[ANNOUNCE] 0.9.32-rc1 released

Ed W lists at wildgooses.com
Sat Dec 25 13:29:59 UTC 2010


Hi

> Once we've done a full SUSv4 audit we could start thinking about something like a 1.0.0.
> This release will be 0.9.32

I personally think you are artificially attached to keeping the version 
numbers low (ie in the opposite direction)?  Why not call that SUSv4 
release version 2? (or 3?)

I do get that the numbering is completely arbitrary, but equally you 
could be seen as arguing completely the opposite direction?  I mean to 
programmers, what is wrong with a major release of functionality 
occurring between version 2.3.4.5 and 2.3.4.6?  However, to less 
technical folks it's a good hint as to the progress of the project

On the other hand it's not entirely unreasonable to point out that 
"much" of the world tends to use an easily increasing first digit for 
major releases (which in a previous thread it was already agreed that 
nearly all uclibc releases are pretty much "major releases"?), a minor 
release second digit and a point release third digit.  Agreed it's 
completely arbitrary, but it does help keep a sense of perspective on 
progress to an outside observer?

Anyway, the whole point is it's completely arbitrary...  (But I do think 
a major addition such as nptl threads is worth of an increase in major 
version number? SUSv4 audit might also be?)

Cheers

Ed W


More information about the uClibc mailing list