Re: draft-ietf-usefor-article-06 and last call

New Message Reply About this list Date view Thread view Subject view Author view

From: Charles Lindsey (chl@clw.cs.man.ac.uk)
Date: Mon Jan 07 2002 - 13:21:49 CST


In <8GKW6e1ocDB@3247.org> list-ietf-wg-apps-usefor@faerber.muc.de (Claus Färber) writes:

>It should also be noted that many users, esp. corporate users, as
>well as newsadmins are reluctant to install new software. Even if
>they had a benefit from it, they might not be the ones who decide.

They will install new software just as soon as their users make enough
fuss.

> 1. A user is not able to read/subscribe to the newsgroup.[1]

So? A chinaman who wants to subscribe to a Chinese newsgroup adhering to
the new conventions will have to acquire a suitable newsreader. We have
always known that. The important thing is that no current usage is
impacted.

> 2. A user is not able to post to the newsgroup.[2,3]

Ditto

> 3. A user is not able to followup to an article cross-posted to
> the newsgroup.[2,4]

Yes, cross-posting to such a newsgroup by a person who is actually reading
one of the "normal" groups in the crosspost indeed has to be considered.

> 4. A user's software creates bogous Followup-To headers in cross-
> posted posts due to character set conversion issues.[5]

Ditto.

> 5. Articles are dropped as illegal.[6]

See (1)

> 6. A sysadmin can't create the newsgroup.[7]

This is a server issue. We have established that servers, per se, seem
immune to UTF-8 problems. If the sysadmin really wants the group, he will
find ways to create it (such as by getting a decent editor).

> 7. A sysadmin creates the newsgroup incorrectly due to
> character set conversion issues.[7]
> 8. Automated creation of newsgroups breaks.[8]

Again, if the sysadmin cares, he will get around it (AIUI, the problem was
simply a check that failed - the check needs to be disabled, and INN 2.4
already disables it).

Groups with UTF-8 names are not going to appear overnight as soon as our
draft appears. They will be formed within hierarchies that have a need for
them, and the administrators will make sure that serving sites that users
of those hierarchies are likely to download from get themselves properly
organized.

The big worldwide backbone sites that pride themselves on carrying
"everything" are clueful enough to ensure that 'newgroup' requests are
observed. That will leave lots of smaller sites whose sys files probably
do not include those hierarchies anyway, and a few sites that do take
those hierarchies whose users may have to prod their newsadmnins to create
the groups manually (which is a common enough situation with small sites
anyway). BUT, once created manually, those sites will not have to do
anything else to their servers beyond ensuring that they are fed the group
from somewhere.

> 9. Per-hierarchy configuration of moderator addresses
> (%s@moderators.example.com) breaks.

Yes, we know that, but that is a problem for the administrators of the
hierarchies concerned. They will just have to refrain from creating
moderated groups with UTF-8 characters in them until they have some
solution they consider workable (it is really a problem for the IDNS
people).

>10. Generic Mail<->News gateways break.[9]

Gateways are usually constructed to cope with specific situations (e.g.
specific mailing lists). The only generic situation is likely to be
mail2news kinds of things. That may inconvenience a few individual
posters, but again only those who want to use the new groups, and nobody
forced them to use mail2news.

>11. Breaks other standards, such as URI and NNTP specs.

The URI people indeed need to get their standard sorted out (and the IDNS
effort should already be prodding them to do so). There is no problem with
NNTP, since the new proposed standard is explicitly written to assume UTF-8.

-- 
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@clw.cs.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


New Message Reply About this list Date view Thread view Subject view Author view


This archive was generated by hypermail 2b29.