On Fri, Mar 26, 2010 at 05:35:19AM -0700, Zac Medico wrote: > On 03/24/2010 08:47 PM, Joshua Saddler wrote: > > On Wed, 24 Mar 2010 16:12:55 -0500 > > William Hubbs <willi...@gentoo.org> wrote: > > > >> On Wed, Mar 24, 2010 at 09:36:52PM +0100, Ben de Groot wrote: > >>> We agree that this is the minimum that should be done. But our > >>> Python lead stubbornly refuses to honor this reasonable request. > >> > >> On the other hand, I can see his point as well. The news item makes it > >> very clear that python-3 cannot be the default python and that python-2 > >> needs to be installed. > > > > Again, if it *cannot* be the default python, then it *should not* be > > installed by default, which is what will happen if it's marked stable and > > users aren't told to p.mask it. Even then, it'll likely get installed > > first, as users will probably learn about p.masking it only *after* they > > install it. > > Do we have a precedent on this, if for example, we look at the last > time that a new slot of java (like 1.5) came out that wasn't > supported by all packages and therefore couldn't be set as the > default system jvm?
There really isn't a precedent since upgrades of this sort typically either have extremely locked down deps, or just plain don't happen till the vast majority of depndencies are updated. If in doubt, look at the past python upgrades- they've been delayed till all of the major consumers played nice w/ the targeted python version. ~harring
pgpH4KkiKHK0p.pgp
Description: PGP signature