economies? I think the answer to that is obvious.

Some of the more interesting findings, some of which don't ring true for me:

• FLOSS market share [is] higher in Europe than in the US for operating systems and PCs, followed by Asia. These market shares have seen considerable growth in the past five years.

  ASAY: I can buy this for operating systems and PCs, but my own experience shows that commercial adoption of open source is generally much higher in the US than in Europe, and far behind in Asia-Pacific.

• FLOSS market penetration is also high – a large share of private and public organisations report some use of FLOSS in most application domains. In the public sector, Europe has particularly high penetration, perhaps soon to be overtaken by Asia and Latin America. In the private sector, FLOSS adoption is driven by medium- and large-sized firms.
  
  
• Almost two-thirds of FLOSS software is still written by individuals; firms contribute about 15% and other institutions another 20%.

  ASAY: This is probably true if the report is measuring open source, generally. But for widely used open source (Linux, Apache, etc.), the software is developed by individuals who largely work for corporations that specifically pay them to write open source software. I doubt the report makes this distinction, largely because it would be difficult to trace who works for whom, given that most developers participate as individuals, though their ability to do so regularly results from a corporation's paycheck.

  
• Europe is the leading region in terms of globally collaborating FLOSS software developers, and leads in terms of global project leaders, followed closely by North America (interestingly, more in the East Coast than the West). Asia and Latin America [Page 9]

  This is probably true, which is why I always find it so odd that European-based open source companies relocate to the US...so that they can be next to VCs(???). VCs are nice and necessary, but they're not buying product. Be where your customers are. Your customers are likely often in the US, but almost never in the Bay Area.

All of which is interesting and perhaps a bit self-congratulatory. But where the report gets really interesting is when it attempts to measure the direct impact of open source on the European economy:

• The existing base of quality FLOSS applications with reasonable quality control and distribution would cost firms almost Euro 12 billion to reproduce internally. This code base has been doubling every 18-24 months over the past eight years, and this growth is projected to continue for several more years.

  If you're a proprietary software manager reading those numbers, you've got to be scared witless. Roughly $12 billion in software out there, doubling roughly every two years. Why would anyone pay you for your software when they will increasingly be able to get equivalent or better quality software for free, and spend their money on service, when they can (and not when the vendor dictates), not code? (Hint: They won't.)

• This existing base of FLOSS software represents a lower bound of about 131 000 real person-years of effort that has been devoted exclusively by programmers.
  
  
• Firms have invested an estimated Euro 1.2 billion in developing FLOSS software that is made freely available. Such firms represent in total at least 565 000 jobs and Euro 263 billion in annual revenue. Contributing firms are from several non-IT (but often ICT intensive) sectors, and tend to have much higher revenues than non-contributing firms.
  
  
• Defined broadly, FLOSS-related services could reach a 32% share of all IT services by 2010, and the FLOSS-related share of the economy could reach 4% of European GDP by 2010. FLOSS directly supports the 29% share of software that is developed in-house in the EU (43% in the U.S.), and provides the natural model for software development for the secondary software sector.

And here's the kicker:

• Proprietary packaged software firms account for well below 10% of employment of software developers in the U.S., and “IT user” firms account for over 70% of software developers employed with a similar salary (and thus skill) level. This suggests a relatively low potential for cannibalisation of proprietary software jobs by FLOSS, and suggests a relatively high potential for software developer jobs to become increasingly FLOSS-related. FLOSS and proprietary software show a ratio of 30:70 (overlapping) in recent job postings indicating significant demand for FLOSS-related skills.

In other words, you have nothing to lose but your chains. It may well be that open source won't lobotomize the Proprietary Bloc so much as make it irrelevant and insignificant over time. It's happening already.

What about innovation? The report addresses this, too:

• FLOSS potentially saves industry over 36% in software R&D investment that can result in increased profits or be more usefully spent in further innovation.

  ASAY: Importantly, these savings apply to everyone, not merely open source companies/developers. Open source isn't biased in distributing its benefits.

  
  
  
• ...• Increased FLOSS use may provide a way for Europe to compensate for a low GDP share of ICT investment relative to the US. A growth and innovation simulation model shows that increasing the FLOSS share of software investment from 20% to 40% would lead to a 0.1% increase in annual EU GDP growth excluding benefits within the ICT industry itself – i.e. over Euro 10 billion annually.

If the investment in open source makes sense for Europe, it also makes sense for the US (to maintain its lead) and for the rest of the world, to close the gap. More open source = more innovation = more jobs = more money for everyone.

So what does this mean for Europe? The report challenges us with the following scenarios:

Europe faces three scenarios: CLOSED, where existing business models are entrenched through legal and technical regulation, favouring a passive consumer model over new businesses supporting active participation in an information society of “prosumers”;
GENERIC, where current mixed policies lead to a gradual growth of FLOSS while many of the opportunities it presents are missed; VOLUNTARY, where policies and the market develop to recognise and utilise the potential of FLOSS and similar collaborative models of creativity to harness the full power of active citizens in the information society.

Amen.

I've hit the highlights from the executive summary, but the real meat of the report is in its detailed findings. I'd encourage you to take a look, and to de-propertize your software lives. You'll be better for it. I promise.

Burn the boats.

Your software sucks (here's why)

I do a weekly podcast with Phil Windley, Scott Lemon, and a guest. Two or three weeks ago we talked with David Platt, author of the book Why Software Sucks. David was engaging, funny, and insightful. It was one of my favorite podcasts yet.

As we talked, I realized just how estranged from "the real world" many developers/companies are (as Mike Pastore writes). Most of us that live in the tech space forget completely that most people don't care about tech, much less know how to use it. They want technology to be an occasional complement to their busy lives, not the substance of their lives. That sounds obvious, but it's shocking how often we forget this when we are in the process of writing software.

Here's a blurb from the Technometria site advertising the podcast:

What is the most important thing to the average computer user? They want their machine to "just work". Why does Google know how to correctly translate a United Parcel Service tracking number, while the actual UPS website requires multiple entries just to get to the point where the tracking number can be entered? Programmer David Platt is the author of "Why Software Sucks...and What You Can Do About It". He discusses his findings with Phil, Matt, and Scott.

Platt believes that much of the problem is related to poor design, with not enough consideration for the end user. For example, he considers open source to be software written for other programmers, of little interest to the typical computer user. He also believes that blaming a particular operating system does little to solve the problem. He talks about the number of programmers who drive cars with manual transmissions to better illustrate how different the programmer thinks compared to other people.

This one is worth a listen, and worth reading David's book.

The BBC on open source

Want to hear smart-sounding Brit Peter Day talk about open source? Look no further than here. The BBC did a highly interesting review of open source and its impact on the software world.

My favorite line?

"This may be more than just a modification to the way computing evolves....[T]o users grown up in the world of handing over large sums of money for a license to use brand-name software, the open source movement may take some getting to grips with."

Amen. Most aren't (coming to grips with it). They're just fighting it. But they'll lose.

Several company execs are profiled/interviewed:

Paul Sterne
chief financial officer of Open-Xchange and general manager Americas.

Peter Yared
chief executive, ActiveGrid

John Newton
chairman, Alfresco

John Powell
chief executive, Alfresco

Scott Dietzen
chief executive, Zimbra http://www.zimbra.com

Paul Radamacher
Google

Di-Ann Eisnor
chief executive, Platial

Paul Saffo
writer and technology forecaster

I highly recommend listening to it. Peter's lovely accent makes it sound even better than it is. ;-)

From engineer to attorney (Luis Villa)

Not sure how I missed the news, but Gnome developer and former Ximian guru, Luis Villa, has to law school (Columbia Law School, to be precise - I was sorely tempted to go there - such a great school).

I just wrote about why it's important for open source companies to have an attorney on staff. I can't think of a better qualified attorney than one that also groks open source at the code and community levels. Luis knocks both of those balls out of the park.

I understand that he has his first summer lined up, but if I were you (and "you" could include every software company on the planet, as well as law firms), I'd be trying to get him on your payroll sooner rather than later. Luis is a great guy (though our first interaction was him taking me to task over a blog entry - it wasn't a pleasant first meeting :-) and heavily involved in strong open source communities. I'd hire him in a second.

(No, he didn't pay me to write this. He doesn't even know I wrote it. Well, now he does.)

Buying open source in 2007? Think applications

LinuxWorld Magazine ran this article yesterday about open source's move up the stack. (Thanks, Russ, for pointing me to it.) Rather than wondering whether open source has arrived (it has), the article asks, "Where?":

“Open source has won the first battle: It is now listed among the default platform decisions,” says Dave Jenkins, CTO at online outdoor sporting goods retailer Backcountry.com in Park City, Utah. The next step, open source users agree, is moving up the stack and figuring out which open source tools are ready for enterprise deployments.

“Infrastructure open source products are essentially a no-brainer at this point, but the adoption of enterprise applications has been slow,” says Curtis Edge, CIO at The Christian Science Monitor, which revamped its Web sites with open source software last year.

So, where do you look? Up the stack.

Not everything is perfect in open source land, of course. You need to do a little homework to determine what's ready, and what's best waiting on. But LinuxWorld provides a rough guide here:

Where things stand

Open source is going mainstream, but not all open source tools are equal. Here’s a look at how enterprise-ready some free software products are:

• Most mature: development tools (Eclipse, Hibernate, Struts), server operating systems (Linux, FreeBSD)
  
• Maturing: application servers (JBoss, Geronimo), security software (Snort, Nessus)
  
• Growing: collaboration (Zope, Drupal), content management (Alfresco, OpenCms), directory services (OpenLDAP)
  
• Emerging: databases (MySQL, Ingres), enterprise applications (SugarCRM, Compiere), portals (Jetspeed, Zope), search engines (Apache Lucene, ht://Dig), virtualization software (Xen)
  
• Embryonic: integration services (openadaptor), enterprise service buses (Open ESB, Mule), process management applications (OpenFlow)

Let's put it this way: whatever you're looking for, open source has it. You just need to determine how far the niche you need to fill has come in the open source world. In many (and, increasingly, most) cases, the answer is "It's ready for prime time."

Or you could spend that $1M on proprietary software to adorn your shelf space. I mean, you do need something ugly to sit there.

EnterpriseDB racks up a great year

EnterpriseDB just completed a great year, its first year as a revenue-generating company. For those who don't know, EnterpriseDB extends the PostgreSQL open source database to basically make it drop-in compatible with Oracle databases. Not bad. EnterpriseDB isn't open source, itself, but is a nice complement to PostgreSQL.

Highlights of its year include:

• Booked millions of dollars in sales. (I know of at least one deal, and perhaps two, that were $1.5M+ in size. That's pretty hefty for any company, much less a startup vendor built on open source.)
  
  
• Raised a $20 million financing led by Fidelity Ventures. Previous investors Charles River Ventures and Valhalla Partners also participated in the round.
  
  
• Significant customer wins, including Sony Online Entertainment and Vonage,
  
  
• Nice industry awards including Best Database (LinuxWorld Expo), Red Herring 100 list, and the AlwaysOn’s AO100 list, as well as a finalist for both the CODiE and Jolt awards.
  
  
• Hired important members of the Postgres community, including Bruce Momjian and Dave Page, two of the seven-member PostgreSQL Core Team. EnterpriseDB also added PostgreSQL performance expert Simon Riggs and noted PostgreSQL author Korry Douglas.

This is all fantastic, Andy and the EnterpriseDB team. What will make it even more significant for 2007? Open sourcing your extensions. I think a dual-license model would suit you and your community, and not your competitors, very well. The kinds of customers you're signing aren't going to run an unsupported database. Just ask Oracle.

John Deere goes "open" source

I received this news today, and at first was really excited. John Deere dumps its proprietary CRM system for an open source system from Queplix. As I did a little research, however, I was dismayed to find that despite Queplix parading itself in open source marketing, there's very little that's open about its source.

There really is no Santa Claus. Or no Santa Source, as the case may be.

Don't get me wrong: Queplix may do many things well. For instance, Queplix sounds like a winner when it comes to data migration:

Queplix's CRM Resuscitation Program eliminates the need for traditional data migration -- shortening conversion time from what is typically many months to a matter of weeks. Customers can ultimately access the same data, the same interface, and the same business rules, using a web-based platform that is more robust, and can be more easily scaled to support emerging needs.

That's great, but it's not open source. There is a weak open source version that no company could effectively use, but if you want anything useful, it's proprietary and pricey ($150,000/CPU and up for <201 users). Yes, this is cheap compared to the big, proprietary vendors, but it's pricey compared to the competition.

Go to Queplix's products page. No mention of open source. You can find it in press releases and elsewhere on the site, but how odd that it's not on the main landing page for its products. Even stranger, go to its downloads page. No code. Just white papers and such. About the only "open"-sounding thing on its website is its OpenForum, which wasn't working when I clicked through to it.

Queplix may well be a good company, but it doesn't look like a good open source company to me. I couldn't even find it on Sourceforge. Nor could I find its license, the code, or anything standard with open source on its website.

Now, I'm not a hardliner on things like this - I understand that there are different variants of open source. And, full disclosure, I'm an advisor to SugarCRM (Sugar takes criticism for its both source approach, but I managed Alfresco's CRM needs on SugarCRM's open source product for many moons - it works exceptionally well and was more than enough for our needs). I just can't accept the open source name on a company that appears to be just as closed as its proprietary competitors.

Good product? Maybe. Looks like it is. Open source? Nope.

Against Leviathan

I'm actually just trying to find a reason to post this picture:



Yes, it's Arsenal, but it's even more significant than normal. This picture represents Julio "The Beast" Baptista scoring his fourth goal today against Liverpool. That's four in the same game. (Which is about how many points Ohio State managed to put on the board in last night's game. Ouch!) Arsenal's reserves managed to trounce Liverpool's first team in the Carling Cup quarterfinal. It took Arsenal's first team to trounce Liverpool 3-1 in the Third Round of the FA Cup on Saturday.

On a more symbolic note, the picture represents the rise of the open source community against the Proprietary Bloc. It's what a rising tide of individuals can do against "Leviathan." Leviathan is, of course, Thomas Hobbes' conception of centralized power in the hands of a dominant state. We find our way out of our "nasty, brutish, and short" existence as "state-less" individuals by ceding authority to a central government with the power to discipline and punish anti-social behavior. Great, right?

Well, as with most things, it's a question of degree. Centralized power is very important and useful, but only to a point. I like the USA's constitutional republican system. I don't think I'd like an autocracy.

In the software world, we have veered too far toward centralized control in the hands of very few vendors. Toward autocracy. In every product space, centralization/consolidation is accelerating or complete. It's happening in my world - content management - even as I type this. It's happening in yours as you read this. It means less choice for customers, worse software for vendors (because competition, not monopoly, drives good software), and fewer opportunities for new entrants.

The antidote is open source, and the community-driven approach it engenders. I'm not Pollyanna, arguing that open source does this perfectly. But open source offers an exceptional way for the individual to compete again against Leviathan. It's an exceptional way to get a company/project started, and to let the software "seep under the doors" of enterprises mired in proprietary technology.

Or as my 18-month old and 4-year old would sing (along withthe Wonder Pets):

Wonder Pets! Wonder Pets! We're on our way,
To help a baby [featured animal], and save the day.
(Ming-Ming:) We're not too big,
(Tuck:) And we're not too tough,
(All:) But when we work together we've got the right stuff!
Goooooooo Wonder Pets! Yaaaaaaayyyyyyy!

What's going to work? Teamwork!

Sorry, couldn't help myself on that one, either.

How to market well (Windows Vista and Amazon)

Here's a lesson in successful marketing, taken from the Microsoft play book. Look at the email I just received from Amazon:



Simple but effective. It doesn't matter that the software I bought through Amazon was for my Mac, not Windows. Very clever. Huge distribution through Amazon and an easy process for making the purchase.

Good work, Microsoft. Now if only the software worked as well as the marketing. :-)

Hire a JD

While I'm on the subject of hiring, whether open source or not (but especially if you are), you really need to hire a JD. A lawyer, that is.

Now, before you call me a killjoy (because, after all, who wants to hang out with lawyers? :-), let me explain.

First off, you need to understand that there is a huge difference in attorneys. Some exist to kill good ideas. Others, however, are fantastic assets to a business. They are "business lawyers" and in the open source world you absolutely need one of these. Because software - and, in particular, open source - is essentially an exercise in licensing, it's important that open source companies have an attorney on staff. Not necessarily a senior IP attorney, but someone that can grok a contract and write one.

Here's what you'll get:

1. The ability to use licensing as a tool/competitive weapon, rather than merely as the ugly stepchild to the "real business" of software. (Savvy open source vendors can quasi-control their competitors by the code they release if they use the right license.)
   
   
2. You'll save a boatload in legal fees. Just by having me on staff (and I am not a "senior IP attorney" (I worked full-time through most of law school, and was generally working during class, so I only trust myself so far...)), Alfresco saves $5,000+ in legal fees. Don't believe me? Ask any software company what they spend on outside legal counsel. I think you'll find that bringing a JD in-house is very, very cheap, whatever their salary.
   
   (This is not to say that you shouldn't retain outside counsel. You absolutely should. The best open source attorneys will help you think about legal issues in innovative, constructive ways. You just shouldn't spend your entire Series B round with them.)
   
   
3. One less person out there that will sue you. :-)

To succeed well in software, you need to be smart about licensing. Just ask Oracle, Microsoft, etc. These are companies that understand licensing cold. To compete with them, as well as to ensure your licensing appeals to and serves the interests of customers, you really need to get an open source savvy attorney on board. The good news is that the law schools are minting these at a nice clip. Save them from firm life. Hire a JD.

(Take my brother while you're at it. He's just finishing up at Stanford Law School and has been thoroughly corrupted with open source by his older brother. :-)

If I wanted a job...

...I'd be sure to go to work for an open source company. I get people asking me all the time for career advice (Not sure why they ask me - it's not like I have a "career." I think that's what older folks have... :-). My advice is always the same:

"Work for an open source company."

The reason is simple economics. The market will basically pay you what it thinks you're worth, and your worth goes up exponentially when you have open source expertise. Open source, according to Gartner and nearly every sane person on the planet, continues to be one of the top three trends in technology. Consequently, if you're an enterprise (i.e., IT person) or an ISV, you want open source people.

And thus, if you're a would-be employee, you want to be wanted. You want open source experience.

It's an unfortunate fact that every person I hire has their salary rate go up 50% or so just by walking in the door of a successful open source company. They join an Alfresco, SugarCRM, Red Hat, MySQL, JasperSoft, etc. and their worth immediately goes up in the market. After a few years of hard work and success with these (or other open source) companies, they can write their own ticket. And it's a hefty ticket. (I should know - we and others are paying a premium for open source experience.)

Do yourself a favor. Find an open source startup. Learn the 21st Century way of writing/selling/supporting/implementing software. It's the very best thing you could do for your career.

Microsoft Vista vs. Mac OS X (InformationWeek)

John Welch of InformationWeek has a great article today comparing Microsoft's Vista with Apple's Mac OS X. Given that I'm covering it here, you can already tell which OS he finds superior. :-) In his words...

While Vista is indeed a major update to Windows, there's a lot of it that is, quite frankly, just Microsoft making up for lost time. The last non-server release of Windows was in 2001 with Windows XP, with only a single major interim update in service pack 2. In the same time, Apple has been steadily releasing updates to Mac OS X on what was a yearly schedule, now around every 18 months....

Microsoft had two serious issues. First, they had to make this update of Windows revolutionary enough that it came close to justifying the delay. Second, they had to come up with something that would stand up well with its main competitor in the desktop OS market, Mac OS X. Have they succeeded at both? I'd argue that the former's almost a non-issue: Vista will sell well, because the world won't have a choice. As far as the latter, well, probably, but you'd be hard-pressed to say Vista's better than Mac OS X.

In a nutshell, Vista vs. Mac OS X is Revolution vs. Evolution. It's about a massive, long-delayed upgrade that has to account for almost 6 years of progress by its competitors, versus a well-executed strategy of regular updates. While updating an operating system is never something that can be called easy, Apple's strategy has been the better one for keeping their OS on top of things, something Microsoft has admitted to in a roundabout way.

This is critical, and often overlooked (outside the Apple camp, anyway). Microsoft's release late and not-so-often mentality means that its users are half a decade behind the Mac world (and Linux, in some areas). Microsoft, because of its massive installed base, arguably has a harder time moving to a "perpetual beta" release mentality. But others, like Google, have shown that this objection is more perceived than real.

Here are some of my favorite snippets from Welch's article:

Mac OS X...[is] the classic English butler. This OS is designed to make the times you have to interact with it as quick and efficient as possible. It expects that things will work correctly, and therefore sees no reason to bother you with correct operation confirmations....

Windows is...well, Windows is very eager to tell you what's going on. Constantly. Plug something in, and you get a message. Unplug something and you get a message. If you're on a network that's having problems staying up, you'll get tons of messages telling you this. It's rather like dealing with an overexcited Boy Scout...who has a lifetime supply of chocolate-covered espresso beans. This gets particularly bad when you factor in things like the user-level implementation of Microsoft's new security features.

An OS should just work, not tell you that it's working. An OS is largely infrastructure - Windows tries to be more than that. It wants to be furniture that walks around and chats with you, but really it just needs to be happy with being valuable furniture, and let its applications (and others') chat with the customer.

Welch also talks up the consistency of the OS X user interface:

This consistency that has been a centerpiece of the Mac OS is something that, even with Vista, Microsoft still can't manage to pull off. Although there are many different UI styles available in Mac OS X, even within those different styles, there is a consistency that Windows just can't seem to hit.

Even with Microsoft applications, there's a feeling that, by and large, the only UI guidelines that Windows applications adhere to is "what we feel like." (I know Microsoft has a lot of UI guideline information, but since no one seems to follow any of it, I'm not sure what the point of it is.)

Welch has other complaints on the Vista UI (it's harder to find some information you need, like network connection details, for example, and the name changes that seem to have been made only for the sake of showing that things have changed), but his complaints about the new security feature, User Account Control, seem more weighty. UAC doesn't offer real security enhancement, in his view, and (in his view) is...

...going to be called "User Annoyance Control." You get what is essentially an "Okay/Cancel" dialog that most users will hit "okay" for without thinking, you may or may not get useful information as to what is going on, and you get locked out of your system until you deal with this. I have a problem with seeing how annoying people is enhancing security. When I say "annoyance" I really mean "infuriate," because you get UAC dialogs all over the place, and you're never sure when or why you're going to get them.

That said, it's a tough problem to solve. How do you give users control over security settings that they won't necessarily understand? I'm not sure how Microsoft could have done it differently, but I do know that I don't have this same problem with my Mac. Not at all.

However, as Welch concedes, it doesn't really matter if Vista is better than OS X. Microsoft only really needs to worry about whether it's better than XP, so as to convince its user base to upgrade. Welch thinks it is (and I do, too, from what I've seen - that said, I hated XP, but liked Windows 2000 quite a bit). As for how it compares to OS X?

However, is it significantly, or even slightly better than Mac OS X? Maybe in a couple of low-level ways, like the randomizing memory address usage function, or being able to use USB memory sticks as additional RAM, but at the human level? Not even close.

I've yet to see anything in Vista that blows away the Mac OS, even a version of the Mac OS that's over a year old. Microsoft still can't manage to make something simple and easy to use. Vista reeks of committee and design by massive consensus, while OS X shines from an intense focus on doing things in a simple, clear fashion and design for the user, not the programmer.

Which, I think, is why I've managed to convert nearly 10 people to the Mac in the last two years alone. Now if the rest of you Mac people would just do your jobs, we'd have world domination within the next few years. :-)

Second Life goes open source

Calling the move "inevitable" (would that all software companies were this astute :-), Linden Labs has open sourced its client. Linden wrote on its blog:

At Linden, we have always been strong advocates of the use of open standards and the advantages of using open source products. Though Second Life makes abundant use of non-standard technologies, our basic UDP protocol message system for example, we rely on open standards and open source implementations when appropriate and available. Since many of the components that will make up this network are not yet done, we are not publishing long white papers or RFCs at this time — instead, we are giving everyone what we have along with a goal of producing those open standards with the input and assistance of the community that has brought Second Life to where it is now.

Releasing the source now is our next invitation to the world to help build this global space for communication, business, and entertainment. We are eager to work with the community and businesses to further our vision of our space.

Welcome to the community, Linden, whatever the life you happen to be living at present (first, second, third....). :-)

For Microsoft Entourage users...

There may not be many of you out there, but for those who (like me) use Entourage on the Mac, you'll want to read this. I was inputing events into my calendar this morning, and grew increasingly frustrated as I tried to enter appointments for March 31. Everything was an hour off. I'd enter it for 10:00 MST, and it would show as 11:00 MST. All of my time zones were in sync on the laptop. I couldn't figure it out.

I turned back to February, and events I added showed up just fine. I went back to March, and events that I had added (Arsenal games, mostly :-) showed up just fine. Well, some of them. I finally figured out that all appointments between March 11 and March 31 were off by an hour. Everything else before or after those dates was fine.

Here's why, and here's when you can expect to have it fixed (words of wisdom from the maestro, Paul Berkowitz). Sounds like it's a known issue and will be fixed soon. So, if you're a Entourage user, just sit tight. Too bad it's not open source so that someone could fix it for Microsoft....

Saturday fun

Ah, what a nice start to the day. i didn't get to ski today because I had to do pictures with my son's basketball team (which he missed because he was...skiing) - I'm the coach. But I had a great consolation prize: watching Arsenal thrash Liverpool in the Third Round of the FA Cup, 3-1.



Nice to have Henry back. Arsenal are beautiful.

Fixed the RSS feed

Sorry to those of you that have had trouble with the RSS feed for AC/OS lately. I upgraded templates in Blogger and it killed everything. I have it fixed now. You can re-subscribe (if the old address isn't working, for whatever reason) here: http://feeds.feedburner.com/Acos.

Mozilla's big numbers in perspective (Matthew Aslett)

Mozilla pulled in $52.9 million in 2005. Not bad for a company/group that actively tries not to make money most of the time. :-)

But that number, which most open source (and proprietary) companies would love to have on their books, becomes even more significant when you look at in in perspective, as Matthew Aslett has:

$52.9m might be small change compared to the likes of IBM, which had revenue of $91.1bn in its fiscal year ended December 31, 2005, but it is more than the likes of Pervasive Software, NetManage, FalconStor, WebSideStory, VA Software, and the good old SCO Group pulled in that year.

According to the figures revealed by Mozilla Corporation CEO Mitchell Baker, the combined expenses of the Mozilla Foundation and Mozilla Corp were just $8.2m in 2005, giving the organization a "profit margin" of 84.5%, beaten only by Google, Akamai Technologies and RealNetworks in 2005.

Meanwhile, with revenue up from just $5.8m in 2004, Mozilla’s revenue growth of 812.1% was unmatched. The nearest anyone got to that was Emblaze with 522.6% growth, followed by TomTom with 274.2% growth.

I just wish we could get an IPO out of this....

PHP vs Java: An interesting perspective from Dries

I'm on the phone with the founding Dries Buytaert, the founder and maintainer of Drupal, a hugely popular PHP-based open source project (web content management). Dries started the project as something to do in his spare time - just a hobby. He's done a lot of work with Java in the past, but elected to go PHP with his project. (He's also written on the PHP/Java question before.) I asked him, Why?

His answer was interesting, and instructive for anyone that wants to launch an open source project.

Basically, the decision came down to accessibility and cost. PHP is a much more accessible language. It's also one of the fastest growing in terms of popularity and usage. If you want a random developer to give you five minutes of her time, you need a language (among other things) that facilitates being able to do something productive with those five minutes.

On the cost side, he mentioned that hosting costs for a Java application are much higher than for a PHP application. To keep costs low, he chose PHP.

Pragmatism that has translated into a huge developer and user following. Sage advice.

Microsoft's new Office: Looks good, works well

Walt Mossberg covers the new Office 2007 in today's WSJ, and comes out very bullish on it. That's saying something as he's not a big Microsoft fan on most things.

I saw Office 2007 on a recent trip to the UK, and it's beautiful. Microsoft has done a great job with the UI upgrade, though (as Mossberg notes) it will likely cause some problems to users as they shift to the new interface. The Mac version won't be out for at least a year, but I'll be intrigued to see how it improves on the Windows version. (Little known fact: Office is better on the Mac than on Windows.)

The biggest headache with the upgrade will be the new document formats. Office 2007 automatically saves (you can change this, of course) into a new XML-based format. Older versions of Office won't be able to open them, which will undoubtedly cause consternation (and upgrades, which Microsoft will be happy about :-) as organizations try to share documents and find it difficult. (Yes, there's a free converter but no, most people won't know about it.) It's an opportunity for OpenOffice, yes, but keep in mind that one reason for the new file formats is to allow Microsoft to add/streamline functionality. It's not just about lock-in. There are real benefits to using the new format. (No, I don't know any, but I'm not an Office expert.)

The real opportunity for OpenOffice, I think, is to find ways to embrace and extend the new XML-based formats. Give its users more functionality than Microsoft Office, and enable it through the file formats. Open source is about more than commodified parity - it's about innovation. Microsoft has innovated with Office 2007 - what's the response?

MySQL license change: more of the same

Matthew Aslett is reporting on MySQL's mostly unnoticed licensing change. As he notes, it's not really a change, so much as putting a stake in the ground to keep MySQL on GPLv2 for the future. As Kaj (VP of Community for MySQL) notes on his blog:

MySQL has today refined its licensing scheme from “GPLv2 or later” to “GPLv2 only“, in order to make it an option, not an obligation for the company to move to GPLv3.

Specifically, this means that copyright notice in the MySQL source code files will change from referring to “either version 2 of the License, or (at your option) any later version” to “version 2” only, in the MySQL 5.0 and MySQL 5.1 code bases.

This is not a once-and-for-all decision, but rather gives MySQL breathing room to wait on a move to GPLv3 based on the results of the drafting process, and customer/developer uptake of the new license. I'm bullish on GPLv3 (though I haven't always been), and assume MySQL will eventually find it useful to move to GPLv3 in the future.

Where are they now? Open source executive moves, IV

Rob Beardan, gruff disciplinarian/operations boss at JBoss, has left JBoss/Red Hat to take the COO chair at OpenSpan. What does OpenSpan do? "Surface Integration." What is that?

OpenSpan, an Atlanta based software company, is the pioneer of Surface Integration™ - a new and innovative approach to desktop integration. OpenSpan's flagship product Composite Studio fully enables rapid integration and automation between applications, without programming, and at a fraction of the cost and time of any other solution. Composite Studio is a visual design environment that allows both IT and domain experts to optimize the existing functionality and any interactions between desktop applications, regardless of their ownership and origin (Windows, web, legacy/host, Java, etc). Solutions can be generated in minutes in order to improve productivity, streamline business processes, reduce costs, improve customer service, and enable tighter integration with partners and customers.

Interesting stuff, but the more so because Matrix is funding it and because Rob is there. Most people at JBoss would give heavy credit to Rob for taking a great project and turning it into a great company.

Good luck to Rob in this venture.

Oh, and Dawn Foster left Intel to run community relations at Compiere. Dawn is great and will do well at Compiere.

An enemy within?

I came across this entry on the Lobby4Linux blog. I don't know who writes th blog, or even if she/he is credible, but I found some of the comments interesting. Interesting and a bit worrisome. Worrisome because they paint Microsoft as it should be acting, and not as the backslapping Boy Scout that it has recently painted itself as.

When I talk with my Novell friends, I'm surprised by how naive they are about their recent pact with Microsoft. I talked with one relative newbie who was chattering about "big deals" being done with the help from Kevin Turner (COO) and other senior Microsoft executives. They apparently think that the loser in all this is Red Hat.

They're gullible in the extreme if they really think Microsoft has Novell's best interests at heart. It can't. That would be illegal. Microsoft's shareholders pay it to win, not hold hands with competitors.

Anyway, true or false, you can imagine that this is a plausible scene:

In the early summer of 2006, a meeting took place at Microsoft in Redmond. There were 9 people in attendance and three of them are of "elevated and notable repute". The meeting lasted for approximately 45 minutes and the sole focus of this meeting was Linux.

The Powerpoint presentations, charts and graphs all focused on the projected acceptance and use of Linux and how Microsoft could counter the growth. It is projected by the numbers presented in that meeting, that Linux will become a viable threat to Microsoft Windows by the year 2009 and that by 2012, Linux could gain as much as 60 percent worldwide market share. There is grave concern over India and China. There is also much wailing and nashing of teeth in Redmond over the GPL. It has been dissected to its finest portions in order to find the weakness within that will allow Microsoft to circumvent it. They can find no such weakness. This has amplified the wailing part of the afore-mentioned reaction. It has been decided that any control Microsoft is to have over Linux must be exerted from within Linux itself. Any attack from the outside will only result in failure.

If I worked at Microsoft, I'd be having this precise meeting. Microsoft can afford to take the long view on threats. (Check this out to see how long they've been talking about Linux, the Internet, and other threats.)

And yes, Linux is a threat to Microsoft. Big time. Which is why the only way to read the Microsoft-Novell pact is as a way to neutralize the Linux threat. (Doesn't make it feel very good if you're a Novell employee, but I'm sorry - business is business and Microsoft's business is not Linux. Just about the opposite, in fact.)

So what does Microsoft do? I remember talking with Jason Matusow at a Harvard Business School consortium on open source back in 2003. I suggested to him that the best way for Microsoft to beat open source was to attack its strength: community. Infiltrate and fragment key open source communities.

Jason, being the gentleman (and sophist :-) that he is, demurred, saying "We would never do that." But the Novell deal can be read as a very real step down that path (as can the deals with SugarCRM and others, but I'm more optimistic about those - I see those as ecosystem-building deals, rather than competitive undermining).

From Lobby4Linux:

There is to be a campaign early in 2007 to openly recruit key members of the Linux Development Community, but again; any specific targeted individuals was kept from me with a smile and a shrug. However, if you recall, shortly after Mark Russinovich discovered the Sony RootKit, he was scooped up by MS and his company System Internal/Winterals was quietly purchased by MS as well. So much for any outside innovation of MS software. MS mainly feared his possible future involvement in the FOSS community, a possibility they seemed to think he was considering. It was discussed that a "tactic" to lure FOSS developers would be to "empathize" with their plight, that no one really appreciated their labor...that they would be allowed a percentage of company time to develop for Open Source and their financial burdens would no longer get in the way of their FOSS work...just remember on which side of your bread is buttered."

Microsoft has already started its open source hiring drive: Daniel Robbins, Bill Hilf, others....It will be interesting to see if it manages to lure others - I'm sure it will. It's got a bottomless wallet and the promise of interesting work. And maybe it just wants to hire the enemy to understand the enemy. I can appreciate that. But let's be clear that Microsoft has a fiduciary duty to kill Linux and dominate with Windows/Office/etc.

All of which makes its efforts with Novell almost farcical. It tried with Red Hat, you know. Many moons, but Red Hat wouldn't play ball on the patents "covenant." (Nor should it have.) I have this from multiple sources, and hereby add Lobby4Linux:

Finally, and I must qualify this with a strong "might-or-might-not-be-correctly-interpreted" statement, the Novell/Microsoft Deal was almost the RedHat/Microsoft Deal. From what was discussed in this meeting, it was alluded to that there had been a pitch made earlier that paralleled the Novell Treason. While RedHat was not specifically mentioned, New York was. A mysterious meeting between Steve Ballmer and Matt Szulik in New York had been expected to yield the same results now obvious from the Novell/MS deal. Microsoft went into New York confident of it. One member present at the meeting was ten points past pissed when this was being discussed. Now I'm not saying there were any chairs thrown, or even if the angry person has a history of throwing chairs...the statement made by said individual is easily remembered and thusly quoted.

"I'll have that son of a bitch eating out of dumpsters in less than two years."

Mr. Ballmer, I somehow doubt this, just as your claim to bury Google has been...buried. I have no doubt that you will try, but in Red Hat you've stumbled upon a competitor just as determined as you, and just as stubbornly devoted to the customer, with a tool (open source) that you don't know how to beat. I think you've got a serious game on your hands, and a game where Red Hat holds the winning ace. It's called open source, and the customer is voting with its wallet.

I think you're going to lose.

What would Novell buy?

A conversation with a friend today made me wonder, "What will Novell buy?" Word on the street is that the company has $200M or so burning a hole in its pocket, with Altiris and XenSource as potential acquisition candidates. If there's any fire behind the smoke, both deals could make sense, albeit for very different reasons.

Altiris would be accretive; XenSource would not. At least, not in the short term. Altiris had a small profit in 2005, and will improve on that in 2006. It's pricier ($750M market cap), but is local to Novell's Utah office (makes geographic sense) but has nice overlap with Novell's ZENworks product line. Actually, it probably has far too much overlap, making it a rough fit, at least on the systems management side.

Except for the fact that Altiris has an interesting virtualization story, which is also the reason for liking XenSource. Here's what Altiris has to say about its virtualization solution:

Altiris® Software Virtualization Solution™ (SVS™) software is a revolutionary approach to software management. By placing applications and data into managed units called Virtual Software Packages, SVS allows you to instantly activate, deactivate or reset applications and to completely avoid conflicts between applications, without altering the base Windows installation.

So, Novell could get a profitable company and roll up more of the systems management market with a nice virtualization bonus. That deal makes sense, though it's pricey.

XenSource makes sense from a different perspective. If Novell were looking for a good way to dump a barrel full of flies into Red Hat's ointment (which has made virtualization a pillar of its strategy), snapping up XenSource might well be it. XenSource wouldn't be cheap - as Dave reported, XenSource recently raised money at a $100M+ pre-money valuation. Not bad for a company with piddling revenues. (Still, I really like the new management team - Frank Artale is a rock star. I'm assuming they can turn "piddling" into "buckets-full.")

Of course, Red Hat (and others) can still benefit from Xen even if Novell owns it, but not quite to the same extent. As I've written before, "source of code" matters as much or more as "source code" in open source. Buying XenSource, where everyone cares about Xen, is very different from buying Ximian, when few people care about the Linux desktop (not yet, anyway). Novell didn't do well - financially - by buying Ximian (though I believe they got great DNA through the deal), but XenSource is a different story. XenSource could be a huge differentiator for Novell in the Linux space.

So, will Novell buy one or both or neither? Not both, but I'm guessing Ron wants to get one. In terms of the biggest risk and potentially biggest reward, XenSource would be it. It will be interesting to see if the rumors bear fruit. (I've been wrong before about Novell's moves, though that JBoss time was through no fault of my own - Novell's offer was pretty tepid compared with the others' offers.)

Oracle starts talking again

It's about time. Oracle has finally opened up a little and Mike Olson of Sleepycat fame will now be blogging. I have a huge amount of respect for Mike and think he's a great addition to the Oracle team.

Hopefully, MIke will follow in the footsteps of others at big companies (Like Jason Matusow at Microsoft, Simon Phipps at Sun, and Jeff Nolan, formerly of SAP) who manage to say interesting things despite their PR departments. Mike will have an uphill battle here - nothing against Oracle, but the bigger the company the less interesting the things people are allowed to say there (generally speaking).

Welcome back, Mike. Now you will have a public forum in which to berate me. :-)

Web 2.0 - Hot and not

Paul Kedrosky has an Alexa-driven list of the top trafficked Web 2.0 websites (as well as the biggest losers). What's interesting to me is that I've only heard of 2-4 of the 20 or so companies mentioned. Guess I'm old school, Web 0.5.

Investing with O'Reilly

My good friend and sometimes ski/mountain biking partner, Bryce Roberts, joined the Technometria podcast last week to talk about the venture capital business and how O'Reilly Alpha Tech Ventures looks at technology. It's particularly interesting given how involved Tim is with the vetting process, and to hear about OATV's fundraising process. Have a listen.