It's taken so long because they've rewritten most of it. The fact that it is "unusable and unstable" is normal for a piece of software a couple of months off beta. I've been watching it for a long time - trust me.
My understanding of the milestone process is that they close the tree every three weeks for stabilisation. This seems to take about a week, so there's roughly about 3 weeks between every milestone.
Being a real programmer is not about doing hard things - it's about getting something real done. It's about being lazy and still getting the job done. And while it is true that small amounts of handcoded assembly will be faster, writing whole programs in assembly is impossible and would be slower overall than well optimised high-level code anyway (as well as buggier). Furthermore, the time spent in rewriting in assembly would usually be better left to tuning algorithms.
Slashdot Mirror
640Mb should be enough for anybody!
I never meta moderator I didn't like.
It's taken so long because they've rewritten most of it. The fact that it is "unusable and unstable" is normal for a piece of software a couple of months off beta. I've been watching it for a long time - trust me.
Erm, XSL and XPath are no finalised standards, and hence "support" counts for diddlysquat.
See "http://bugzilla.mozilla.org/show_bug.cgi?id=3826" .
My understanding of the milestone process is that they close the tree every three weeks for stabilisation. This seems to take about a week, so there's roughly about 3 weeks between every milestone.
Being a real programmer is not about doing hard things - it's about getting something real done. It's about being lazy and still getting the job done. And while it is true that small amounts of handcoded assembly will be faster, writing whole programs in assembly is impossible and would be slower overall than well optimised high-level code anyway (as well as buggier). Furthermore, the time spent in rewriting in assembly would usually be better left to tuning algorithms.