[JDEV] Standards-track HTTP polling JEP? (was: jabberwebclient)

[Joe Hildebrand]

David Yitzchak Cohen [mailto:dave at dave.tj] wrote:
> Eeeek ... please don't use HTTP Polling if you don't have to. 
>  It's not good.

/me reacts to the drive-by troll.

Sure, it's resource-intensive, but it will always work, if you can get to
the web.  And it's no worse than some of the solutions proposed.  If you've
got new requirements for the new standards-track HTTP polling JEP, I'd love
to hear them.

Here is the list of things I'd like to see in a new JEP:

- Use of application/x-www-form-urlencoded name/value pairs instead of funky
, and ; syntax.
- Ability to either use key stuff or not (enforcible by either side)
- Ability to try multipart/mixed, if client supports it
- Ability to recover from multipart/mixed socket being truncated by a
firewall (potential race condition; potential data loss condition)

-- 
Joe Hildebrand