Open Source (was Don't you just wish Rev would do this?)

[Bob Warren]

Chipp:

What I didn't mention about Ubuntu's procedure is the following. It 
after release they find  significant bugs, they correct them in a new 
"bugfix" version, submitted to the normal procedures of global testing. 
However, it is the exception rather than the rule. This is NOT what I 
suggested in my original proposal, but the preparedness to make 
exceptions in this way if really necessary would also allay your fears 
about stability I imagine.

But I am still not convinced that it would be necessary in the kind of 
system I proposed. That bugs should occur is natural and normal, even 
with the best prevention. What matters is the TURNAROUND, i.e. the time 
between discovering the bug, fixing it, and returning the fix to the 
user. Three months (or in the case of Linux, 2 years or more) is not 
good enough. If there was anything wrong with post-production patches 
downloaded in the way I suggest, you would soon know about it!** And 
provided they were given absolute priority for correction and the 
issuing of new patches, the system would still be better than the 
current one.

[** And it should be remembered that even good beta testing doesn't 
catch everything anyway. A great number of bugs are found post-release, 
and always will be. But hopefully, these should be the least significant 
bugs.]

Bob