conference call (fwd)

[John Unsworth]

At 09:58 AM 6/25/2002 +0900, you wrote:

>I might not be completely awake yet, but I can't seem to figure out
>what a RFP is.  (I did manage to expand MM2002 -- CW) On a third
>thought, is it similar to a CFP?

Yes--a request for proposals.

> > Council agreed, on the question of certification, that it prefers the idea
> > of reviewing software to the idea of producing a test suite or certifying
> > software.  Therefore, we will produce a call for reviews (Matthew and 
> Syd),
> > and collectively encourage knowledgeable individuals to submit software
> > reviews to TEI-L, and the TEI webmaster will cull from those reviews to
> > post on the web site.
>
>Does this has to be an either or?  With something as complex and
>diverse as the TEI set of DTDs, I can't really see how a review could
>provide much more than a general thumbs up or down.  The needs I have
>differ substantially from the average TEI user (if such an animal
>exists -- I personally haven't yet encountered one), but I would be
>happy to provide the necessary files for a test suite that check for
>all kinds of CJK compatibility.  While it might not be feasable in the
>short term, I hope that we can keep the production of a test suite as
>a future work item.

If you're interested in working toward a test suite, you could probably 
persuade the Council that this is a worthy goal--the short-term sense of 
the discussion yesterday was that it would be a pretty difficult task to 
produce this, though, given the many forms the DTD can legitimately take, 
and certain kinds of vagueness (you refer to one below) in the way TEI 
specifies some things.  Your offer of files to test for CJK compatibility 
does suggest, though, that we might *ask* whether people out there in the 
user community have already developed similar test files for features of 
importance to them....

>Sebastian brought up, among other things, interoperation with W3C
>standards.  Has there been any further discussion about that?

During this phone call, only in a general way.

>I would
>be especially interested to know what the Council things about the
>future interaction with the CSS group of recommendations.  Do we
>propose to use them in any specific way (for example as the content of
>the "rend" attribute, announced somewhere in the header).  I certainly
>feel the need for some more formal declaration of this.  [and, to come
>back to the test cases issue, I think test cases really make only
>sense if we have some kind of overall processing model where we can
>say 'under these and these conditions, we expect to have this and that
>result' -- but currently the overall model is far from being clearly
>specified.]

Interaction with CSS didn't come up, but we can of course take it up here 
on the list.

> > On character encoding: Lou estimates that this workgroup will keep to its
> > schedule, producing a draft report by MM2002, and a final set of
> > recommendations by the January Council meeting.  Thanks to Christian
> > Wittern for keeping the group on task and on time.
>
>OOh? I have to admit that I am far less optimistic.  While we might be
>able to produce some kind of report, I am not so sure about the final
>recommendations.  There are a lot of issues far from being solved and
>some architectural decisions will depend on the development towards P5
>as a whole.

Could you imagine making recommendations within the scope of P4 in January, 
with a separate set of recommendations for P5?

John