[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: D-I Etch Beta2 - Status update



On Wed, Feb 15, 2006 at 01:10:49AM +0100, Frans Pop wrote:
> On Tuesday 14 February 2006 02:49, Steve Langasek wrote:
> > I wonder where these defaults came from?  The last I heard was that
> > yaird would be favored in the installer, but initramfs-tools would be
> > listed as the first alternative in the kernel package dependencies to
> > provide a better upgrade path from sarge.
> 
> Personally I've always slightly favored initramfs-tools for d-i based on 
> my experiences during installations.

Just for information, i was upto now a strong yaird supporter, since i like
its design better, but two issues have recently made me change my mind :

  - yaird upstream seems MIA (no response to email since early december, no
    other kind of communication from him), and the debian maintainer of yaird
    seems to be unable to take the responsability of fixing bugs without
    consulting with his upstream.

  - the yaird debian maintainer is difficult to work with, as shown by not
    only his not fixing the bug making the kernel uninstallable on
    powerpc/pegasos since now more than 45 days, including a patch, but his
    stuborn refusal to even consider said patch, because it may break some
    working cases and he needs upstreams approval, which he is not getting,
    but also not actively searching. The patch in question is a 3 line patch
    reverting on powerpc a breakage introduced by a hastily done hacky patch
    regarding the infamous ide-generic issue, and as powerpc doesn't build
    ide-generic, i believe it is proven that this module is not needed, or we
    would have known earlier :)

So, all in all, altough i still like yaird's approach better, it is clear that
it's poor maintenance and difficulties with the debian maintainer make it
unsuitable for default. This is unless someone else cames over and hijacks the
package, someone more prepared to handle a missing upstream even.

Friendly,

Sven Luther



Reply to: