Re: [xtm-wg] A challenge on "the graph"

[RA Poell <poell@fel.tno.nl>]

Bernard Vatant wrote:
> The model they propose is ridiculously simple. Just a translation of nodes
> and arcs into ... vertices and edges :)
> ...
> ... In fact, it's as simple and obvious as Topic Maps looked to me ...
> before I began to look into the syntax :o)

I fully agree with you.
The things we are handling in a TM (as well in RDF and in Notion System)
are pieces of information that are form together an information network.
IMO there is no better way to "visuallize" a network than a graph. There
are multiple ways you can store (make persistant) that information:
relational databases, OO-databases, XML files (and in the case of TM XTM
files),...

The querying mechanisme you are going to use in a particular application
depends on how the information can be accessed. The example of Bernard
(graph) needs (and allows) completely different mechanismes then the
querying of a relational database. A particular problem I am actually
dealing with is the querying on distributed data (without global
indexing).

I would like to resume the situation like this:
- we have pieces of info: nodes, relationship between these nodes and
things we want to say about the nodes and relationships 
- TM is one way of defining what are the things in the network (topics,
associations,...)
- XTM is a formalisation of how you can express a TM in a written form
(allowing storage in an .xtm, communication of these files,...)
- the usage of the information in the network, like querying and
inferencing, depends on how you can access the data (application
specific) (directly in an xtm file, a parsed xtm in memory, a graph in
memory, distributed amoung different servers,...)
- the representation to the user can be various (with the same data):
2D-graph, hyperlinked text (these exists) and new forms like 3-D and 4-D
representations (still some work to be done). The "user" here might be
an implementor, a modeller, the final content user, or whatever so.

Each of these "layers" can impose its own rules of how and what is to be
done building on the rules of the previous "layers". Some of these
"layers" will impose meaning or exploitation rules (sort names in XTM,
in NS (application) all text must be associated with a language, the use
of a relational database can impose data integrity rules, etc)
Further implications, this time on what content we are going to put in,
derive from what you want to do with it: authorization rights, validity
attributes and NameIt. The work Cyc, DAML+OIL and others working on
ontologies and my visions on how we might be able to realize automatic
maintenance on the data do have a lot of implications on this level.

IMHO the actual discussion (altough very interesting) mixes some these
"layers" which makes communication a bit difficult.

Friendly greetings

Ronald Poell

-- 
Ronald Poell
Consultant Knowledge Management
Netherlands Organization for Applied Scientific Research (TNO)

Oude Waalsdorperweg 63
P.O. Box 96864
2509 JG The Hague
The Netherlands

T +31 70 374 02 00
F +31 70 374 06 52

http://www.tno.nl

email poell@fel.tno.nl

------------------------ Yahoo! Groups Sponsor ---------------------~-~>
Secure your servers with 128-bit SSL encryption!
Grab your copy of VeriSign's FREE Guide,
"Securing Your Web site for Business." Get it now!
http://us.click.yahoo.com/4cW4jC/e.WCAA/bT0EAA/2n6YlB/TM
---------------------------------------------------------------------_->

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

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

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/