[osis-core] Index Syntax

[Patrick Durusau]

Harry,

Harry Plantinga wrote:

>>Not with the <index> element or its attributes (as far as I can tell). 
>>The attributes are level1 - level4 and appear to correspond to what I 
>>have suggested as key1 - key4, since they are not really levels 
>>of indexing.
>>
>
>My understanding was that these _were_ levels in a sense:
>
><index index-name="subject" key1="Job" key2="theodicy" key3="parallel 
> literature" key4="Babylonian sufferer">
>
>would result in an entry in the subject index that looks like this:
>
>Job
>  theodicy
>    parallel literature
>      Babylonian sufferer [id]
>
>If that doesn't correspond to your thinking, we need to talk.
>
Doesn't but that makes the TEI levels make a lot more sense. I do think 
we need to make it an explicit semantic (for those prone to mis-reading 
like myself) that what you have described is what we intend as a result. 
That each key down (up?) the list is a sub-topic of the one that comes 
before it. (key2 is a subtopic of key1, just for absolute clarity)

With that as the understood semantic, I withdraw the suggestion that 
there is an implied see-also relationship, at least of the kind I was 
contemplating.

Ah, and the ID applies only to the last entry in the list of keys.

OK, now that I think we are on the same page, what about this see/see 
also mechanism?

Patrick

>
>----------------------------------------------------
>
>A "see also" line might add this:
>
>      <i>See also</i> Suffering - Theodicy
>
>Could we handle like this:
>
><index index-name="subject" key1="Job" key2="theodicy" key3="parallel 
> literature" key4="Babylonian sufferer" see="Suffering - Theodicy">
>
>which the software could render as
>
>      Babylonian sufferer [id]; <i>see also</i> Suffering - Theodicy
>
>This would add only one attribute, and it would solve the problem. Only
>problem is you have to parse the see attribute value to figure out how
>to link to the appropriate (see also) spot in the index, but MOST see 
>alsos probably only reference a key1, not key2 key3 key4.  And software 
>that doesn't parse the see attribute gets decent fallback behavior -- the
>reader will know where to look.
>
>-Harry
> 
>

-- 
Patrick Durusau
Director of Research and Development
Society of Biblical Literature
pdurusau@emory.edu