FW: [osis-core] osisCore_Candiate_1.1_003 - 12 - splitID
Todd Tillinghast
osis-core@bibletechnologieswg.org
Mon, 26 Aug 2002 17:24:28 -0600
Patrick,
>
> Todd,
>
> Thanks for the repost!
>
> Taking your post into account, then the syntax should read:
>
> refsys[work]:canonical@grain-canonical@grain
>
> with a separate splitID, which should be xs:string, that users could
use
> either to indicate order (your case 2) or portion of segmented element
> (your case 1). If it is xs:string, doesn't that get you numbers and
> letters?
The trouble with using splitID attribute value to replace what you
intended within the regex is:
1) there is no splitID in a <reference> element
2) there could be two split extensions in a range reference and there is
only a single value
3) there can be more than one self-id in an osisID and there is only a
single value in splitID. There would be not way to determine which
self-id the splitID extension applies to.
It seems I should go back and try to explain what I did not clearly
communicate in my first post about splitID.
There are TWO purposes that we seem to be tring to accomplish with
splitID.
The first is to provide a mechanism to extend the predominant reference
system when an identifier from that system applies to MORE THAN ONE
<verse> element. In this case the thing that is split is the logical
concept of the verse but there is NO overlapping hierarchy as far as
element go.
The second case is where there are overlapping hierarchies and we must
split what would be a single element into two (or more) elements so that
elements from another hierarchy can hold their position in the XML
hierarchy. In this case we need a mechanism independent from osisID to
indicate that an element has been split as well as to match up the
pieces of the split element. (If the splitID attribute is used as a
mechanism to extend a reference system (as in the first case above)
then it will not be clear when an element has been split and when a
reference system is being extended.)
PROPOSAL FOR SPLIT ELEMENTS DUE TO OVERLAPPING HIERARCHIES:
1) Use the splitID attribute to indicate that an element has been split
due
to overlapping hierarchies of elements.
2) Have splitID be xs:string.
3) Have splitID be the exact same token for all pieces of the split
element.
4) Have the sequence of the split element be implied by their order in
the document.
5) Have the splitID strategy be TOTALLY independent of osisID.
6) Put splitID attributes on the elements that can be split and not in
globalAttributes. (div, q, verse, p, lineGroup, line, not sure if there
are others)
PROPOSAL FOR EXTENDING HIERARCHIES:
1) Not use splitID in any fashon
2.A) use the [] syntax to indicate a reference system extension OR
2.B) use the grain to accommodate reference system extension
<snip>
Are there any reasons that this does not work?
Todd