[sc34wg3] Almost arbitrary markup in resourceData

Murray Altheim sc34wg3@isotopicmaps.org
Wed, 12 Nov 2003 19:44:10 +0000

Lars Marius Garshol wrote:
> * Patrick Durusau
> |
> | Hmmm, depends on what you mean by XML data doesn't it? I assume that
> | I would want to interpret the markup that I find, but if all I have
> | is the markup, I am guessing about its significance. 
> That is true. So long as the topic map is only used within a single
> organization or project this is no problem, but once we allow topic
> map fragments to be interchanged in the open we can be pretty certain
> that sooner or later someone will receive an internal occurrence
> containing XML markup the recipient does not understand and does not
> know how to display/process.
> The question is whether we should try to handle this in any way, or
> whether we should just say "beware of sending such stuff around", and
> "don't expect that you'll always understand the markup in an
> occurrence". At the moment I'm leaning towards not making any attempt
> to handle this.

Why not simply consider that any proprietary markup sent around doesn't
pass the sniff test for interchange, and leave it at that? That the
interchange format for Topic Maps in XTM doesn't allow proprietary
markup, full stop? If you want to interchange with other people, and
you want to guarantee interoperability (which is the whole point of
an interchange syntax), you lose the proprietary stuff when you export
into the interchange syntax. Yes, I repeat, you lose the ability to
interchange proprietary "semantics" when you send them to other
people. That only makes sense.

> I think what we will probably do in the Omnigator is to pass through
> markup in the XHTML namespace when rendering, but to strip out all
> markup in other namespaces. Or, we could just display the markup in
> escaped form. Or we could strip all markup. Or we could pass through
> all markup and leave it to the browser to figure out whether it can
> make sense of it.
> | Without it being in XTM, I would say it is an XML issue. Not sure
> | that changes if it is inside XTM.
> I would tend to agree.

I'm not sure where this division between what is an XTM and what is
an XML issue arises. Sounds like hand-waving. If by "XML issue" you
mean "it's not our problem" that's not true. The decision to allow
"XML issues" to infect XTM is a conscious one, within our control.
We were compliant with XML Namespaces with XTM 1.0, but we
prohibited things that were allowed by it, such as arbitrary markup.


Murray Altheim                         http://kmi.open.ac.uk/people/murray/
Knowledge Media Institute
The Open University, Milton Keynes, Bucks, MK7 6AA, UK                    .

   Entitled Continuing Collateral Damage: the health and environmental
   costs of war on Iraq, the report estimates that between 22,000 and
   55,000 people - mainly Iraqi soldiers and civilians - died as a direct
   result of the war.

   Entitled Continuing Collateral Damage? ...a euphemism for BushCo.