[sc34wg3] Use cases for occurrence variants

[Nikita Ogievetsky]

Kal Ahmed wrote:


> Steve Pepper wrote:
> 
> > At 20:16 04.03.2003 -0800, Nikita Ogievetsky wrote:
> >
> >> To get occurrence variants discussion down to Earth
> >> I suggest that we try to summarize use cases.
> >>
> >> Here are those that I know of or have seen mentioned so far:
> >>
> >> 1. Thumbnail(s) of an image.
> >> 2. Description of an image (displayability).
> >> 3. Alt text (displayability).
> >> 4. Sorting occurrences (sortability).
> >> 5. Alternate address depending on device context, protocol, etc.
> >> 6. Model inconsistency between base names and occurrences.
> >>
> >> Any others?
> >
> >
> > I think that pretty much covers it.
> >
> > #5. is perhaps the most important, so I would put it at the top and
> > call it "Alternate renditions" rather than "Alternate address". It
> > would include alternate notations, resolutions, sizes, etc. For
> > example, in the XML Papers application, the PDF and HTML versions of
> > a paper might be variant occurrences of the XML version.
> >
> > As someone who has been playing devil's advocate and didn't really
> > want variant occurrences, your list scares me. What it tells me is
> > that there is a whole range of very useful and necessary functionality
> > we are excluded from using unless we are prepared to use associations
> > instead of occurrences. (And even then we don't have an application-
> > independent way of capturing the relationship between variant forms
> > of a resource.)
> >
> > I would almost go as far as to say that without variants, occurrences
> > are virtually useless. It surprises me that more people haven't thrown
> > up their hands in despair. Maybe no-one is actually using topic maps?
> 
> Maybe they have just found other ways of expressing their models.
> 
> 1 - 4 could all be handled by reification of the occurrence resource.

Do you mean: by reification of occurrence assertion?
 
> 5 could be handled either by reification or by making use of facilities 
> in the HTTP protocol for user agent identification.
> 

Kal, nobody argues with that.
Same way occurrences could be handled by associations.
But we use occurrences and name variants and other syntactical constructs
because they are convenient and make our documents more readable.
That is all.

--Nikita.

Nikita Ogievetsky, nogievet@cogx.com;
Cogitech Inc.        http://www.cogx.com   
Topic Maps Tutorials and Consulting. 
phone:  1 (917) 406 - 8734