Meeting Minutes 7/06/2004 -- DITA Technical Committee

["France Baril" <France.Baril@ixiasoft.com>]

Title: Meeting Minutes 7/06/2004 -- DITA Technical Committee

Agenda:

1. Roll call

Voting Members: Indi Liepa, Eliot Kimber, Robin Cover, Mike Wethington, David Schell, Paul Grosso, Michael Priestley, France Baril (scribe), Don Day

Observers: Christopher Kreiler, Deborah Lepeyre, Erik Hennum, Eric Sirois

No quorum

2. Review/approve minutes from 29 June
http://lists.oasis-open.org/archives/dita/200406/msg00070.html

Not approved, because we didn’t have quorum.

3. Progress on issues currently in discussion:
- Conref and XInclude? (Plan for now: explain & contrast with SGML conref)

TODO : Michael to look if contrasts with SGML.
TODO : Eric Hennum to wrap up his thoughts and put them on the list.

Eliot talked about his experience with Xiruss-t (project site is
http://xiruss-t.sourceforge.net).
He wrote a paper for XML Europe where he talks about basic
challenges in Content Management functionality for XML based tech
doc. He says there is no complete solution at this point that are
completely acceptable. He makes a demonstration on managing
hyperdocuments, talks about Xinclude and conref. The code is
available on sourceforge, not for production use, but educational.
The papers demonstrates the basic fact that most of the pain is at
the boundary where the import/export process takes place. He
developed a generic import framework to work with compound
documents. People should be able to download it from the sourceforge website.
Don asked if it is possible to make a bin package for downloading
Eliot to look at license issue and see what he can do. He doesn’t
want people to use it for production.
Discussions about: would we want to provide DITA toolkit as open source?
The Issue of licensing with open source came out.
The idea of having an IBM statement of self interest in making DITA
public to answer questions in the industry about wanting to force
this DITA std on others.
Michael asked others to bring up the business value of DITA for them
or their organization. Michael to post this question on the list to
invite others to comment on the issue.
Indi has information for business case that she can share.

- Namespace for DITA?
Eliot has demonstrated that he can add namespaces into schema and
operate on it. He created 2 ns, one for maps and one for topics. He
had to modify the xslt processor so that all maps statement referred
to the class attribute and made sure all templates matched on *
instead of on element names. XSLTs were already using * matches at
98%. There is nothing more to it.

Corresponding change on document: add the namespace declaration.
Because dita namespace become root namespace so no implication for
other changes.

Paul Grosso: supporting namespaces does not bring extra problems to vendors.

Why need to have namespaces?
Eliot: Always was a problem in SGML. There was
no way to say unambiguiously this document relates to this set of
rules. XML still had the same problem. You could guess with doctype
declaration. With namespaces and schema we have a way to make
unambiguous info. Schema spec gives syntactically distinct mechanism
to map to a namespace. You have to be able to do that to develop
generic processes.
Don other benefit: gives DITA content possibility to compound with
other document types. Michael: also with namespaces you can use
elements from 2 different specializations for companies that use
the same element name in different ways within the same tree.

Issue: we need to know how many namespaces we are going to need. We
need to determine what names we need.

Each basic topic type from DITA will have its own namespace.
Specialization from other companies will be develop in these
companies namespaces.

- Which version of CALS table model?

Not discussed.

- Are the current domains adequate?

Not discussed.


4. Continue working on outlines for the initial specifications:
- Guidance from Michael and JoAnn? Latest outline still is this:
http://lists.oasis-open.org/archives/dita/200406/msg00035.html
- Report from Eliot on substructure conventions? Specs we should emulate?
- DITA Map/task map for the spec documents?

Not discussed.

5. AOB?
Nothing new.

ACTIONS:
Don encouraged using the mailing list this week to drive progress
on the discussion issues, particularly to understand the full set
of namespaces we should declare for DITA.
Action: Eliot to wrap up the "set of namespaces" as a proposal for the
TC to approve next week.
Action: All, contribute to Michael Priestley's posted question on
"What is the business value of DITA?"
http://lists.oasis-open.org/archives/dita/200407/msg00003.html

France Baril
Documentation architect/Architecte documentaire
I X I A S O F T

+1 514 279-4942
france.baril@ixiasoft.com
[   www.ixiasoft.com   ]