[standards-jig] Checking for implementations

Iain Shigeoka iainshigeoka at yahoo.com
Wed Apr 17 17:01:29 UTC 2002


On 4/16/02 8:45 PM, "David Waite" <mass at akuma.org> wrote:

> Of course, I make no secret that I dislike browse. It is directory
> services and pub/sub and introspection (and even feature negotiation, by
> some people's definition) all rolled into a tiny ball. I've always
> gotten the feeling that a large part of its design was based solely on
> the need for a better agents retrieval mechanism and a better way of
> representing users in a groupchat. But I am perfectly happy to help get
> it accepted as a standard; then the community can decide if they need
> something greater later on down the line we can always transition - we
> would already have to do transitioning if that something greater was
> available today.

I'll throw my hat in with you David.  In addition to your points of
contention, one of my problems with browse and several other protocols is
that they seem torn between a design for generic use and particular
applications.  This results in a very inconsistent feel to the protocol, and
to implementations taking one approach or the other leading to them being
compatible but practically unrecognizable as the same protocol.

It would seem we should either be specifying very particular, application
specific protocols (like message and presence) or purely generic protocols
with particular use cases in well-known application areas (browse and the
pub-sub stuff going on now seem like good candidates here).

-iain


_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com




More information about the Standards mailing list