[osis-core] lang inheritance....

[Patrick Durusau]

Steve,

Did we ever solve this one?

Not really sure how much of a problem this will be with milestones. 
Should be fairly rare (Troy? Todd?) where you have overlapping 
quotations that are in different languages you are marking with xml:lang.

Rather than prohibit it, since it could well be useful on Troy milestone 
elements, suggest we say that the meaning of xml:lang in the case of 
overlapping milestones is undefined.

If use of xml:lang is a real issue for a project, they can always fall 
back onto <seg> for those critical cases.

Inelegant I know but I think that will be the best we can do.

Comments?

Hope everyone is having a great day!

Patrick


Steven J. DeRose wrote:
> Sorry I've been pretty absent on the list. But let me throw a small 
> wrench into the works to think about:
> 
> Writing this indexer, I'm trying to get it to track xml:lang values. 
> This works trivially for the normal element tree (well, after I add a 
> couple lines of code it will). *But* -- how, exactly, is the language 
> value affected by Trojan milestone elements?
> 
> Since XML defines the inheritance of xml:lang in terms of element 
> containment, the value doesn't inherit onto the content of a 
> milestone-delimited element. This is a potential problem. We could 
> easily state that OSIS applications are also supposed to inherit lang 
> through milestone-delimited elements, but I don't see any way to specify 
> the real meaning of that unambiguously. If you've got several phenomena 
> overlapping, just whose "lang" is in effect at any given time?
> 
> The Occam's razor solution (or perhaps, less virtuously, the Gordium 
> knot solution), is to prohibit xml:lang on milestoned elements. Not so 
> bad for chapter/verse, but could be real bad for multiply nested 
> quotations....
> 
> S


-- 
Patrick Durusau
Director of Research and Development
Society of Biblical Literature
Patrick.Durusau@sbl-site.org
Chair, V1 - Text Processing: Office and Publishing Systems Interface
Co-Editor, ISO 13250, Topic Maps -- Reference Model

Topic Maps: Human, not artificial, intelligence at work!