On Thu, Oct 19, 2000 at 11:24:50PM +0300, Eray Ozkural wrote:
> Jason Gunthorpe wrote:
> > > * Ideally, all packages would lie flat in a single directory. example
> > > /package-pool
> > No. This is not ideal, it impeads the ability of the ftp team to
> > manipulate the archive.
> Yes, but isn't the idea to provide automated tools that will require
> minimal manual intervention? I guess tools can check for consistency or
> correctness better than people, and the admins can just use a set
> of tools that will do all necessary tasks.
Tools break, or don't work how they're meant to, or don't do something
obscure you might want. Being able to understand things independently
of your tools is a *huge* win when you're trying to reason about whether
your tools are correct or not.
> Well, a tool can run a hashing function. Once the tools are stabilized,
> they will have no problems.
Except that we don't really want the tools to stabilize all that much.
We want to be able to keep moving forward, implementing better ideas,
getting rid of old crufty problems, and so forth.
> For the current set of packages, this may be true. What I'm questioning
> is whether this will hold for 20.000 or 50.000 packages? What's the
> asymptotic time complexity? ;)
20k packages probably imples about 100 gigabytes of archive space,
multiplied by a factor for however many new architectures we start
supporting. Depending on how much more bandwidth is common and how much
disk space is available, we may well have other things to worry about
than how well our hash function works or doesn't.
Stuff in the future is inherently unpredictable outside of academia
(inside of academia it's only predictable because you can say "assume
these are the problems we'll face, how can we solve them"). Sure, keep
an eye on what might happen next, but focus on making things work right
*now*.
Cheers,
aj
--
Anthony Towns <aj@humbug.org.au> <http://azure.humbug.org.au/~aj/>
I don't speak for anyone save myself. GPG signed mail preferred.
``We reject: kings, presidents, and voting.
We believe in: rough consensus and working code.''
-- Dave Clark
Attachment:
pgpwLwQRBCNpO.pgp
Description: PGP signature