ietf-822
[Top] [All Lists]

Re: RFC 2047 and gatewaying

2003-01-08 20:12:47

In <20030108003655(_dot_)GA11638(_at_)ussenterprise(_dot_)ufp(_dot_)org> Leo 
Bicknell <bicknell(_at_)ufp(_dot_)org> writes:

In a message written on Tue, Jan 07, 2003 at 03:20:55PM -0800, Russ Allbery 
wrote:
* Extend MIME (eg, define types) to allow for "message/rfcM".  Boom,
  e-mail is good.

You've lost me here.  How does defining message/rfcM let you deal with
UTF-8 headers in e-mail?

message/multipart
- text/plain
   "This is a news article you need to approve/disapprove."
   or
   "This is a news article that has been gatewayed to e-mail."
- message/rfcM
   - UTF-8 article, base64 encoded

The beast you are describing exists already. It is called
application/news-transmission, and it is perfectly suitable for tunneling
any news article (with UTF-8 in its headers) through any mail system.

It is indeed an ideal tool for sending articles to moderators (it even
comes with a parameter saying "this is for you to moderate"). It is exactly
what Usefor wanted to use.


This sounds like you're talking about encapsulating submissions to
moderators rather than gatewaying them.  This breaks almost all existing
moderation software.

I submit the number of moderators is /vastly/ smaller than the number of
people who read e-mail, or read news, or are on a list gatewayed between
them.  If the fallout is that moderators need to update their software
I'd say that's affecting the smallest group possible.

They are also people who are likely to update to the latest and greatest
with any news change, as I think they are likely to want new bells and
whistles, or at least be able to understand them as they moderate.

And that is what we thought, but sadly it did not turn out to be so (even
though it is already used to some extent in Usenet). The moderators just
did not want to know, and so we had to backtrack and write detailed rules
on how to send email to moderators without encapsulation.

In the longer term, I hope that the encapsulation method (which we all
agree is technically superior) will win out.

And in principle it would also work for other gateways, except that User
Agents currently do not know what to do with it except store it into a
file for later examination. So it does not solve the present problem,
except maybe in a few very specialized applications.

-- 
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_)clw(_dot_)cs(_dot_)man(_dot_)ac(_dot_)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

<Prev in Thread] Current Thread [Next in Thread>