[osis-core] osis_0108.3 Making milestone OSISID and OSISIDREF match.

[Patrick Durusau]

Todd,

I am glad that you have kept pursuing this issue as it allows us to make 
clear the differences between (and the need for a mapping) between 
reference systems.

Todd Tillinghast wrote:

>I most cases verse milestone markers are for a single verse.
>
>However, there are several cases (in the TEV anyway) where multiple
>logical verses are grouped.  In this case there is a start of Matt.1.2
>and an end of Matt.1.6a and a start of Matt.1.6b and an end of
>Matt.1.11.  (Look the printed text to make it clearer.)  Patrick has
>suggested in the past that this is purely a rendering issue.  I believe
>that this could be the case in in Matt.1 where the verse markers are not
>place within the lineage.  However, I don't think that it explains what
>we see in Gen.20.  It seems to me that what is traditionally know as
>Gen.20.17 and Gen.20.18 have been inseparably mixed by the translators
>to make the text flow.  I have provided the NIV version of the same two
>verses and the flow of ideas is reversed between the two cases.  (I'm
>sure the Hebrew would be a better comparative text but I don't read
>Hebrew.)
>
There is NOT a grouping of multiple logical verses in TEV! It is a 
difference in reference systems, as shown by your examples:

NIV: uses the traditional verse reference system as shown in your example:

>
>Bible.NIV  (Provided as a comparison of the ideas in the text and how
>the NIV decided to break up the verses.)
><verseStart OSISID="Bible.NIV..Gen.20.17" />Then Abraham prayed to God,
>and the God healed Abimelech, his wife and his slave girls so they could
>have children again, <verseEnd
>OSISIDREF="Bible.NIV..Gen.20.17/><verseStart
>OSISID="Bible.NIV..Gen.20.18" />for the Lord had closed up every womb in
>Abimelech's household because of Abraham's wife Sarah.<verseEnd
>OSISIDREF="Bible.NIV..Gen.20.18/>
>
>Bible.TEV
>
The TEV takes two verses that are treated as separate in the NIV (and 
other reference systems) and combines them together.

But, it tries to preserve a reference to the other systems by a 
reference to the other system (NIV and others, as 17-18).

Essentially, the CEV (the version I have on hand that follows your 
point) does not have a specified reference system but has pointers to an 
outside reference system.


>
>Option A: (This is not valid because Bible.TEV..Gen.20.18 is an xs:IDREF
>with no matching xs:ID.)
><verseStart OSISID="Bible.TEV..Gen.20.17" />Because of what had happened
>to Sarah, Abraham's wife, the <divineName type="yhwh">LORD</divineName>
>had made it impossible for any woman in Abimelech's palace to have
>children. So Abraham prayed for Abimelech, and God healed him. He also
>healed his wife and his slave women, so that they could have
>children.<verseEnd OSISIDREF="Bible.TEV..Gen.20.18" />
>
Agree Option A is a non-starter.

>
>Option B: (This works from the xs:ID and xs:IDREF perspective.  But this
>creates a new identifier that is unique to the Bible.TEV reference
>system, which will require a lookup to map to other references.  It also
>means that an OSISIDREF to Bible.TEV..Gen.20.17 would be invalid if
>place else where in the same document.)
><verseStart OSISID="Bible.TEV..Gen.20.17-18" />Because of what had
>happened to Sarah, Abraham's wife, the <divineName
>type="yhwh">LORD</divineName> had made it impossible for any woman in
>Abimelech's palace to have children. So Abraham prayed for Abimelech,
>and God healed him. He also healed his wife and his slave women, so that
>they could have children.<verseEnd OSISIDREF="Bible.TEV..Gen.20.17-18"
>/>
>
Agree that it creates a unique ID for TEV, but it is the TEV "reference" 
so one would not expect to have an OSISIDREF to Bible.TEV.Gen.20.17 
since it does not exsit in the TEV reference system. Correct?

Assuming (not sure this is the case) that such cases always involve 
verses that are contiguous, could resolve the mapping problem with:

Option D:

<verseStart OSISID="Bible.TEV..Gen.20.17-18" CommonStart="Gen.17.17" CommonEnd="Gen.17.18"/>Because of what had
happened to Sarah, Abraham's wife, the <divineName
type="yhwh">LORD</divineName> had made it impossible for any woman in
Abimelech's palace to have children. So Abraham prayed for Abimelech,
and God healed him. He also healed his wife and his slave women, so that
they could have children.<verseEnd OSISIDREF="Bible.TEV..Gen.20.17-18"
/>

At the expense of adding two new attributes to verse milestones. But 
this get you the mapping without having to hold the material in separate 
tables. And, it puts the burden of doing that mapping on people who are 
constructing non-specified reference system that refer to outside 
systems. If they want a mapping, it is their obligation to provide one. 
(I am sure you, Troy and Chris could autogenerate such mapping 
attributes for the TEV.)

Note that I think we should default to Gen.17.17 for references and 
allow the more specific as an option only.

>
>Option C: (This causes there to be two cases for how a <verseStart> and
><verseEnd> would behave, because in this case the OSISID of the
><verseEnd> holds vital information that would often be ignored if the
>OSISID equals the OSISIDREF in a <verseEnd>. Note:There can be verses
>that have not ID representing them in the text if the range of verses of
>this type is greater than two.]
>)
><verseStart OSISID="Bible.TEV..Gen.20.17"
>OSISIDREF="Bible.TEV..Gen.20.18"/>Because of what had happened to Sarah,
>Abraham's wife, the <divineName type="yhwh">LORD</divineName> had made
>it impossible for any woman in Abimelech's palace to have children. So
>Abraham prayed for Abimelech, and God healed him. He also healed his
>wife and his slave women, so that they could have children. <verseStart
>OSISID="Bible.TEV..Gen.20.18" OSISIDREF="Bible.TEV..Gen.20.17"/>
>
Could  do C but I think it would be more confusing that suggesting 
people follow D.

I am working my way through all the other posts and probably will not 
issue the next iteration until late today. Would like to get some 
consensus on this and other issues before the next release.

Thanks for all the hard work!

Patrick


>
>QUESTION:  Is the current structure sufficient?  (I say yes.)
>QUESTION:  Is option C is the "best practice" for this sort of case.
>
>My vote is for Option C.  I know the standard as it sits allows for both
>Option B and Option C, but I am trying to make sure we are comfortable
>that this case can be encoded.
>
>Todd
>
>

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