notes on confernece call of 29 july

Notes on HTML-Math ERB Conference Call
29 July 96
---------------------------------------------------------------------

In attendance:

Robert Miner	The Geometry Center
Robert Sutor	IBM
Ron Whitney     American Math Society
Ralph Youngen   American Math Society

Other people whose first names began with a letter other than "R" were
asked to leave.

[Notes prepared by RW.  Corrections welcome.]
---------------------------------------------------------------------

We proceeded in the order of the proposed agenda.  Sutor is "Bob"
in these notes; Miner is "Robert".


Acknowledging Bob's recent posting on Stephen Watt's return to IBM
this fall, Ron asked if others had opinion on giving our materials to
Stephen and the Safir group for critical comment.   Robert felt that
it might be more difficult to modify our work later to accommodate
Safir suggestions, so we should send our materials sooner rather than
later.  Bob mentioned that all of France is on holiday in August, so
that we might not get any reaction for a while.  No one on the ERB
has spoken yet against Watt's proposal, but then it's also not clear
that everyone has had a chance to even hear of it.  Please do let
RW (and the rest of the list) know if you have reservations.

There was brief discussion of Ralph's overview.  Robert said he agreed
generally with it.  See more below on where we are and what we hope to
accomplish in our Boston meeting.

Discussion of the Hammond posting centered on the validity of
Hammond's contention that it is very important that we produce
something of MIME type text/html.  Bob felt that we would actually be in
*better* position by producing something of another MIME type since
this would facilitate the use of plug-ins.  If our markup is good
and there is plug-in support, major browser manufacturers may pick
it up themselves, but they needn't from the onset.

Bob reported on discussion of HTML-Math at the OpenMath Workshop which
met in Zurich over the weekend.  He made a brief presentation in
Zurich which outlined the objectives listed on our ERB page, showed
the list of committee members, and mentioned the 4 viewpoints
(Wolfram, MINSE, Pike, OpenMath) outlined in a message from RW to this
list a couple of weeks ago.  Comment at the Workshop sounded generally
favorable toward the efforts of this ERB, and in any case it appeared
that the Consortium members wanted their relation to the HTML project
to be advisory and not competitive.  There was some strong sentiment
that HTML Math should allow specification of "context" information
(this had been called "lexicon" information at one time) so as to
account for OpenMath's "object" layer of transfer.  As in the Pike
DTD, contexts are domains of mathematical discourse (e.g. linear
algebra, differential geometry, quantum mechanics; domains form some
form of tree or DAG), and context information is supplied by pairing
an object with a domain.  (Bob can fill this in further.  I'm
imagining the pairing as between notation and object [e.g. <"i",
complex unit of Complex Analysis>], but this appears more as a triple:
<notation, object, context>.)  The four attendees of the conference
call felt this capability is well within the contextual information we
have discussed as being part of HTML-Math.  We should remain aware of
OpenMath's work in contexts.  Bob mentioned that he would post URLs
and other information resources as they become available.

Concluding the call, Robert said he felt that things had come "undone"
to some degree and was concerned about where we would be come October.
Ron said that he felt the Wolfram proposal had been just that, and
that although there had been some discussion on it, the original view
of it was as a proposal and not as a formally agreed upon
specification or even direction.  Given the difficulties of carrying
on business with email and with weekly telephone conversation, it is
hard to feel that consensus is being reached.  Ralph suggested
instituting a formal process whereby specific proposals are discussed
and voted upon in a certain time frame.  Ron said he felt that we
could institute such procedures, but that he would prefer pushing
discussion by email until the October meeting and discussing such
administrative procedures then.  Ron will make an effort to
focus issues and conversation between now and the October meeting.