[topicmapmail] Regarding Facets

[Murray Altheim]

Guy Murphy wrote:
> Hiyas.
> 
> This is where we approach what I believe to be the crux of a schism (and by
> that I don't mean to imply that you're sat on the wrong side of it).... I
> still cling to the tenant that XML mark-up should be whenever possible human
> readable, whether the intent is for the XML to be edited by human or not.
> There's reasons why this was a stated goal of XML which has a lot to do with
> whether developers can understand the data they're working with, or whether
> companies have to pay out fat cheques to domain-expert contractors when it
> all goes pear shaped.

Well, during my career in markup I've always advocated markup simplicity,
which is one of the reasons I really hate xml namespaces, have a tough time
with RDF and XML Schema, etc.  They're "aesthetically" ugly to me. I don't
think the approach Steve and I have advocated is complicated or difficult
to understand, it's just verbose.

> Either way the reply of "you want to read the mark-up?" isn't likely to
> placate the developer who asks "how do I express a topic property?".

I consider XTM markup readable. Verbose, but readable*. My remark in my
last message about a <property> element being limiting should be considered
perhaps more seriously, though. XTM can in theory be used for damned near
anything, and property value pairs *could* have been implemented just as
could a ton of other domain-specific features. But doing it as a *simple*
element-attribute thing (ala what I see commonly in XML preference files)
might seem to developers simple, but it solves only one very small
problem, and does it in one very limited way. Using the more verbose but
more flexible <occurrence> route leaves open a lot more possibilities in
expressing aspects of facets that <property> could not touch absent adding
a bunch more domain-specific attributes to express further attributes of
the facet, which is itself an attribute.

> [quote]
>       "In Las Vegas Mr Gates also demonstrated a prototype
>        fridge magnet which can be programmed to receive traffic
>        reports, sports results and advertisements from local
>        restaurants using the same FM signal as the wristwatch."
>                                    -- The Guardian, 10 Jan 2003.
> [/quote]
> 
> Which is a good illustration of what happens when one tries to cover every
> eventuality of function rather than concentrate on the simplest production
> catering to major usage... stick to the fridge.... are you not advocating
> your own super fridge magnet?

I'd challenge you to demonstrate what the "major usage" of XTM is. I'm
certain there are just about as many approaches and uses currently as
there are projects. Certainly within my own work, I've seen quite a variance
of my own use of XTM markup.

> [quote]
> your proposal is quite limiting in its simplicity.
> [/quote]
> 
> I maintain that your actuality is more limiting in its complexity. =)

I honestly don't think it's so complex as you make it out to be. It's
simply a link to another element that has a child used as either a
reference or a CDATA value, which can be characterized by use of
other topics as a facet. Yes, it *can* be given a name, a type,
can be scoped by other topics. Those are *options* I'd hate to
eliminate. You don't have to use them.

I think we're on the same page philosophically, we just see simplicity
in different ways.

Murray

* as you may know, I've got a fair amount of experience with markup,
   but I still find the ISO 13250 meta-DTDs very difficult to read,
   even though they're extremely concise and to the point.
......................................................................
Murray Altheim                  <http://kmi.open.ac.uk/people/murray/>
Knowledge Media Institute
The Open University, Milton Keynes, Bucks, MK7 6AA, UK

     "In Las Vegas Mr Gates also demonstrated a prototype
      fridge magnet which can be programmed to receive traffic
      reports, sports results and advertisements from local
      restaurants using the same FM signal as the wristwatch."
                                  -- The Guardian, 10 Jan 2003.