Open source - feel good, but work better.

Over on The Guardian, Andrew Brown has applied his unique brand of gentle melancholia to the subject of open source software. Open source, he says, is often acutely dependant on big corporations, not the warm fuzzy community group hug of legend. There is no reason for Oracle to keep OpenOffice going, nobody else writes code for it, and even Google has overlooked OO in its Summer of Code – inspired, he thinks, by its adherence to the Google Docs dogma.

And as for professionalism – even Ubuntu broke his sound card on the last upgrade, and there was nowhere to turn for help. You need professionals doing a professional job, and millions of dollars, to make things work: who's going to bother?

I hope I can make Brown feel better – but first, I must make him feel worse. While it is certainly true that open source software isn't perfect and support can be patchy, it is just as true for the stuff that comes in boxes. I have upgraded Windows and had things not work, and been unable subsequently to find out why or fix them. I have struggled to get to the bottom of bugs in OS X, with Apple's tech support mute on my pain. It is certainly true that while most things work first time in Linux, anything that takes longer than five minutes to fix is not going to be worth it. Yet Ubuntu keeps getting better at the former and less liable to the latter. Brown might not be able to put his finger on why or how – and the dynamics of open source development are certainly not straightforward – but there it is.

Propriety software and support has got as good as it's going to get. Open source keeps getting better, and it's already up there with the paid-for stuff.

As for OpenOffice and Oracle: even if nobody did a stroke of work on it ever again, it remains a fine piece of software – reliable, fast, flexible. For most purposes, like a well-designed typewriter it'll work as well in a hundred years time as it ever did. (You can still run 50 year old software perfectly well on your PC).

I don't believe Oracle will stop work on OO anyway; it enjoys annoying Microsoft far too much, and it will be thinking hard about extending its new top-to-bottom server strategy to encompass the desktop. You or I or Andrew Brown might not have the skills to add new features to OpenOffice, but open source software was never about you or I or Andrew Brown doing any such thing. Someone can, and if the need or desire is there it will happen – because nobody can stop it. And if the need or desire isn't there, then they won't and quite right too: it's not like Microsoft Office, where a burning need for new revenue forces massive feature churn every five years.

Benevolence is a poor substitute for professionalism, as Brown suggests. But open source isn't about benevolence, for all that's a very pleasing side effect. It's about focus on utility, about making things work for the sake of what they can do, not for what can be sold. That's why open source software, for all its frustrations and raw edges, is so often a better experience than the paid-for stuff, and why so many companies find it worthwhile to pay their employees to work on things to give away. A community may be a benevolent place to be, but it creates and enables more than any individual can – and that's why it survives.

And that's why open source will survive, and prosper. It's too darn useful to stop.