[sc34wg3] XTM 1.1 issues

[Murray Altheim]

Lars Heuer wrote:
> Hi Murray,
> 
> [...]
> 
>>Since you guys are in charge of the process and feel it within
>>your rights and duty to make such substantive changes to XTM,
>>such that to my eyes at least it doesn't look or act remotely 
>>like the syntax of the current markup language, why not call 
>>it something else? You've plainly got a much different agenda
> 
> [...]
> 
> Well, like everything else, XTM evolves.
> From my point of view it would have been a mistake to change the
> namespace but not to take the chance to clean up the syntax.
> Either the TM community has to keep the XTM 1.0 namespace or take the
> chance to make XTM consistent to TMDM terminology.
> I think breaking backwards compatibility implies the chance to make
> XTM ready for the future.
> Because of the backwards compatiblity break I am much in favour to
> rename XTM 1.1 into XTM 2.0, but it is just a number.

The question is not namespaces or version numbers, but one of
semantics. If the fundamental vocabulary of a language has
changed, it's no longer the same language, and therefore does
not have the same name. They started calling it "English" at
some point -- "XTM" is no longer appropriate to a markup language
based on the TMDM given the TMDM doesn't use the same vocabulary
as XTM, i.e., ISO 13250:2000. It's something new, and needs a
new name.

> Anyway, I think it is a failure to keep the mergeMap element in the
> XTM 1.1. This might belong to the upcoming CTM syntax which is
> designed for _authoring_ topic maps, while XTM should be designed for
> _interchanging_ topic maps (XTM should be readable by humans, though).

Perhaps you have a limited view of interchange. I think of being
able to interchange either an entire vocabulary, or, when we talk
about modularization, a subset of a vocabulary. When looking at
large, monolithic vocabularies (such as the Cyc ontology), very
few people would want to exchange the entire map; they want to be
able to pick and choose amongst components, and <mergeMap> provides
precisely the functionality necessary. At an application level, it
permits either a human or program to decide whether or not to
traverse the link bringing in the module. "Interchange" in this
sense is the interchange of one or more components, which has zero
to do with authoring.

And BTW, while XTM had "human legibility" as a design goal (i.e.,
we weren't creating a binary format), it never had terseness as
one (see goals 6 and 10 of XTM [1], following the lead of XML [2]).
XTM is readable by humans, but nobody except geeks should ever have
to look at XTM or XML markup.

> Due to the lack of the specialized Topic Maps authoring language
> it seems that mergeMap has to stay. IMO this mergeMap "feature" is
> very artificial. If you receive a PDF and the Acrobat Reader tells
> you, "Well, I've to fetch the page 'X' from location 'A' and page 'Y'
> from location 'B'" it would be very strange. The PDF is a complete
> document and a XTM should be a complete topic map.
> The sender should be responsible to keep everything in place and send
> the topic map (or fragments of a topic map or a number of merged topic
> maps) in one file.

XTM is not PDF. Comparison between the two seems pretty pointless,
as they're not even near each other's application domains. As for
XTM having some requirement of being a complete Topic Map, the
CTM work covers that, which you seem to be confusing with XTM.

>>I think it's time to call your duck a dog, if it barks instead
>>of quacks.
> 
> :)
> It's still a topic map interchange syntax.

Yes, it still is a Topic Map interchange syntax. But it sounds
like a new Topic Map interchange syntax if the fundamentals of
the language have changed, and they have. XTM wasn't written
in mind of harmonisation with the TMDM (obviously), and if the
ISO committee has that as a goal, they then seem to be creating
a new language as an interchange syntax for the TMDM. If that's
new, fine, but it's not "XTM" anymore.

Murray

[1] http://www.topicmaps.org/xtm/1.0/index.html#goals
[2] http://www.w3.org/TR/REC-xml/#sec-origin-goals
......................................................................
Murray Altheim                          http://www.altheim.com/murray/
Strategic Services Development Manager
The Open University Library and Learning Resources Centre
The Open University, Milton Keynes, Bucks, MK7 6AA, UK               .

    As late as 1855, New York newspapers reported that Presbyterian,
    Baptist and Methodist churches were closed on Dec. 25 because
    "they do not accept the day as a Holy One." On the eve of the
    Civil War, Christmas was recognized in just 18 states.
    http://www.nytimes.com/2005/12/04/opinion/04sun3.html