[sc34wg3] clarifications in PMTM4 1.0.2

Steven R. Newcomb sc34wg3@isotopicmaps.org
Wed, 25 Jul 2001 15:27:05 -0500

PMTM4 (http://www.topicmaps.net/pmtm4.htm), the
Biezunski/Newcomb Processing Model for XML Topic Maps,
is today at revision 1.0.2, the first revision since
March 15, 2001.  Implementers have asked some questions
that prompted Michel Biezunski and Steve Newcomb (us)
to incorporate some clarifications in the text.  For
those who are interested to see exactly what has been
added to the text, the new text appears in red.

We know there are at least four implementers of PMTM4,
and at least three different implementations are
working and/or are under construction.  (There may, of
course, be more that we don't know about yet.)  Those
who are implementing PMTM4 say they have found the
model very satisfactory to work with.  No
inconsistencies, inelegances, unworkabilities, or other
problems with the model itself have been reported to us
by any persons who have actually committed themselves
to implementing it.

However, there have been many requests for
clarifications from the implementers; we get calls
almost every day, and sometimes several calls in a
single day.  This is not surprising; we're all
pioneering.  So far, we have been able to help the
implementers to understand exactly what PMTM4 is
intended to mean.  As this latest revision of PMTM4
demonstrates, we're continuing to work to make the
PMTM4 spec as clear as possible.  We need it to be
very, very clear because our ability to spend time on
the phone with implementers and other researchers is
very, very limited.

The Extreme Markup Languages Conference in Montreal
(August 12-17, www.extrememarkup.com) will have several
presentations that are based on and/or influenced by

All this leads me to believe that consensus is building
around PMTM4, and that PMTM4 "has legs".

PMTM4 establishes a context within which the usefulness
of existing approaches can be fully appreciated,
compared, discussed, and improved.  PMTM4 does not
directly constrain applications, implementations, data
models, APIs, interchange syntaxes, query languages,
constraint languages, or anything else.  PMTM4 merely
proposes to establish a framework within which the
*meaning* of the XTM syntax can be rigorously defined,
and then it proposes a definition of the meaning of the
XTM syntax, using that framework.  Then, the same
framework can be used to define the meaning of any
syntax for interchanging the kinds of information that
topic maps syntaxes can be used to interchange.

We hope PMTM4 will *influence* the development of many
implementations, data models, APIs, interchange
syntaxes, query languages, constraint languages, and
much else.  We hope PMTM4 will hasten and facilitate
the development of all of these things, and of
standards for all these things, by providing a frame of
reference within which the developers of all of these
things will be able to communicate precisely and easily
with one another, and with the community of
implementers and users.

Michel Biezunski and Steve Newcomb

Michel Biezunski, InfoLoom
Tel +33 1 44 59 84 29 Cell +33 6 03 99 25 29
Email: mb@infoloom.com  Web: www.infoloom.com

Steven R. Newcomb, Consultant
voice: +1 972 359 8160
fax:   +1 972 359 0270
1527 Northaven Drive
Allen, Texas 75002-1648 USA