[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [seul-edu] Educational binaries and binary directory
On Tue, 29 Feb 2000, Doug Loss wrote:
> rpmfind database and Debian's equivalent don't quite cover this, as this
> shows all the package types available for each app rather than showing
> all the apps for each package type.
Hm, true. If I'm a Debian user looking at Freshmeat, I only care about
1. is there a .deb for my version (slink, potato, woody, etc.)
2. if there isn't a .deb, where do i get the tarball?
So, if I could set a preference for the site (.deb, tarball, or both)
and have pages just list links to the formats I care about, it would
be nearly perfect.
Just about the only other thing that is missing that I might be interested
in seeing (and as I've pointed out before, this may be more of interest to
developers than users) is a "work in progress" status if a package is
*not* available for a certain distro. This sort of info is best
maintained by each distro rather than at Freshmeat, so a link back to
Debian for the .deb's (and various versions if they exist) and other info
about the package (such as work-in-progress status) would be helpful. For
instance, at present, if you visit http://www.debian.org/distrib/packages
you find a form for looking up existing packages. There are some
checkboxes for "version" and "release". Perhaps an additional checkbox
could be added for "in progress" could be added to "Version" and the
in progress status could be added to the index as developers announce
their "intent to package". In any event, none of this would need to
be tracked by Freshmeat. A single link could be provided on the package
name to "Search Debian" in addition to any hard links to .deb format
packages that Freshmeat provides. For example, to search for my package,
"typist", the link would be:
This searches all versions and all releases for packages containing the
keyword "typist". If we added an "in progress" release, you'd get that
addtional info "for free".
Of course, there is one imperfection in this scheme. Package names at
upstream don't always match character-for-character the package name in
the distribution, so that might need to be adjusted manually as
discrepancies are found. Likewise, our "Search on: package names only"
searches for a substring, and doesn't allow for the fact that the same
substring might appear in multiple (and possibly unrelated) package
names. But I suspect the # of "false hits" will be low enough to make
this a useful tool.
nSLUG http://www.nslug.ns.ca email@example.com
Debian http://www.debian.org firstname.lastname@example.org
[ pgp key fingerprint = 7F DA 09 4B BA 2C 0D E0 1B B1 31 ED C6 A9 39 4F ]
[ gpg key fingerprint = 395C F3A4 35D3 D247 1387 2D9E 5A94 F3CA 0B27 13C8 ]