On Sat, 2005-01-22 at 11:00 -0600, Daniel Goller wrote:
> you could even consider games.eclass as frozen and leave it as is, 
> avoiding any work to change ebuilds using it
> then start with games-1.eclass, should this thread lead to an accepted 
> solution

You don't seem to understand that what you are proposing is ludicrous in
many situations, yet you propose to *force* this asinine behavior on us
all.

My answer to this is simply that I don't go piss in your pool, so don't
piss in mine.

> then games-2.eclass is used for ebuilds after a change is introduced
> the versioning scheme is not the problem to me, id like it finer grained 
> as you wouldnt end up with any more or less eclasses either way but have 
> more flexibility as to how big you think your change was

The current system is very simple.  A change to the way things works
requires a new eclass.  Versioning has *zero* to do with this, as it
doesn't matter if I call the eclass "games-1.eclass" or
"games-q3mod.eclass", except that the latter has some instant
notification of purpose that a simple version does not.

This is my whole point.  Forcing version numbers onto eclass that do not
require them is a waste of time and energy for all involved, and will
probably be fought to the bitter end simply due to its complete lack of
merit.

If an eclass needs a new "version" then make one.  It is that simple.
Don't force all of us to follow your versioning simply because you think
you need it.

-- 
Chris Gianelloni
Release Engineering - Operations/QA Manager
Games - Developer
Gentoo Linux

Attachment: signature.asc
Description: This is a digitally signed message part

Reply via email to