Compound Locations

In todays call, we discussed compound locations.  I have to express some
scepticism over how much we can reasonably hope to do in EARL.

I have no problem with the simple case:

<Assertion>
  <Location>
     [one or more *instance of* location where the assertion applies]
     [ e.g. here's a bunch of IMGs that each need ALTs ]
  </Location>
</Assertion>

But the compound case is more problematic.  We discussed a few
examples where an assertion might refer to more than one location,
but I think those are rather contrived (e.g. a bunch of unseparated
links is more natural to refer to as a single instance - the surrounding
container - than each for itself).

Most problematic is the fact that *related* links implies a *relationship*.
If we are to embrace the concept of a compound location, we have to
be able to express that relationship.  And that's a whole new can of
worms.  If Tool A (eg APrompt) expresses a location comprising more
than one point in the testsubject, how is Tool B (eg Valet) to infer
anything more meaningful than an _unstructured_ list from that?

There are a couple of simple cases we can perhaps express:
* Range (Start-point + End-point)
* Main+subsidiary locations

But I think anything more complex has to be tool-specific.

I've also just created a trivial case for discussion.  A document with a
validation error that appears in one place but refers back to another.
See how the online validators all deal differently with it:

http://valet.webthing.com/page/validate?url=http%3A%2F%2Fwww.webthing.com%2F%7Enick%2Ffoo.html&parser=OpenSP&resultsMode=traditional&parseMode=sgml
http://validator.w3.org/check?uri=http%3A%2F%2Fwww.webthing.com%2F%7Enick%2Ffoo.html
http://www.htmlhelp.com/cgi-bin/validate.cgi?url=http%3A%2F%2Fwww.webthing.com%2F%7Enick%2Ffoo.html&input=yes

-- 
Nick Kew

Received on Wednesday, 21 September 2005 01:11:02 UTC