Re: [gentoo-dev] RFC: ion license

[Ciaran McCreesh <ciaranm@ciaranm.org>]

[-- Attachment #1: Type: text/plain, Size: 1912 bytes --]

On Sun, 13 May 2007 09:57:05 +0200
Matti Bickel <mabi@gentoo.org> wrote:
> Ciaran McCreesh <ciaranm@ciaranm.org> wrote:
> > Supporting this would be a huge policy violation, and not so merely
> > as a technicality.
> 
> How's that? I agree that this timely response clause will mean ion-3
> will never go stable. That's the only thing i could envision to be a
> policy violation.

Right, and packages that aren't aiming for stable eventually shouldn't
really be in the tree at all.

A larger issue, though... It requires some way of pushing updates to a
user who hasn't synced for >28 days.

> > I suggest simply removing ion support from the main
> > tree, and sticking it in an overlay that comes with a big warning
> > telling users that they cannot expect any level of QA for those
> > packages.
> 
> Care to expand on "no QA"? Tuomo fixed several QA warnings upstream
> (missing strlen, etc. includes) when i told him (there will be
> patches on our side until the next _rc).

If upstream release a new version that has a serious bug, Gentoo would
be required to include it as the most visible package within 28 days,
even if it is completely unusable.

> Additionally i'd like to point out the bit where he says he don't
> want this license to hinder distributions who just stick with
> upstream, which our policy explicitly recommends. That's why i'm
> trying to reach a compromise on those USE patches we apply. That's
> why the next build will tell ppl to bug me first.

If he doesn't want to hinder distributions, get him to fix his licence.
The way it is now makes it impossible for distributions to do their job.

> In general: i don't think forking is an option. I won't be
> maintaining a fork myself to begin with.

Probably true, from a Gentoo perspective. If there's a significant ion
userbase, someone else will do the work.

-- 
Ciaran McCreesh


[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 189 bytes --]