Strict Open Source Policies Lead to Fail

Ubuntu and Firefox are two of the biggest open source projects in terms of desktop end users. Yet, the same open source philosophies that made these projects so popular are a weak spot that the competition can, and will, use to get ahead. If they do not change their policies I see Firefox being destroyed by Chrome, and Ubuntu losing ground to Windows 7 and Chrome OS.

HTML5 is on the move. Pretty much every browser other than Internet Explorer is supporting it pretty well. However, the codec issue on the video tag is going to be a major headache for Firefox. YouTube, which is really the only video site that matters, has decided to go with H.264. Firefox can’t include H.264 because of patents, and can only support Ogg Theora. Chrome and Safari do not have this problem. IE will also not have this problem, if and when they get their act together.

This is due to the balance of open source that people like RMS do not understand. If you demand everything be purely open source, you will be missing out on something. The same goes if you demand everything be purely closed source. Some applications only exist closed, and some only exist open. You will have to use both to get an optimal user experience, especially on the desktop.

The overwhelming majority of users do not care about whether something is open or not. They only care that it is free as in beer, and that it has all the features they desire. Regardless of speed benefits, Chrome will beat Firefox because of its willingness to not be 100% pure open source, and include things like the patented H.264 video codec.

For those of you who do not know, Ubuntu has a very strict release cycle. Every six months they release a new version of Ubuntu. The most recent release is 9.10, released in October 2009. It includes version 2.6.31 of the Linux kernel. If there is a bug or security patch to that kernel, Ubuntu will push out an update to the users, such as the current version 2.6.31-17. The thing is, the current version of the Linux kernel is actually 2.6.32.5. Ubuntu 9.10 users will never get version 2.6.32. In order to get a newer version of the kernel, they will have to build it by hand, or wait for Ubuntu 10.4 in April.

What’s the big deal? The kernel version barely makes a difference in the user experience. Most users will probably not notice or care, even if they are stuck with a very old kernel version. Well, what if you apply the same update policy to something like Firefox? Firefox 3.6 was just released, but Ubuntu users will not get that update. Windows and Mac users will, because Firefox updates itself on those platforms. Ubuntu users will have to wait for Ubuntu 10.4, or jump through hoops to update Firefox by hand. They have to wait 3 months for a very significant user experience update.

The problem with these two open source projects, and others, is not that they are open source. Open source is awesome, and it’s the only reason these projects can even compete in the first place. The problem is that unlike say, the Apache web server, these target every day users. However, their philosophies and policies do not make the experience of the user their number one priority.

Yes, there are good reasons behind these policies. I support stable software releases and being as open as possible. It’s simply that the user experience is even more important than either of those. You need to make exceptions to policy when the user is suffering. If a store didn’t make a reasonable exception to its return policy, you would give it a bad review and maybe even stop shopping there. Why can’t open source projects also make reasonable exceptions? If they don’t, they’ll just slowly lose ground to their competitors.

This entry was posted in Technology. Bookmark the permalink.

One Response to Strict Open Source Policies Lead to Fail

  1. James says:

    “Willingness?” I’m not sure H.264 support has *that* much to do with ideology. Opera has stated that they won’t include H.264 support because the licensing fees are too expensive. I’m sure that factored into Mozilla’s decision.

    When Chromium ships in a lot of distributions, it won’t support H.264 either; distributions will have to strip it out due because of the legal issues. Fedora is considering starting work on integrating Chromium with Gstreamer in order to have a legal way to distribute the browser for free and let the user plug the codec in without Fedora Project or Red Hat getting sued. Or maybe everyone will just download licensed binaries from Google. Hardly ideal, but hardly unlikely, either. *shrug*

    YouTube decided to go with H.264 over two years ago; I think up until now it’s mostly been used for Apple TV and iPhone (and other mobile phones?). Given that relations with Apple have been weakening somewhat as of late, it’s not completely ridiculous to postulate that they’ll start moving toward Theora in the future.

    I dunno. It’s pretty similar to the Linux MP3 situation, which hasn’t drastically changed much in recent years.

Comments are closed.