[sword-devel] XML idea: modular spec

David Burry sword-devel@crosswire.org
Fri, 12 Oct 2001 10:29:58 -0700


At 08:49 AM 10/12/2001 +0000, Trevor Jenkins wrote:
>On Thu, 11 Oct 2001, David Burry <dburry@tagnet.org> wrote:
>
>[Keep in mind throughout my comments below tht I was a member of the ISO
>committee that developed SGML.]

okayyyyyyy... I take it you're a big fan of SGML CONCUR..... ;o)  I'll have 
to go study up and figure out how that exactly works now, thanks.


> > What I mean is that, suppose the Bible were stored in a binary/text
> > compressed and/or indexed format, but available for query _as_if_ it were
> > in this kind of format:
> >
> > <version name="kjv">
>
>If you draw your examples from <version name="cev"> or <version
>name="tev"> then this nice regular structure breaks down. Both these
>translations have occasions where several verses appear together. These
>verses are printed with references like 12--13 or 25--28.  In one of the
>long list of names in, I think, 1 Chronicles there's even a 28--60.

True, not every translation supports everything, it was just a simple 
example to illustrate a point.


> >    <book name="genesis">
> >      <chapter>
> >        <verse><paragraphmarker/>contents of verse 1</verse>
>
>I suggest that there's an atribute on these elemenets for chapter and
>verse numbers as appropriate. Granted that an SGML/XML application could
>count them for use but some translations, eg The New English bIble or the
>Complete Jewish Bible have moved some verse around; I think most occur in
>Isaiah but I can't remember the exact passages.

That might be practical, I was leaving them out in this example because a) 
it makes it simpler to read for an example, and b) their position can 
easily show what number they are, for instance, the XPath expression 
"book[@name='john']/chapter[3]/verse[16]" (for example in XSLT) will get 
you the expected verse without the clutter of numbering them.  This does 
not address the problem of alternate verse numbers in other translations, 
but again following the same idea there could easily be multiple views, one 
for the native numbering scheme and one for the "standard" kjv numbering 
scheme (again only where possible, more simple interfaces that require 
standard complete kjv numbering schemes simply wouldn't be able to handle 
versions that don't contain them!).

> > ... perhaps
> > someone else has even already thought of and done stuff like this.  Anyone
> > know of any?
>
>I'm currently looking at XSEM (XML Scripture Encoding Model) from
>SIL/Wycliffe as to whether it address the issues that absence of "CONCUR"
>raises. XSEM might address all your requiremetns and more, given that it's
>come from a group of professional Bible translators whoa re working with
>the text 24/7. Anyone interested in this project can find more at
>http://www.sil.org/computing/xsem/index.htm and by following the links can
>download their DTD and other materials. They have some interesting ideas
>but I'm not sure that I'll sit in church following the Bible reading on my
>WAP phone. :-)

woo hoo! thanks!  It does seem very complex, it does seem like they've put 
in everything but the kitchen sink, and maybe even that too...  Gonna take 
me a while to digest this link.

Dave