One Linux, Or Many, Or None?One Linux, Or Many, Or None?
DistroWatch <a href="http://distrowatch.com/weekly.php?issue=20090202#feature" target="_blank">chatted</a> with none other than Linus Torvalds about Linux in general, wherein he admitted "multiple [Linux] distributions aren't just a good thing, I think it's something absolutely required". I'd agree, but is it required as a first step or a last one?</p>
DistroWatch chatted with none other than Linus Torvalds about Linux in general, wherein he admitted "multiple [Linux] distributions aren't just a good thing, I think it's something absolutely required". I'd agree, but is it required as a first step or a last one?
It's hard to ignore the concept of having a single unified Linux base to draw from, since it solves a number of problems at once (e.g., the continually egregious issue of binary incompatibility between distros). Linus, on the other hand, prefers plurality as a fairness measure:
... having multiple players just keeps everybody honest, and allows you to compare them. It may all look a bit messy and complex, but I'd much rather have a multi-party system over a single-party one. Even if it's more complicated.
Sure, but that brings up another question: is it possible to keep everyone honest without resorting to such a messy, complex system? I'd like to think that it is, especially after having spent as much time as we have dealing with the consequences of things like multiple competing audio architectures. The answer to the criticism that this is all a messy and complex business should not be "live with it".
I've said -- and I'll continue to say -- that Linux does not need to beat Windows at its own game, or anything that superficial. Linus echoes this when he says that he doesn't believe there's going to be a year of of the Linux desktop, but rather a series of incremental advancements towards making Linux a solid choice for everyone.
But what's becoming clearer is that without a single unified base for all distributions to draw on -- a base that's broader than just the kernel or the toolchain -- Linux will have problems that have nothing to do with gross metrics like market share. Things like the Linux Standards Base are a good start, but they're not remotely deep or far-reaching enough; they need to be the basis for much more ambitious work.
Linux, whether we like it or not, has become more than just a kernel. It's high time the Linux community accepted this and acted on it. We can still have a future where Linux comes custom-tailored for every device you can think of -- but it would be nice if we had a one-size-fits-all default as well.
About the Author
You May Also Like