[Gnu-arch-users] Re: Automatic archive discovery, take 1

[Miles Bader]

Colin Walters <address@hidden> writes:
> > Note that Tom's two 2003 archives have completely different urls in
> > Andrew's list!
> 
> But those one of those URLs is just a simple mirror for the other, I
> believe.

I'm not sure if the older archive is mirrored, but I do know that the
mirror of the newer archive is preferable in general (because it
contains .listing files), which I presume is the reason it's on Andrew's
list instead of the other one.

IOW, in this case the flexbility -- of choosing a mirror or not -- is a
_good_ thing.

> >  I suspect that this may be a common event -- to use the change to a
> > new archive to also move to a new host.
> 
> Really? I wouldn't suspect it's very common, but I guess we don't have
> enough data to know.

Well considering all the other responses to your post, it looks like
it's _very_ common.  In my case, I hope I'll be able to use an FSF host
for my future archives, but my current `public' archive is at
arch.linuxguru.net.

> > My feeling is that people _will_ move their archives around over
> > time -- but will often want to keep their same email address (email
> > addresses are often much easier to maintain because they require
> > almost no resources) -- and that we shouldn't penalize this case.
> 
> Yeah.  It's a tradeoff between making moving archives a little easier
> for the archive maintainer, versus making registering and using archives
> easier for everyone using that archive.

I'm a bit confused by what exactly you mean, so I'll state what I
understand:

  * A `central' index for a particular user, which can be used to
    locate his archives, would be useful both for the ordinary user
    (less pain tracking archive changes) and the maintainer (less pain
    making sure users track archive changes!).

  * A restriction such as you earlier proposed (that having `disimilar'
    URLs for different years requires using a different `main name')
    will make using archives _harder_ for ordinary users, because
    it will put some of the onus back on them to track archives.

Anyway I have no idea why it's a hard problem, surely a `meta archive'
such as you propose could be a simple text file reached by the
meta-archive url, which could have a list of real archives and their
URLs.  This shouldn't place any restriction at all on the URLs of the
real archives.

-Miles
-- 
`Suppose Korea goes to the World Cup final against Japan and wins,' Moon said.
`All the past could be forgiven.'   [NYT]