[sc34wg3] occurrence - basename fuzzy border

Murray Altheim sc34wg3@isotopicmaps.org
Fri, 21 Feb 2003 14:57:22 +0000

Steve Pepper wrote:
> At 11:13 21.02.2003 +0000, Murray Altheim wrote:
>> This smacks so strongly of featuritis that I can't help but do the
>> common thing of accusing you of being an application vendor.
> In that case I refuse to engage in any discussion with you.

A bit childish a response, Steve. My point was, while obviously
we are all here trying to develop a technology, your suggestion,
similar the one Michel made during our TopicMaps.Org days on
adding base names to occurrences, seem based on a vendor's ideas
or needs for their application. In the case of Michel's, we in
the end didn't add the feature for the reasons I cited in my last
message, and applications have been able to come up with other
ways of doing it.

I had hoped to engage you in a discussion of why adding a
presentational structure to occurrences is in this case *not*
simply a case of vendor featuritis, rather than shutting down
the conversation entirely, which is obviously always an option
but hardly productive. IOW, I wasn't making an attack on you
as a vendor, I was saying your proposal sounds like one based
on the needs of a vendor rather than architectural considerations.
We always must strike a compromise between these -- the one
usually called "80/20".

So I'm not saying you're doing that, I'm asking to hear why your
argument is *not* that, since so far I've only heard that it
would improve your application's ability to sort and display
topics (I say "your" since it seems to me there a number of
ways to approach this problem that don't require changes to
the current spec, and obviously current topic map applications
are currently operating without this feature).

We've seen what happens to specifications when basing the
justification for new features is based on the "need" for new
features, as in HTML/XHTML: there's an initial good idea, and
then over time it bloats out with further "good ideas", features
that differentiate products, that may improve a GUI but make
implementations more complex, raise the bar of entry for newcomers,
and often destroy interoperability and interchange. A "fully
compliant" web browser nowadays is near an impossibility to create,
a far cry from Mosaic. [I think the W3C has long since taken XML
past that point; another discussion entirely...] Ed Nixon under
a different thread ("Mapping files") is making *exactly* this point.

Adding the recursive <variant> to <occurrence> would add significant
complexity to occurrences, as we've already heard people complain
about them in <baseName>. At this point in TM's history, while we're
all trying to increase its "market penetration", the last thing we
should be doing is unnecessarily making the technology more complex.
Someone had posted a note on their office door here that's now gone
that said in effect, "any technology that takes longer than a week
for a web developer to figure out is doomed." Like it or not, it's
probably true.


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.