[sword-devel] dynamic versification

[Jeremy Brown]

sword-devel at crosswire.org on Sunday, March 11, 2007 at 11:00 AM -0800
wrote:
>> which is currently imported as a SWORD General Book (General Books use
>a 
>> TreeKey index, as you can see from the left navigation) and to make it 
>> accessible via a new specialized VerseKey descendant (VerseKey is what 
>> SWORD Bibles use).  The new TreeKey subclass will be implemented to 
>> merely walk the existing TreeKey index of the General Book module
>format 
>> to get it's data (books, chapter max, verse max, etc) and will position 
>> the TreeKey to the appropriate node when it is positioned to, say,
>"John 
>> 3:16".

Is the goal in the new VerseKey to
1. display and interact with Bibles in their original versification scheme?
2. also allow two Bibles with different versifications to accurately
display their verses in parallel?

It seems like it is at least accomplishing #1, I don't know if you are
trying to accomplish #2 or not at this point. It is something that can
probably be tacked on later, but #1 is probably critical to accomplish
before you can do anything with #2. 

If you are trying to work on #2, then I have some tools that might be
useful, maybe I would need to rewrite them in a useful format for you
though.  I would like to help if it is needed, but if you aren't working
on #2, you probably don't need these.

One is a tool (currently a web app) that looks at the verses in the NRSVA,
and the verses in some other version, and shows where verse mismatches
occur (i.e. a verse that is in the NRSVA does not exist in the second
version, or vice versa.) It then lets you look at those chapters, find out
where the verse misalignment occurs, and add custom verse mappings, until
every verse in one version is mapped to a verse in the other. I can then
export these verse mapping rules, and know the correspondences between
verses in NRSVA and the other version.  This is used in the Unbound Bible
to show versions in parallel - the NRSVA verse scheme is used as the
default, and then all the other versions (that have been mapped) are shown
in parallel using these custom verse mappings.

The second tool is modified from a set of Microsoft published perl scripts
for comparing translated texts.  It takes Bible input in an Unbound Bible
format (book, chapter, verse, subverse) for two Bible versions, and tries
to find mismatches in the parallelity of the verses.  This can help find
places where the verses do not match up from one translation to the other.
>From that data, a person can start examining both translations around the
areas of mismatch and decide what verses in one version really match up to
verses in the other.

So I'd be happy to try to share these tools, if you think they are
relevant.

Jeremy