[
Lists Home |
Date Index |
Thread Index
]
robin.berjon@expway.fr (Robin Berjon) writes:
>> Thanks - that makes it clear. Seems like an odd omission.
>
>It is indeed. But there is no work that I know of going on around HTTP
>so I'm unsure how much can be done.
I don't think this work needs to happen in HTTP specifically - it seems
like it could be broken out as a more general MIME-related issue. This
kind of encoding seems useful in contexts way beyond HTTP.
On the other hand, getting that break-out to happen might be a larger
challenge all its own.
>> Has anyone taken this particular challenge up with the IETF?
>
>I've talked with IANA people about the registration of new content
>codings (since they're in charge of it), and asked if adding
>parameters would be ok. They kindly answered that they had to dig up a
>very old form from some random location and that they had no idea.
>Something in the kind tone suggested that they might as well have been
>communicating with a perfect alien. Either way, a new content coding
>needs to be a RFC -- which is where the issue would likely be banged
>on -- so I had no chance to look further into it.
You (or the W3C binary folks) might write up a trial balloon
Internet-Draft and see where it goes. That's relatively easy and it
would at least start the discussion, though it might be something that
should proceed as part of the W3C Binary whatever work and probably in
concert with the IETF/IESG/IANA folks.
>No one I could find seems to know why parameters are not allowed there.
It could be an omission, or there may be something that isn't obvious.
>> (I worry that other people will read it the way I did and start
>> hacking encoding parameters into Content-Type rather than use
>> Content-Encoding.
>
>But then we can point fingers and laugh at their bad taste.
Doesn't tend to work. See RFC 3205 for one effort at this:
http://ietf.org/rfc/rfc3205.txt
Larry Masinter's even posted a brilliant exposition of how it applies to
SOAP, to www-tag, no less:
http://lists.w3.org/Archives/Public/www-tag/2003Feb/0208.html
Result: nothing noticeable.
--
Simon St.Laurent
Ring around the content, a pocket full of brackets
Errors, errors, all fall down!
http://simonstl.com -- http://monasticxml.org
|
