[NNTP] News and nntp URIs

[Charles Lindsey]

In <87psv63gd3.fsf at windlord.stanford.edu> Russ Allbery <rra at stanford.edu> writes:

>Charles Lindsey <chl at clerew.man.ac.uk> writes:

>> There are now two drafts for your consideration:
>> draft-lindsey-news-nntp-uri-00.txt
>> draft-ellermann-news-nntp-uri-00.txt

>> The first is exactly what I put before you a few months back, except
>> that the '!' has been removed as a possible newsgroup-name
>> character. You all seemed happy with that text last time.

>I personally would prefer to see the NNTP URI scheme wait on the
>publication of the new base document and then follow the approach that
>you're taking of picking up the new NNTP work.  I'm still not sure that
>the full wildmat support in the URI is really the best way to go, but I
>can't come up with any strong argument against it other than "it looks odd
>and I'm not sure how many people will bother using it," which doesn't seem
>strong enough.

Sure. The "lindsey" draft makes normative reference to the new base
document, so it would not escape from the RFC editor until that had been
published.

The real question is whether to have a 'wildcard' notation which is the
same as something that servers are likely to provide (i.e. wildmat), or
whether to invent yet another one. If you write

    news://some.server.example/*.test

(expecting to receive a list of available test groups), then for sure the
search for matching groups must be done on the server (which just scans
through its active file). The 'ellermann' draft seems to imagine that the
scan will be performed by the browser on which you click that URI, which
is crazy because it would have to download the whole active file first.

Even present NNTP servers will support wildmats, though not necessarily in
the precise form of our latest draft, so they might not be able to cope
with

    news://some.server.example/*.test,!de.test

I would have no problem saying in the draft that only part of the
wildmat functionality would be available with old-style NNTP servers (and
a browser could soon detect whether it was attached to a new or an old
one).

I will also post this message to the uri at w3.org list if I can figure out
how to do it.

>-- 
>Russ Allbery (rra at stanford.edu)             <http://www.eyrie.org/~eagle/>

-- 
Charles H. Lindsey ---------At Home, doing my own thing------------------------
Tel: +44 161 436 6131 Fax: +44 161 436 6133   Web: http://www.cs.man.ac.uk/~chl
Email: chl at clerew.man.ac.uk      Snail: 5 Clerewood Ave, CHEADLE, SK8 3JU, U.K.
PGP: 2C15F1A9      Fingerprint: 73 6D C2 51 93 A0 01 E7 65 E8 64 7E 14 A4 AB A5