[osis-core] osisWork, refSystem - Summary

Steve DeRose osis-core@bibletechnologieswg.org
Thu, 22 Aug 2002 17:40:16 -0400


>The obvious issue is that there is redundant information in the
><header> and in the <work> element. Do I really need a <work> element
>to define augustine.confessions.SPCK1912, and if so, should we drop
>the header elements?

I don't mind the redundancy, actually; I'm not sure we would often be 
defining reference systems one-to-one with works. For Bibles I think 
we'll have a few widely used refsystems, and a few idiosyncratic 
ones. For classical works, many will have either just Loeb, or Loeb 
and one other. And we could define "Loeb" as a single reference 
system that extends across all or nearly all classical works -- so we 
would need a refsystem declaration that points to the definition of 
how Loeb does referencing (basically section or page and line of 
their first printing, I think), but no more than that unless we want 
to do reference value validation for all the Loeb works (in that case 
we might want some special way to do it; seems a bit excessive to 
list the number of lines on every page of every work; I think most 
would be happy with validation that their Loeb ref is always 2 
integers, and neither is over 1000; or some such.

-- 

Steve DeRose -- http://www.stg.brown.edu/~sjd
Chair, Bible Technologies Group -- http://www.bibletechnologies.net
Email: sderose@speakeasy.net
Backup email: sjd@stg.brown.edu