Winning in a good way

Jos wrote a series of posts about Being Free, and I feel compelled to contest some of his conclusions, or at least show a different perspective on some of them.

First, let's make it clear, I want the world to use free software. I'm not here the technical thrill alone, though I'm a true nerd at heart. I believe in FOSS and its importance to users all over. I also believe that how we reach our goal is as important as finally reaching it, if not more.

Jos mentions some examples, like Microsoft and Novell, and Canonical and Ubuntu One. You cannot understand the contempt experienced by the community about the Novell-MS deal, unless you see it in a broader light.

The Novell deal

Microsoft had been threatening Linux in a direct manner or by proxy (hello SCO!) for a number of years. The Novell deal was the first chance that the Microsoft PR machine had to say: "See? Linux vendors are aware of our patents and they pay us for it". Second point that I really don't like, the deal only protected Novell's customers, not the community at large. Why, you say, would Novell protect someone who is not a customer? Because it's a community, damn!! We care about each other.

So, you say, this is a company and companies don't do that? Then look at the Red Hat-Firestar deal, which did exactly that. It took a patent, settled and it was done in a way which covered the community, not only Red Hat's customers. Sorry, Novell. The bar has been raised, and I'll never be your customer if you act like that. Remember we are a community next time. Remember you profit with code developed by someone else. Those people deserve your protection, not a new threat by the Microsoft PR based on your petty deal. So, Novell was looking for money, and money was all it got. Goodwill was not part of the deal, on neither side.

And it allowed Microsoft to go after other vendors, including TomTom, which was sued over a patent allegedly infringed by Linux. Remember that next time you talk about Microsoft being good to open source. This was last year, the "new open-source friendly Microsoft", not ancient history. And remember that patents can be used against developers and users alike, like SCO did.

Ubuntu's developments

Next one in line is Canonical's Ubuntu One. Great service, and a good example of Canonical's MO for development. They have to be kicked on the butt to release things in the open. Remember Launchpad, Rosetta, et al? Why? Don't they trust us? I'wouldn't have cared if they developed Ubuntu One as open source, and the service was charged. After all, maintaining all those servers and the bandwidth costs money. But what's the point of keeping the server source closed? It doesn't matter if eventually the code is released. The modus operandi in this and other cases where the source code was eventually released makes me think that Canonical doesn't really like to be open. They do not develop major projects in the open, not at first at least. And it strikes me as odd, because to me it looks that they feel the closed method is better, they just can't do it right now, but they will when they have a chance.

To me, the open bazaar methodology is technically superior in many senses. Not taking advantage of that, you lose many of the advantages you get by being part of this community. So, are they here because they believe in us, or only because a businessman thought it was an untapped niche? I'm sure it's the former, and I really hope they won't let me down. What would happen if the community won't insist them to release code? They just won't? Why? Why not start open from the begining? Or release it the minute it's working? We have waited years for them to release the code for some projects. They did, but only after a few kicks.

Why do we have to be the watchdogs? Why don't they do it on their own, like others do? I think opportunities were lost in this case. Look at OwnCloud. What if, instead of developing yet another piece of code, we would have had Ubuntu One's server to develop on top of it? Just add a KDE interface on it, start it on your own PC and off you go. That didn't happen, and from looking at GSOC's ideas, there is a lot of interest in OwnCloud. There goes a bunch of developers that Canonical won't use. All of that, because it uses closed development methodologies.

Canonical's marketing work has been amazing at getting people to know about Ubuntu/Linux. So, it would be bad to all of us if that name had a staint on it.

These are just some of the examples mentioned by Jos. I wanted to go over them to show a different perspective. Here are some others of my own.

Microsoft's interoperability efforts

Some people are touting Microsoft's interoperability efforts as the birth of a new Microsoft. Nice image, but a little naive. It should be reminded that Microsoft was dragged to the interoperability table, kicking and screaming, by the EU. And the EU didn't do that by itself. It was led by the antitrust litigation, where the Samba team and those loonies of the FSFE played a main role (A really big thank you for that). Come back to me when that company makes a true interoperability move, not forced by a judicial stick. In the meantime, just take the released documentation and try to port as much as possible out of their lock-in. The same goes for other half-hearted efforts by other companies. Grab your stuff and run like hell!

Pragmatism meets real life

Remember the BitKeeper events? That was a moment for pragmatics to just settle for a license and move on. Instead, a tough call was made, and Git was born. Allowing the Linux development process to be hampered by a license would've made an irreparable damage to it, limiting the number of contributors to the Linux kernel.

There are two other cases I want to mention. Canonical's use of vBulletin, and on the opposite side, KDE using phpBB. Canonical just said it uses vBulletin because it serves its purpose, and that's it. KDE negotiated with phpBB to release its code, and keep building on it helped by the community. If an NGO can do it, how come a company the size of Canonical, with a community the size of Ubuntu, can't do even close? Is it lack of power, or a lack of desire to do it, or just lack of interest?

I mention these cases, because as a former Windows and Ubuntu user, I know about these cases, but there are plenty of others.

We also have a dear case, with the license of Qt, finally solved by the KDE Free Qt Foundation, and the release of Qt under the LGPL.

Conquer the world, in our own good way

So, I want FOSS to conquer the world, just not in any way. We've made it this far because of who we are and how do things. Let's improve on that and make things even better. I see many people saying: "Ok, you've made it this far. Now start acting like grown-ups". I say: "If acting like grown-ups is doing what you do, then something is wrong.". I really hate when someone says: "This is the way things are, so you better get used to it". That's typical corporate monkey act, and it only serves to keep the status quo. We are here to change that status quo, not to replace one convicted monopoly company by a monopoly-wanna-be NGO. If want things to be different, we have to act different. We have demonstrated far and long that we can build better, faster, nicer, better-looking desktops doing things in a fun, coordinated, motivated way. Don't let people tell you that "you are a grown-up now, act like one and follow the rest of the monkeys doing the corporate act".

Working with companies

So, we can coordinate our work with corporations and interest group, but our first objective should be our users. Work openly, make it clear from the start that we work in a certain way and they have to live with that. Companies have policies, we do too. Clarify from the start that policies will be enforced, no matter how much money is at stake. Of course, we can take special actions, like we would do with distros or other groups, but when it makes technical sense, and it helps or at least does not hinder users.

And perhaps policies will need to be adjusted if we get a flow of corporate contributors. I see that happen in KOffice, with a flow of patches by paid contributors having to be reviewed by maintainers who work on a volunteer basis. If those patches don't have the required quality it can take the fun away very quickly. In the end, a volunteer might feel in charge of training people who actually get a salary to work on it, which seems unfair. Perhaps more strict quality policies have to be set for those cases.

Those who mentor SoK know how long it takes to review work done by new contributors. Imagine a team of developers working full time sending code your way. We have to get ready for that. Now, because it's happening and it will probably happen more often in the future.

Epilogue

We do things differently and it works. Let's improve on that, without losing sight of who we are , what we do and why we do it.