RE: [huml-comment] Interoperable Language Standard (ILS)

["Bullard, Claude L (Len)" <clbullar@ingr.com>]

I agree with that.  Out of line 
or as once called, independent linking is an 
old and much neglected idea.  I hope that 
it is the path that Huml developers take 
for annotation systems.

Code lists are very effective.  We use them 
a lot in our systems for the customizable 
selections, because a table of two level 
codes is easy to implement, and easy for 
the local user to extend.

len

-----Original Message-----
From: Emmanuil Batsis (Manos) [mailto:mbatsis@humanmarkup.org]

Bullard, Claude L (Len) wrote:
> Note in the sample XML that they are going down 
> the path of enumerations instead of types.  They 
> are essentially building code lists, aka, controlled 
> vocabularies for annotating a text.

True, but those are just mental notes, very similar to ours back in the 
yahoogroups list (emotions etc).

I bet they will end up using metadata in RDF (section three) linked 
using something like the Annotea technology (XPointer). I have this 
belief mostly because XPointer allows annotation without actually 
altering the target document, while it is not affected by the nesting 
restrictions of XML (you can point to ranges etc). For inline 
annotation, the only efficient mechanism would be something like LMNL, 
but it wouldn't enjoy the support XML does and I still preffer 
seperation of things(document VS annotations).

Promoting the enumeration list items to actual classes would increase 
the expressive power of the annotations. But then again, certain 
use-cases would benefit from the simplicity of simple enumerations as 
the most efficient solution for their requierments.

Manos