Re: [xtm-wg] XTM Requirements

["Jack Park" <jackpark@verticalnet.com>]

Well, the "heavies" are beginning to weigh in not in favor of my proposal.
Consider it withdrawn.  It doesn't stop there, though. Nobody has offered a
suggestion of a way to put them in, save Alexander Sigel in which, instead
of attributes, another kind of relation is suggested.

I remain open to finding a solution to this issue. As Steve Pepper points
out, there remains the need for standardizing presentation semantics. I do
not agree with the emphasis Peter Jones puts in his otherwise well presented
arguments: there, IMHO, will be a lot of stuff done visually in the
knowledge domains in which TMs will serve.  Otherwise, what would be the
point in Benedicte doing a whole dissertation in that field?  I believe that
her work, including the papers she has already published, will be a valuable
contribution to our field of effort.

I am unable to visualize how an XSL stylesheet would help here.  As Peter
correctly points out, there could be thousands of nodes, and, given the high
dimensionality of the space, they may need to be layered such that 2-d views
(maybe 3-d if Benedicte is right) can be presented.  Rather, some kind of
lookup table of suggested coordinates, some kind of relational inclusion
right into the TM itself, or perhaps another approach that completely
escapes me now will work.  For the time being, I have implemented attributes
in the open source editing tool I have built.  The tool will eventually
include Kal's engine such that it will, theoretically speaking, be capable
of performing all the operations we envision on TMs.  The tool can be
changed to suit our evolving specifications.  For now, gaining experience
with a visual editor combined with a text editor should help us resolve the
display issue.

Cheers
Jack

From: Steve Pepper <pepper@ontopia.net>

> At 16:10 30.06.00 +0100, Peter Jones wrote:
> >>Won't having fixed coord numbers in a given map risk
> >>privileging certain views?
>
> I agree.
>
> This discussion is about style sheets. I believe topic
> maps should represent abstractions only. We don't want to
> "contaminate" them with presentational or other processing
> information because, as you say, that would privilege
> certain views.
>
> Having said that, Jack does have a valid point: There are
> situations in which it would be useful to have a standardised
> way of expressing presentation semantics. I can imagine a
> tool which allows the user to set certain display parameters
> interactively (e.g. which icons to use to display which topic
> types, which colours to use for the arcs representing binary
> associations of a certain type, etc.) and then save those
> parameters as preferences in a style sheet. If the syntax
> of that style sheet were standardised it could be interchanged
> between applications. Very useful.
>
> This is what DSSSL was designed to do for SGML, and what XSL
> does for XML.
>
> However, at this point in time, we should concentrate on getting
> an XML representation of the *abstraction* out of the door. We
> can perhaps start gathering requirements related to presentation,
> but I think we have a very long way to go before we're anything
> like ready to create a useful, general solution to this problem.
> Just the fact that my examples introduce two display attributes
> shows that Jack's requirements are only the tip of an iceberg.
>
> Steve
>
> --
> Steve Pepper, Ontopia AS (+47 908 27246)
>
>
> ------------------------------------------------------------------------
> Create professional forms and interactive web pages in less time
> with Mozquito(tm) technology.
> Form the Web today - visit:
> http://click.egroups.com/1/5770/4/_/337252/_/962381295/
> ------------------------------------------------------------------------
>
> To Post a message, send it to:   xtm-wg@eGroups.com
>
> To Unsubscribe, send a blank message to: xtm-wg-unsubscribe@eGroups.com


------------------------------------------------------------------------
Replace complicated scripts using 14 new HTML tags that work in
current browsers.
Form the Web today - visit:
http://click.egroups.com/1/5769/4/_/337252/_/962387851/
------------------------------------------------------------------------

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

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