Skip to Content.
Sympa Menu

sm-discuss - Re: [SM-Discuss] stable/testing branches

sm-discuss AT lists.ibiblio.org

Subject: Public SourceMage Discussion List

List archive

Chronological Thread  
  • From: Phil/CERisE/KG6MBQ <cerise AT littlegreenmen.armory.com>
  • To: Eric Womack <eric AT lasvegasdata.com>
  • Cc: Ari Steinberg <ari.steinberg AT stanford.edu>, <sm-discuss AT lists.ibiblio.org>
  • Subject: Re: [SM-Discuss] stable/testing branches
  • Date: Thu, 25 Jul 2002 17:49:17 -0700 (PDT)

Eric Womack said:

> But even to say that we will determine if a package is is stable because
> Debian says it's stable is too much. WE need to test is and determine for
> ourselves if something is stable. We are already developing a clear
> guideline for the proper use of each grimoire is, and I think we should
> continue in that direction.

I think that's setting ourselves up for way too much work.
I think we should simply list as stable what the authors would list
as stable.

> I disagree, and I believe your extension of my reasoning creates a logical
> fallacy. We are dependent on the kernel team for source code, but not for
> what version is more stable/usable (linux-vanilla, linux-xfs, linux-ac,
> etc.). In other words, the multitude of authors we "depend" on are
> producing a product (of sorts) for our use, where as the usage of the
> Debian stable choice is an opinion. While that opinion is based upon
> their testing in their software, and certainly worth looking at, to fall
> back on them for rating a package still seems wrong.

I concur that falling back on Debian for a rating is silly. I
believe that the code authors should be consulted. After all, we use
their version numbers.
One thing that I think should be done is some sort of internal
versioning. It drove me nuts while I was writing a slackware upgrade
script was that the package names and versions were not consistent at
all. For instance AbiWord was packaged as "abi" until 8.0, when it
became 8.1.
I'd recommend distributing a list of versions in descending order of
preference. That way, if an author wants to retract a release, then we
don't end up leaving people out of the loop with an oddball version
number. If the user's current version matches the first line of the
list, then it's considered up to date. Otherwise, it is updated to the
one on the first line of the list.
One minor worry is if a release is considered the best and is
somehow lost by the author. It's improbable, but I suppose that in
that case, a universal mirror might be a good idea (although I've
argued with AFK to the contrary).

-Phil/CERisE/KG6MBQ





Archive powered by MHonArc 2.6.24.

Top of Page