[osis-core] naming mechanism for OSIS identifiers.

[Todd Tillinghast]

Just for Bibles:

The following period separated values:
1) Bible
2) language (either the two letter ISO code in lower case or the three
letter Ethnologue code in upper case)
3) common translation name (use the three or four letter translation
name adopted by CPSG if applicable)  (KJV, NIV, etc...)
4) year of publication


Examples:
Bible.es.RVR.1960 
Bible.en.KJV.1611
Bible.en.KJV.1995.Zondervan
Bible.en.CEV
Bible.en.CEV.1995
Bible.en.CEV.1995.ABS
Bible.JVN.SIL.1990
Bible.en.NIV.1975
Bible.es.NIV.1980
Bible.es.NIV.1980.IBS

All identifiers should have at least one value after the language code.

End of first screen :)

Issues:
1) For languages where there are multiple translations there are common
translation identifiers (NIV, KJV, CEV, etc...), but for minority
languages the Bibles are known as the Bible in that language.  The
problem is there is no value to fill the third position.  Propose we use
the translating organization if no other common value is available.
(SIL or UBS should work in most cases, but we should get input from Kees
and others.)

2) What do we do about cases where there is more than one translation in
a single year? (198003 for the third month of 1980?)

3) Should we enumerate standardized organization names (IBS, ABS, WBTC,
PBT, etc...).  The Forum of Bible Agencies has standard two, three, or
four letter codes for its member agencies.  Does the UBS have
standardized values for its member Bible agencies?

4) An alternative is to ALWAYS put organization name in the third
position (Bible.en.ABS.CEV.1995 and Bible.es.ABS.RVR.1960 and
Bible.en.IBS.NIV.1975).  This would reduce the likeness of name
collision because each organization would be allocated the equivalent of
a domain in the third position but this will not always be obvious to
many users.  This would be bad for fall back for the KJV, but would be
fine for most other translations.

Todd
End of second screen :)