[tei-council] Location of <app>s for external apparatus

Fri Jun 15 18:21:19 EDT 2012

See below ...

On 6/15/2012 12:08 PM, Martin Holmes wrote:
> On 12-06-15 01:49 AM, Sebastian Rahtz wrote:
>>
>> On 15 Jun 2012, at 09:45, stuart yeates wrote:
>>
>>> It would be useful to have a global clue that something is standoff (I
>>> anticipate a need for it on the SOM group, but we're not reached there
>>> yet).
>>>
>>> Is there one?
>>
>>
>> I dont think so. in the case of eg<link>, its _always_ standoff, but<app>
>> and<note>   can be used in either way, which is confusing (to me).
>>
>> one could almost posit _another_ sibling of<text>   called<standoffText>   where
>> all this stuff lives.
>
> I think one should seriously posit such a thing. A born-digital document
> is a very different scenario from a traditional
> transcription-from-source, and leaving born-digitals aside for the
> moment, we have a situation where the original source document probably
> has<front>,<body>  and<back>, and all the real original content would
> be transcribed there, in the order in which it appears. Adding in
> standoff content to<back>  is ambiguous, and problematic for anyone
> writing stylesheets or other processors. And I share James's unhappiness
> with suggesting that something as ill-constrained as @type be used to
> distinguish the two cases.
>
> So I like the idea of a fourth child of<text>, which explicitly
> contains stuff which is out-of-sequence, editorial, or otherwise not
> part of the normal flow of the source document. I don't like the name
> <standoffText>, though, because it may not always be textual stuff;
> <linkGrp>s and<link>s belong there too, for instance.

And <interpGrp>!  It has long driven me crazy that you insert this in 
<back> even though it basically never occurs in the source document but 
that there isn't a recommended way to indicate that such an element is 
not meant to be rendered as part of the encoded text.

So I would be very interested in the fourth child of <text> as well. 
But I wonder whether we can solve the <listApp> problem separately, even 
only for those who are encoding an apparatus that occurs in the source 
document.

> We need a name that incorporates the idea that these items are not part
> of the source document flow, although they may include text which ends
> up being rendered into that flow.

<supplements>?