Open Development: diversity matters

Kaj Arnö, of Mysql fame, has been commenting my now quite old Open Development rant, providing some insightful comments and interesting food for thought. Given where Kaj comes from, I’m not surprised he doesn’t quite like my arguments about community development needing neutrality to thrive, and he does have a few points when he provides an alternate and articulate proposal for “Participatory Open Source”.

I’m happy to see Kaj agrees that Open Source should be about community as well, and I applaud to what MySQL has been doing to get external developers involved. Still, I have to stomp my feet somewhat and insist on neutrality as a main component of Open Development (or whatchamacallit, for that matter). The keyword here is diversity, as I mentioned before. Sustainable Open Source, built upon the classical advantages of multiple eyeballs, absence of vendor lock-in, distributed innovation and cooperation among developers needs a community of technical interests, with participation being driven by diverse objectives who happen to share a common goal. A diverse community drives innovation, acts on technical merits rather than business objectives, and protects from orphaned/abandoned code or from whatever might happen to a single company backing a project. Diversity is good. Diversity matters.

A diverse environment is hard to build if neutrality isn’t part of the picture. Not impossible, but highly unlikely. For one, it’s hard for people to give free lunches away on a continuous basis: casual contribution are fine, and there is no problem in sharing a bugfix a developer cares about, but there is little evidence of communities with a thriving developer base from different backgrounds helping out actively the average commercial Open Source project on a structured basis (partner companies need not apply, thanks). Think of it as “no taxation without representation”: where is the motivation to contribute if there is no way to have a significant say in the project governance and roadmap? Why should I bother contributing ideas and code when there will be a business scrutiny getting precedence over technical merit? A neutral and meritocratic environment which encourages participation is the single most important feature individuals and companies should look for when deciding to contribute precious time and resources.

Kaj, if you need any further proof, look no further than your post when you question the notion of non-discriminatory access to the code:

The interesting middle ground is whether it’s OK to say “no” or “maybe later” to technically sound contributions that don’t fit with the business interest of the owner of the main code base. I would argue that such “discrimination” is OK, as the “Participatory Open Source” requirements would otherwise impose so severe purity limitations, as not to be interesting for companies with a commercial agenda, like MySQL AB.

As you can see, this is a chasm which is not easy to cross. MySQL AB, ultimately, wants the final word when it comes to a business agenda clashing with a technical proposition: as much as your position is understandable, I trust you acknowledge the little motivation third parties will have in contributing. Saying that contributions are welcome as long as they fit your business is OK from a corporate perspective, but let’s call a spade a spade and consider how your position from a community perspective is less than inviting.

You just can’t have your cake and eat it too: you’re not going to get any structured contribution without giving some control away, and you’ll have to be content with what you get from a community built upon a few that happen to share your business goals and a number of “useloper”, if you allow me to forge a new term that tries to describe users that might contribute something from time to time. If you happen to be as widespread as MySQL is, that’s no small stuff, yet you can count on a Simpson’s hand how many commercial Open Source companies are so lucky. The other ones either will have to deal with a few casual contribution here and there, and fully carry the burden of internal development. Or just drink the Open Development Kool-Aid, adjust a little their business model, and enjoy participation from a diverse, neutral, community which will provide the foundation to their business solution.

See you in Philly!

I just got home from the Open Source Think Tank, and it’s almost time to pack again and cross the Atlantic one more time. Next week I will be in Philadelphia, talking about Open Source in Corporate Environments at Emerging Technologies for the Enterprise.

As my schedule gets tighter, this time I will have no room to hang out and have a look around: I will be landing in Tuesday and fly back on Thursday evening, which will make jet lag interesting stuff to master. If you happen to be around the Philadelphia area, just wave and I’ll be glad to have a chat over a beer or so while I’m attending the conference.

Hopefully travel won’t be too bad: I decided this was a good occasion to jump the cliff and give the “all-business” airlines a chance, so I’ll be flying via Paris with L’Avion, spending 70% than the coach class rip-off for short-length trips and enjoy a better seat, possibly with some sleep included on my way back. Bonus treat: I will escape the Heathrow security jokes. As it seems I’ll be crossing the ocean a number of times in the future, I hope this proves to be a good compromise between sustainable prices and travel comfort: where do I sign to have those guys open routes to the West Coast?

The walled garden of Open Source

I gave a shiver when reading what Eben Moglen is going to be talking about at the upcoming OSBC:

Copyleft Business Models: Why It’s Good Not To Be Your Competitor’s Free Lunch

Abstract: Now that the GPL wars are over, and we have two good GPLs to choose from, it is time to re-ask some fundamental questions about business models and software licenses. In this talk, I explain why smaller software-focused businesses will soon be deserting Apache- and BSD-style permissive licenses for GPL[2 3] and their successors.

This is a perfect example of what’s wrong with Open Source nowadays. I keep hearing the argument that GPL is the best way to avoid your competitor to run away with your code, and I just can’t stop thinking how this is the correct answer to the wrong question. First of all, in Open Source (or should I say Open Development?) there shouldn’t be such thing as “your” code: commons-based peer production of software is all about sharing technology, and build value upon it.

Code is just part of the game: ultimately, what you definitely don’t want to happen is your competitor running away with your value, and there is much more to value then just code. Your competitor can run away with your code just fine, even in the GPL world, but this isn’t going to make a substantial harm to your business: Unbreakable Linux anyone?

At the end of the day, the GPL is a powerful tool for the new wave of Open Source to build walled gardens and keep business as usual in software-land, while giving a casual wink at Open Source as the only way that makes sense today to bring a product to the market. Software is developed behind firewalls, external contribution beyond casual bug fixes are not welcome, most business models revolve around upselling users to paying customers with little added value other than a different license. No economies of scale, no distributed innovation, no collaboration to share a common technical platform and work on adding business value to each proposition. Can you say “BORING”?

The not-so-subtle hint the players of Commercial Open Source have in common is that users should be paying because it takes time and effort to build a software, and this needs to be compensated. And hey, they’re right: if the burden of building software, paying engineers, doing QA and all that falls entirely on their shoulder, it’s just fair they seek compensation even before looking for profit. Then again, this is another correct answer to the wrong question. A successful Open Source ecosystem works when different entities (companies, research, users, individuals) collaborate on building a technology, the commercial part being adding business value to the platform.

Isn’t it exactly what RedHat does, by the way? They stand on a huge community cluster, they contribute quite significantly to the ecosystem, and they sell the packaging and oversight as the added business value. Do they use GPL? Sure, but not as a tool to protect themselves against the competition. Their ultimate protection is brand, quality and ultimately guess what? Value. Even Oracle had to learn that the hard way.

This is why permissive licenses are not going away, no matter what the FSF may think. As long as we are going to see companies who grok Open Source as the best way to collaborate on technology and compete on business value, we will need permissive licenses to allow ecosystems of participation. Software vendors in disguise are of course free to (ab)use the FSF and the GPL to keep on pretending they are Open Source players while burning money in expensive internal engineers and QA teams: just remember that, in the end, the community wins. No matter what.

It’s a commoditized world

There has been quite a buzz in the blogosphere about the excellent slides from Brent Williams, and there is most definitely a reason for it: the document is very well written, entertaining and informative. A must read, and I found myself resonating a lot with what has been said at EclipseCon just a few days ago.

I have to note I don’t fully agree with the author’s views when it comes to debunking the so-called myth of software being a commoditized market. Brent runs a very convincing reasoning along the lines of commoditization being applicable just to industry sectors where it’s impossible to distinguish between products of different producers: think frozen orange juice, crude oil or steel. There is no question software is different given that, as the author notes, people don’t exchange database “just because”. Too bad this is dangerously close to a straw man argument: no one would argue that software is a commodity, yet it would be very hard to deny that we are assisting to a gradual shift (some would call it a landslide) in the software business, which is best described as a (asymptotic, if you like being meticulous) commoditization process.

Excuse me while I hop on the stage and start taking my turn at debunking myths: there is no such thing as a purely commoditized market in the business world. No matter how you look at it, every market has a number of factors to consider before even getting to the price of goods: a steel plant will undergo tons of negotiations before committing to a supplier who is able to guarantee a constant flux of raw material, doing its best to apply price pressure, get better conditions, arrange logistics and possibly integrate the vendor’s IT system into its ERP software. It’s highly unlikely that a manufacturer will change its supplier “just because”: resistance to change and, ultimately, some kind of lock-in, is omnipresent in day to day business. Sticking to economics books, there is little or no way we can find a single commoditized market out there.

If we take the notion of commoditized market with a grain of salt, we have to add a few “almost” and “mostly” here and there. Well, the software market then is clearly moving towards commoditization, with Open Source playing a significant role in the process. Reduction of acquisition cost, flexibility in licenses, much better predictability of associated costs, leaner adoption process: these are just some of the factors that make the software market much more lively than it used to be. It is easier for a camel to go through the eye of a needle, than for an enterprise to change a software before full amortization/depreciation kicks in: as Open Source moves a large chunk of costs from CapEx to OpEx (license vs. subscription), resistance to change is much lower as there is little to depreciate. As software gets more fungible, using Open Standards, interchangeability is much less of an issue. As we move to a service/on demand/hosted model, matching functional needs is nearly all you need to make the switch to a different supplier.

Pardon me for calling that commoditization, while we consider moving our email infrastructure from a Zimbra install which is only two months old to the new and shiny Google Enterprise offering, effectively switching from a fully depreciated infrastructure (yes, we have been cheap bastards using the community edition) to a non-depreciable and fully interchangeable SaaS solution. To us, e-mail has been commoditized as much as it can possibly be: maybe you should give Open Source a try, after all.

Thoughts from the tank

Now that the jet lag has been somewhat mitigated by a lot of sleep, and the impressive pile of backlog has reduced to something more manageable, it’s time to write a recap from what’s been going on in Napa Valley at the Open Source Think Tank.

The whole experience has definitely been a blast and a revelation: sitting in the same room with 90 or so among the most prominent people in Open Source industry was both terribly exciting and extremely informative. Kudos go to the folks at Olliance for making this happen, and for their spotless organization. For a change, the program wasn’t about sessions (apart from a few presentations): we have been split in brainstorming groups, focusing on the prominent issues and promises of Open Source, trying to find answers and proposition to guide the commercial Open Source community in the years to come.

What I’m taking home is a bag of mixed feelings. For one Napa has been the ultimate proof, if ever needed, that Silicon Valley is the most powerful IT Old Boys’ Club ever, as being there makes very easy to understand how powerful local buzz and networking can be: the impression I’ve got is that there’s little need to bother about business plans and forecasts, when all you need is a bottle of Napa’s Chardonnay and your tennis buddy, which by the way happens to be a VC, to help your sipping on a sunny terrace while you talk him over investing a few millions in the next Open Source gizmo. OK, I’m taking this to the extreme, but if you are on the hunt for funding, do consider moving over there, as it can really make the difference.

This said, commercial Open Source is definitely thriving, as witnessed by the notable number of startups that were mingling at the event. As Fabrizio notes, it’s good to see that the hype is almost over, the playing field is level and the Open Source companies are now working on a peer level with traditional software vendors. The event has been an incredible confidence builder for me: the Napa audience (almost fully involved in the business of Open Source products), pretty much agreed in considering system integrators (I happen to know a pretty good one) as a core component of the value chain, bridging product companies to enterprise customers and building powerful agglomerates of technology from the vast choice of Open Source and Open Source related stuff out there. This is even more true for good ‘ole Europe, a market that US companies find quite hard to grasp, where system integrators have been traditionally playing the role of technology advisors and implementors. More good news for us, then: having our business plan validated and applauded by everyone in the crowd has been definitely good and highly motivating.

To be completely honest, though, I see some dark clouds over the horizon. Maybe it’s just me being over-cautious, but whenever I see such a massive amount of VC funding, I start to smell bubble. As much as statistics may be a powerful weapon of mass deception, we all know that the failure rate of VC-funded companies is very high. I’m wondering what will the scenario be in three years from now, when one company out of ten will be successful, four will have survived by selling out their assets and soul, and five will just have vanished.

What’s really sad is considering how very few will be attending their funeral service: despite sharing the same room with the big Open Source guys for nearly three days, there was little or no consideration for the community aspect of Open Source. And I mean none. Zilch. Nada. Of course communities have been mentioned here and there, but mostly as user gatherings, cheering crowds of enthusiastic fans eager to help fixing bugs and making suggestions. This is quite far from my notion of community as the diverse environment fostering innovation and providing sustainability to a software project. Call me narrow-minded, but I can’t quite see how “Commercial Open Source” can be considered sustainable at large when in a lot of cases we’re just talking about a bunch of coders behind a firewall, throwing some stuff to their users here and there and taking their products to the market mostly with a “bait and hook” approach (or worse). As much as the CIOs at the event told us how they don’t really care about Open Source in itself, as long as a solution solves their problem at the right price, I still contend there is much more to OSS in business then a cool distribution model with a somewhat enhanced role for the customer. We’ll get there, eventually, even if that would need some reality checks once a few of those companies start to vanish as the ruthless Darwinian laws start to kick in.

Oh well, enough ranting: time will tell, and of course I hope to be proven wrong. I’ll leave my fellow readers with one of the most shared visions from the gathering: everyone is expecting that in a few years from now Open Source and commercial software are going to meet somewhere in the middle, raising some sort of hybrid distribution and business model as software gets fully commoditized and SaaS takes the world by storm. I can clearly see some of that happening already, and it will be most interesting taking a look at the software market in 10 years from now. For the time being, I’m happy to keep on digesting the wealth of information I’ve got, eager to get the most from the impressive deck of business cards I managed to snatch, and impatient to find out whether having a similar event on a European level (as Matthew gently hinted) might be as fantastic as it looks on paper. And yes, Tuscany would be great!

Stupid airlines…

Taking the advice of a few seasoned traveler buddies, I thought I’d ditch Alitalia in favor of British Airways for my US bound long haul flights. So far, I’m less than impressed as it clearly seems BA is doing his best to make a traveler’s life miserable: the new policies (looks like the UK has a passion for stupid procedures) explicitely state that it’s impossible now to change your seat once the boarding pass has been issued. Too bad their online check-in is utterly buggy: when confronted with the worst seat ever being assigned to yours truly from the system, while being unable to change it apparently due to their poor browser support, I tried to cancel the whole procedure and try to sort me out at the airport, yet it seems that “cancel” to BA means “confirm the stupid seat you gave me and issue the boarding pass with no way to roll back. No, really, hell will freeze over before you can change your seat”. Net result: I’m stuck to a cramped seat for the next 14 hours or so.

To add insult to injury, it looks like BA now requires to show at the airport the credit card being used to buy your ticket, apparently because of fraud prevention. I’m lucky enough to be the boss and have the corporate credit card with me, yet this piece of plastic is what we normally use to fly our guys around the globe, and I don’t think we’re the only ones doing so: as much as it’s likely I’ll try to avoid BA in the future if this becomes a widely accepted tribulation, I’m starting to wonder what will the next step be to make our life harder.

Update: I was lucky enough to find a disgruntled yet helpful BA employee at Heathrow, who told me how they were pissed off with the new “no seat change” policy and how their group managed to find a way around the “inhibitor” that the company set up on their systems. He asked me to keep on complaining to see the policy reverted, and who am I not to oblige, given I should have a decent seat now?

$ find /var/log/stuff/ -newer ~/blog/$LATEST -exec cat {} >> ~/blog/new \;

Blog hiatus, again. It’s been more than a month since my last post, and despite my frivolous attempt at writing about what I’ve been up to (Japan, work stuff and lots of travel), I haven’t managed to jot anything worth publishing.

Thing is, life is really crazy over here. Since I got back from Japan, I’ve been literally swamped in dealing with the backlog first, and doing lots of going around later on. I think I know every single employee of Schiphol airport by now, not to mention I could probably drive a cab through Amsterdam or Norwich without getting lost. During the first two months I flew some 45 hours already (that’s nearly 45 minutes per day in average, if you like statistics), slept in 9 different hotels for a total of 23 nights out (which amounts to two days out of five, including weekends) and had 33 business dinners (basically killing my diet more than every other day). I don’t even dare to count emails, meetings and phone calls, but I guess you get the idea.

And it’s way from being over yet: Wednesday I’m taking off to beautiful Napa Valley for the long anticipated Open Source Think Tank. I’ll be flying back on March 11, with a day to kill in SFO, so if you feel like meeting up for brunch or so just drop me a line. I will barely have the time to unpack before flying out to Germany, then it will be just about time for my usual weekly trips to the Netherlands and the UK. The final week of March I will be speaking at Emerging Technologies for the Enterprise in Philadelphia, most likely with a full week-end to spend in the NYC/Jersey/Philly areas so, again, I would most appreciate some company for a beer and a chat. I hope to enjoy Easter home, then I’ll be traveling lightly to Rome, Norwich and Amsterdam before heading to the Netherlands again and speak at ApacheCon Europe early May before leaving, most likely directly from AMS, to SFO where I’ll be speaking at JavaOne (woohoo!).

Incidentally, this means I’ll be crossing the Atlantic no less than three times in three months, earning a hefty amount of FF miles in return but getting nearly killed by jet lag and traveling altogether. I’m considering planning to turn my MBP into the ultimate entertainment center to escape the flying hell: given I’m now the proud owner of an Empower MagSafe cable, I expect the thingie to make the difference, assuming I can find a suitable seat, so I’m now chasing movies and games to make my traveling life less miserable.

Are you still wondering why blogging is so light over here?

$ find /var/log/stuff/ -newer ~/blog/$LATEST -exec cat {}>>~/blog/new ;

Blog hiatus, again. It’s been more than a month since my last post, and despite my frivolous attempt at writing about what I’ve been up to (Japan, work stuff and lots of travel), I haven’t managed to jot anything worth publishing.

Thing is, life is really crazy over here. Since I got back from Japan, I’ve been literally swamped in dealing with the backlog first, and doing lots of going around later on. I think I know every single employee of Schiphol airport by now, not to mention I could probably drive a cab through Amsterdam or Norwich without getting lost. During the first two months I flew some 45 hours already (that’s nearly 45 minutes per day in average, if you like statistics), slept in 9 different hotels for a total of 23 nights out (which amounts to two days out of five, including weekends) and had 33 business dinners (basically killing my diet more than every other day). I don’t even dare to count emails, meetings and phone calls, but I guess you get the idea.

And it’s way from being over yet: Wednesday I’m taking off to beautiful Napa Valley for the long anticipated Open Source Think Tank. I’ll be flying back on March 11, with a day to kill in SFO, so if you feel like meeting up for brunch or so just drop me a line. I will barely have the time to unpack before flying out to Germany, then it will be just about time for my usual weekly trips to the Netherlands and the UK. The final week of March I will be speaking at Emerging Technologies for the Enterprise in Philadelphia, most likely with a full week-end to spend in the NYC/Jersey/Philly areas so, again, I would most appreciate some company for a beer and a chat. I hope to enjoy Easter home, then I’ll be traveling lightly to Rome, Norwich and Amsterdam before heading to the Netherlands again and speak at ApacheCon Europe early May before leaving, most likely directly from AMS, to SFO where I’ll be speaking at JavaOne (woohoo!).

Incidentally, this means I’ll be crossing the Atlantic no less than three times in three months, earning a hefty amount of FF miles in return but getting nearly killed by jet lag and traveling altogether. I’m considering planning to turn my MBP into the ultimate entertainment center to escape the flying hell: given I’m now the proud owner of an Empower MagSafe cable, I expect the thingie to make the difference, assuming I can find a suitable seat, so I’m now chasing movies and games to make my traveling life less miserable.

Are you still wondering why blogging is so light over here?