Paul Everitt <paul@...> wrote:
<about the locator; here are my reactions>
>1) Centralized or distributed?

Centralized; there aren't that many Python modules, and new
ones don't show up every day. A distributed system will be more
complex.

>2) Anyone can enter data, many can enter but data must be reviewed, or
>only one/several can enter data?

Anyone, initially.

>4) Is this a module registry, or should we also try to track authors,
>documents, or other artifacts?

Stick to modules initially; there are already Web pages at
python.org listing documents, & new documentation is even more rare
than new code.

>8) Is browsing (as opposed to searching) desirable?

Yes. We want something that can produce a large text file,
for inclusion on CD-ROMs, posting to news.answers, etc.

>9) Are we trying to do value-added things, e.g. confirm that the author
>exists, weed out entries that get stale, etc.?

No; again, the volume doesn't warrant the programming effort involved.

>As for the schema, I propose the Linux Software Map as a starting point.

Good idea; the LSM has been discussed before on this SIG.

In general, simplify, simplify! Better to start small...


Andrew Kuchling
amk@...
http://starship.skyport.net/crew/amk/

_______________
LOCATOR-SIG: Discussions about a Python Locator for resource discovery

send messages to: locator-sig@...
administrivia to: locator-sig-request@...
_______________