[osis-core] osisRef (finally)

[Patrick Durusau]

Troy,

Troy A. Griffitts wrote:

>> Come to think of it, we were going to identify referenced works in the
>> header,
>> right?  So we'd never use Bible.KJV.Matt.1.2 -- instead we'd have
>> <work id="KJV" ...> in the header, and we'd use
>>
>> KJV.Matt.1.2.
>>
>> Is that correct?
>
>
> In an age long ago, we decided to specify a default reference work and 
> schema in the header and allow overriding with something like:
>
> <reference work="Bible.NIV" refID="John.1.1">
>
> If the 'work' attribute was not present, then the default from the 
> header would be assumed.

I think I am probably guilty of this more than anyone but it would be 
helpful if we all used the most recent OSIS schema element and attribute 
names (or noted "proposed" somewhere in the post) when we are discussing 
constructs. I tend to focus on the syntax and strange usages tend to 
throw me off the intended message.)

To restate Troy's example:

<verse osisID="John.1.1"> the osisID defaults to whatever work attribute 
is declared in osisText.

Overriding happens in the following case:

<verse work="Bible.NIV" osisID="John.1.1">

Both cite and outCite should work differently with osisRef. The cite 
should default to inside this document and hence subject to the default 
declaration of work in the header. In other words, it is pointing to 
another location in this document instance.

outCite, on the other hand, should (don't think it does at the moment or 
at least is unclear) always require that the work be specified as part 
of the pointer.

>
> I would like to hear some discussion about the:
>
> "Bible.KJV"
>
> text.  What exactly do people (you people) mean when they say this?
>
> 'Bible' as opposed to???  Commentary? Lexicon? DailyDevotional?  Are 
> we really wanting to classify works other than Bibles at this level?
>
> 'KJV' meaning versification scheme? Actual work? If so, what edition?

Hmmm, I suspect that we need to unpack or at least make explicit what is 
meant by our Bible.KJV reference.

At one level, it is a reference system that has been followed by any 
number of other translations, but differing ones exist (Jewish, Korean 
by pages, others).

This touches on Harry's question yesterday about mapping and reference 
systems and I think I was assuming a standard system that was not being 
explicitly stated.

I see edition, translation, etc. as separate issues (although important 
ones) from the notion of references.

Thus, I should be able to say (in some syntax): Versification - 
Standard, Translation/Edition (to specify a particular translation), etc.

Perhaps we should return to Harry's question of having a standard 
versification (includes all apocrypha, etc.) since the greater includes 
the lesser and simply say that is the standard versification by default 
and that all others map to it? (Without attempting the mapping at this 
late date.) I could probably cast together a pattern of "or" statements 
that would validate any standard bible reference. (In other words, 
unlike the regex that Todd and I labored so long on, it would have range 
operators for the individual books and verses. Anyone who wants to use 
an alternative versification, such as the Jewish one, could subtype that 
standard and extend the range statements.

Therefore, our reference syntax would be: 
Bible.Standard.(translation/edition)?.(book).(chapter)?.(verse)? with 
the Bible.Standard.(translation/edition)? being held in "work"?

(After all, the chances of someone finding the incorrect reference when 
someone calls out: John 5:39 are fairly small, no matter what 
"translation/edition" they are using. I think we may be trying to solve 
the odd case (primarily the Psalms but a few other places) elegantly 
when there may be no elegant solution. I suggest "Let'em  map" and be 
done with it. ;-) (Troy, you should appreciate the citation.)

Would it be sufficient to simply refer to the Oxford Study Bible (just 
pick some edition) as the standard reference system for bible citations? 
And perhaps offer a mapping mechanism for those who wish to use some 
other system? 

In terms of ease of use I think we should simply adopt Harry's 
suggestion of declaring a standard system (like the Oxford bible) and 
simply say that if you are different, for whatever reason, you have to 
map back to that system. (Where it falls short is when other systems 
declare headers as verses but that could be verse "0" in terms of 
mapping to the standard system.)

This is one of the few really hard issues remaining so I would like to 
see lively discussion and some resolution on this particular one. I can 
clean up whatever syntax (with help from you guys) so let's hear some 
proposals.

Patrick


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