[osis-core] Users Manual Bug List - 6. Q in transChange

[Chris Little]

Hi folks,

That Feb 15 deadline for comment on the buglist is creeping up fast, so 
I thought I'd comment at least on the more important (to me) issues from 
the list....



<q> can contain <transChange> but not vice versa.  I think this is 
appropriate, so I would NOT recommend adding <q> to <transChange> as 
suggested.

In the example cited, we have italics (used to mark a translation 
change) straddling a quotation boundary.  Apologies for being so 
dogmatic, but I think the markup supplied by Jim simply incorrect.  If a 
<transChange> straddles a <q>, it is, by definition, two distinct 
<transChange> elements.  It has to be two separate (in this case) 
additions, supplied to make two separate clauses grammatical (the first 
level quotation and the second level quotation, respectively).

The correct markup is simply:

<verse osisID="Matt.16.14>
They said, <q>Some <transChange type="added">say</transChange>, <q
sID="Matt.16.14.q.1"/><transChange type="added">He is</transChange> 
<name type="human">John the Baptist</name>,<q eID="Matt.16.14.q.1"/> but 
others, <q><name type="human">Elijah</name>,</q> and others, <q><name
type="human">Jeremiah</name> or one of the prophets.</q></q></verse>

And this requires no changes to the schema.

--Chris


> 6. Q in transChange
> 
> I am trying to encode the following text (Matt 16:14): 14They said, "Some 
> <say, 'He is> John the Baptist,' but others, 'Elijah,' and others, 
> 'Jeremiah or one of the prophets.'" In this text, <..> should translate to 
> <transChange type="added"> This means that I get
> 
> <verse osisID="Matt.16.14>
> They said, <q>Some <transChange type="added"> say, <q 
> sID="Matt.16.14.q.1"/>He is</transChange> <name type="human">John the
> Baptist</name>,<q eID="Matt.16.14.q.1"/> but others, <q><name
> type="human">Elijah</name>,</q> and others, <q><name
> type="human">Jeremiah</name> or one of the prophets.</q></q></verse>
> 
> The problem with this is that <transChange> does not allow a <q> element 
> to appear in it. Submitted by: Jim Schaad, jimsch@nwlink.com
> 
> Resolution: I suggested:
> 
> <verse osisID="Matt.16.14>They said, <q>Some say, <q
> sID="Matt.16.14.q.1"/><transChange>He is</transChange> <name
> type="human">John the Baptist</name>, but others, <name
> type="human">Elijah</name>, and others, <name type="human">Jeremiah</name>
> or one of the prophets.</q></verse>
> 
> at least based on the presentation of the text in the NIV.
> 
> Followup note from Jim makes it clear that he wants <transChange> to cross 
> element boundaries. From my followup post to Jim:
> 
> When you say:
> 
> 
>>I cannot milestone a transChange, and I
>>cannot start a speech (or a quote) inside of a transChange.
> 
> 
> You do not mean that the <speech> or <q> should be
> contained inside the <transChange>? Correct?
> 
> In other words, what you are really looking for is for
> <transChange> to become milestoneable such that you can start a
> <q> or <speech> inside a <transChange>?
> 
> Taking your earlier example:
> 
> "Some <say, 'He is> John the Baptist,'
> 
> You want:
> 
> <q>Some <transChange sID="0101" type="added"/>say <q>He is<transChange
> eID="0101"> John the Baptist,</q></q>
> 
> So it is <transChange> being able to cross element boundaries
> (along with the appropriate content models) that is the issue?
> 
> Sorry, you're right, I missed that from the original post.
> 
> The other question will be, what advantages/disadvantages are there to the 
> following:
> 
> <q>Some <transChange type="added">say</transChange><q><transChange
> type="added">He is</transChange> John the Baptist</q></q>
> 
> Captures the <transChange> without crossing the boundaries.
> 
> May need to look at where transChange can occur, presently:
> 
> catchWord
> cell
> chapter
> closer
> div
> item
> label
> l
> mentioned
> p
> q
> rdg
> salute
> seg
> speech
> verse
> 
> Children of <transChange>
> 
> a
> abbr
> date
> divineName
> foreign
> hi
> index
> lb
> milestone
> milestoneEnd
> milestoneStart
> name
> note
> reference
> seg
> w
>      
> More comments from Jim on a solution: Yes, you now have the sense of the 
> issue. My suggestion would be to keep <transChanges> as non-milestone 
> able, but to allow for a <q> milestone (and relatives) to be placed inside 
> of it. I would be leery of the suggestion of breakin g up the 
> <transChange> as you make it much harder to determine where to place the 
> brackets (using my scheme to mark these items) and generally any notes on 
> the <transChange> want to apply to the entire item.