[standards-jig] Software Standards

DJ Adams dj.adams at pobox.com
Thu Aug 16 23:26:07 UTC 2001


On Mon, Aug 13, 2001 at 05:41:45PM -0600, Michael Bauer wrote:
>
> Here's a draft charter on Jabber Software Standards.  It's intended to
> capture our collective thinking on what a "standard" and what "conformance"
> with a standard currently means in Jabber.  Forgive me if it doesn't seem
> very clear.  Perhaps you can help me clarify it with your feedback.

This makes a great start - nice one.

I think the original idea as previously discussed in this list still is
the one we should be aiming for - and indeed fits in with the idea of
the diference between /conforming/ and /complying/ as expressed in the
draft charter.

The intention of the namespace usage document was to be a starter at the
lowest level of granularity for conformity focus. It seems from the comments
that we generally agree that a particular software 'candidate' will
implement a set of these protocol 'granules' - namespaces, as we already
know - against which we can test conformity.

I think the SC JIG sits at a rather central place in a lot of the activities
within Jabber and the Foundation, and that's going to make the job a little
harder - we are trying to define conformity to standards that are emerging
as we go along. And I'm not just referring to the 'draft' set of protocols
on the docs site. I've had first hand experience at trying to wring the
essence of the protocol - which encompasses /rules/ as well as /definitions/ -
in putting together the draft reference for s2s.

As Julian's already said, we also need more than just the protocol
descriptions ("it looks like this") to go on. So there's a lot to do, but
at least it seems we know which direction to go in.

Why don't we pick a fairly harmless namespace, such as jabber:x:oob, and
try and come up with what we think ought to be the underlying documentation
to base the standards compliance work upon? We can then use that as a
model to progress.

What do you all think?

dj




More information about the Standards mailing list