[xtm-wg] Deadline for deliverables proposals

["Michel Biezunski" <mb@infoloom.com>]

-------------------------- eGroups Sponsor -------------------------~-~>
Thousands of Great Jobs, One Great Location!
Austinatwork.com. Great Jobs, Great Life!
http://click.egroups.com/1/7847/4/_/337252/_/968075570/
---------------------------------------------------------------------_->

To all members of the XTM authoring group:

Thank to all of you for the excellent work.

After a period of intense work by email, we are
now at the end of the period where each of the subgroup
should have now a list of deliverable proposals and the relations
with the other subgroups.

We have fairly recent documents for the Conceptual Modeling Subgroup
(sent by Daniel Rivers-Moore) and a document I sent on Aug. 28 for the
Interchange Syntax Subgroup. For the Use Cases Subgroup, the document
I am using is part of a mail sent by Patrick Durusau. I am enclosing these
3 documents at the end of this mail.

Next deadline is the London meeting mid-October. This is the date for
delivering the materials we just defined. After this period, remember,
we decided that the subgroups will be dissolved and we'll have to
make everything fit together. Because of the tight deadline, we will not
complete what the subgroups will not have the time to do before, but
rather take what we will have and improve it.

Because there is a risk that the amount of work needed to complete
the work exceeds the time available, it would be good if the members of
the subgroups would put some priority tags on the list of deliverables,
basically 2 levels: 1) critical 2) if time permits.

Reminder: all correspondence should be done using the same address
(xtm-wg@egroups.com  for the time being, and using prefixes to distinguish
the work from different subgroups: [xtm-cms], [xtm-iss], and [xtm-ucs].
This allows each of us to follow the work done by the other groups.

It would also be good that the members of XTM who did not attend the
Montreal
meeting choose one subgroup in which they actively contribute. We have
decided that because of the amount of work to be done, it's not realistic to
be part of more than one subgroup.

Here is what I have as deliverables from the 3 subgroups. If I have
omitted something, please send email.
(It is possible that I didn't get all recent emails).

Michel

==========================================================
1. CONCEPTUAL MODELING SUBGROUP

THE FOLLOWING IS THE CONCEPTUAL MODEL SUBGROUP'S LIST OF DELIVERABLES AND
RELATIONSHIP TO OTHER SUBGROUPS.

DELIVERABLES:

1) Completion of 13250 Conceptual Model
	a) Conceptual model implicit in 13250 (expressed in UML)
	b) Property Set for that conceptual model (expressed in SGML
conforming to HyTime property set DTD, and expressed in UML)
	c) 13250 Syntax Constructs : (expressed as a DTD (the ISO 13250 DTD)
and expressed in UML
	d) Formal mappings between a, b and c (expressed in UML)

2) XTM Conceptual Model
	a) Conceptual Model of XTM (expressed in UML)
	b) Property Set for that conceptual model (expressed in SGML
conforming to HyTime property set DTD, and expressed in UML)
	c) XTM Syntax Constructs (expressed as a DTD (this is the DTD that
the Interchange Syntax Subgroup will produce) and expressed in UML)
	d) Formal mappings between a, b and c, (expressed in UML)
	e) Description of the differences between 2a and 1a
	f) Justification of those differences (based on clarification or
resolution of ambiguities in 13250, or based on Use Cases that suggest
enhancements or simplifications to the 13250 model

3) Description of relationship between XTM conceptual model and RDF
conceptual model.
	a) UML representation of RDF Conceptual Model (expressed in UML)
	b) Fomal mapping between 3a and 2a (expressed in UML)


RELATION TO OTHER SUBGROUPS

1) Interchange Syntax Subgroup
	a) Dependency: To allow us to do 2c, the ISS needs to inform us of
its decisions regarding syntax constructs
	b) Impact: 2d might reveal the need to enhance or modify the
syntactic constructs. They must be formally mappable to the conceptual
model.

2) Use Cases Group
	a) Dependency: Some use cases may dictate simplifications or
enhancements to the conceptual model
	b) Impact: The conceptual model may have power that suggests further
use cases
==========================================
2. INTERCHANGE SYNTAX SUBGROUP

Interchange Syntax Subgroup -- Design decisions and Deliverables
We propose the following list of deliverables:

Topic Map Interchange DTD
Explanatory document
Convention for Public subjects
Cookbook for Topic Map templates

and a list of non-deliverables, which we consider interesting, but that we
decide to exclude from our consideration for the deadline of Dec. 3, 2000.

Deliverable: Topic Map Interchange DTD

We propose to use the DTD formalism (XML form) to express the interchange
syntax. This decision does not exclude other possibilities of expressing the
syntax in future steps (such as XML schemas, Relax, etc.) but since 13250
expresses a DTD. However, considering the short deadline, choosing an XML
DTD will enable us to work in the quickest and efficient manner to hit the
deadline of Dec. 3 for delivering the version 1.0 of the spec.

Design decision: Hyperlink expression
The hyperlinks will be expressed using Xlink Extended Links.

Design decision: Use of namespaces
We will use two namespaces:
- xlink
- topicmap
We will use the mechanism defined by the XML Namespace specification to
declare our own namespace (topicmap).
Use of xlink will be part of the Topic Map Interchange DTD

Design decision: Addressing
We propose to limit addressing to URIs (including XPath).
Use of URIs will be part of the Topic Map Interchange DTD.

Design decision: Packaging topic maps
We should provide a mechanism for declaring metadata for the topic map
itself. It is probably something similar, (or identical?) to the addthms
attribute. We might want to use or refer to the Dublin Core to provide this
information.

Design decision: ISO 13250 conformance
The DTD will be interpretable as having inherited from the ISO DTD.
If it appears that there are things in the XTM spec which have been added
(because they were not provided by ISO 13250), we would propose to the ISO
JTC1/SC34/WG3 responsible for dealing with the Topic Maps standard to add
the features that will maintain compatibility between the ISO standard and
the XTM
spec.

Design decision: Miscellaneous
By studying the various DTDs that have been currently proposed, other issues
will be raised and will have to be resolved.
Here is a provisional list of such issues:
- provision for merging topic maps (proposed mergemap attribute)
- notion of DTD integration support
- checking whether removing many HyTime features (bosspec, maxbos, boslevel,
etc.) leaves us in a safe position.
- do we need two mechanisms for addthms (element and attribute)?
proposed supertypes attribute on topics. Do we need it?
- do we need the linktype attribute in the xtm spec? It seems that this is
only here because it's needed by HyTime.
- proposed addition of a notion of "name service"
- proposed addition of considering topic associations as topics. Do we need
it ?
Should the id attributes on topic, assoc, etc., have a topicmap namespace
added?
etc. [List not closed yet]
- Available DTDs:

Deliverable: Explanatory document

The explanatory document accompanying and documenting the DTD should be
simple, short, and should contain examples.
We should start writing this document as soon as possible, to have time to
improve it until delivery time. Many parts of it are not going to be
altered, and therefore much writing can be made even if all design decisions
are not finalized.
Examples could be provided from the Use Cases Group.

Deliverable: Convention for Public subjects

We propose to deliver a convention for using public subjects. Basically,
public subjects will be part of topic maps that will be available on the
Web.
There should be an explanatory document that makes it easy for every topic
map user to declare public subjects.
Note: We should discuss whether we should allow the identity attribute to
point to several public subjects rather than one.

Deliverable: Cookbook for Topic map templates
We should first clarify what we mean by topic map templates.
If we agree to consider templates as topic maps with predefined semantics,
then we should provide a cookbook with examples of how to create and use
topic map templates.

Non-deliverables

Accompanying white papers

What follows is a list of proposed documents that may represent high value
added pieces of information. Instead of integrating as parts of the spec as
deliverables, we propose instead to include them in a list of things to add
to deliverables, and being submitted for further thinking, testing, and/or
future insertion into the spec.
XSLT transformation for expressing topic maps into SVG (Scalable Vector
Graphics) for display

Items deferred to the Authoring Group

We propose to not take into account the following list of items, because we
consider it should be the responsability of the Authoring Group as a whole
to take care of those.

Interoperability issues

Interoperability issues should be resolved by adding the conclusions coming
from the Conceptual Modeling Subgroup to the ones coming from the
Interchange Syntax Subgroup.

Topic Maps / RDF complementarity

The way Topic Maps and RDF will be mapped together will depend on the summit
being planned between RDF and Topic Maps experts. We prefer not anticipate
an answer before having studied and resolved most of the issues. We
anticipate that the results of these discussions should not directly impact
the work being proposed by the Interchange Syntax Subgroup.
==========================================
3. USE CASE SUBGROUP


1. A list of use cases sorted by a typology created by the UCS
2. A representative sample of the use cases explored in greater detail
3. A list of capabilities for XTM derived from the use cases
4. As an appendix to the deliverable, a sorted list of "raw" responses in
XML


To Post a message, send it to:   xtm-wg@eGroups.com

To Unsubscribe, send a blank message to: xtm-wg-unsubscribe@eGroups.com