[tei-council] Making absolutely bare a little more bare

[James Cummings]

David, et al.

I think I'd be one of those to argue against providing as a starting 
point the customisation David is suggesting.

Though it is fairly easy, you just add  start="TEI teiCorpus text" to 
your ODD's schemaSpec and hey presto, you are also able to start your 
documents at the <text> element.  In fact, I have one of these that is 
tei_all that uses <div> as an optional start, but _only_ because these 
are separate documents I want to validate separately (and for oxygen 
support when editing them), but I know that they will certainly be 
incorporated into a larger TEI document. I'd worry about people feeling 
it is recommended practice to create TEI documents without a header.

While I'd be the first to defend the flexibility and occasional 
redundancy of multiple methods in the TEI world, I think where we can 
encourage best practice (e.g. in documenting your file) then we should.

The header-related bugbear that I think we *should* tackle when we come 
back to this is the number of instances in the header where use of 
macro.phraseSeq or macro.paraContent or similar really allows entirely 
unnecessary child elements. For a randomly selected example, look at 
everything that  <extent> can contain. 
http://snipurl.com/tei-p5-doc/ref-extent.html Does it really need all 
those?  (Yes, I know the problem is it is also used elsewhere...but 
maybe we should start using schematron or something to constrain this?) 
Maybe this comes under the revision of ODD though?

-James

David Sewell wrote:
> I would agree.
> 
> When we come back to this discussion, we may find that there are other
> possible options. For example, we might want to make available an
> explicitly non-TEI-conformant schema that would allow <text> as a root
> element, to address the particular case where one wishes to encode a
> document provisionally without creating a TEI header. (I'm not
> necessarily advocating this, just suggesting it as an alternative to
> stripping more required elements from teiHeader.)
> 
> David
> 
> On Mon, 8 Dec 2008, Laurent Romary wrote:
> 
>> I had this in mind when asking the question in the context of TEI
>> bare. It was just a good opportunity of expression my frustration at
>> keeping producing TEI documents where part of the header would consist
>> of:
>> <pubStmt><p>...<p></pubStmt> just to make sure that the thing would
>> parse.
>> I suggest the following:
>> - simplify TEI bare in accordance with the simple and obvious  parts
>> of the proposal I made
>> - keep room in the next F2F for a debate on optionalities in the
>> header, since I think it's an important debate
>> Cheers,
>> Laurent
>>
>> Le 8 déc. 08 à 12:59, James Cummings a écrit :
>>
>>> Arianna Ciula wrote:
>>>>
>>>> James Cummings wrote:
>>>>
>>>>> However, I don't really care that much, but perhaps a
>>>>> representative of
>>>>> the TEI Library community might have some thoughts on this?
>>>> I agree with your points James and wouldn't like to see <sourceDesc>
>>>> disappear from TEI Bare...but then I am not a librarian.
>>> Well, librarians may or may not care about making it optional. I don't
>>> really know. My only concern has been that I don't want us to make
>>> such
>>> a drastic change hastily. ;-)
>>>
>>> Best,
>>> -James
>>>
>>> --
>>> Dr James Cummings, Research Technologies Service, University of Oxford
>>> James dot Cummings at oucs dot ox dot ac dot uk
>>> _______________________________________________
>>> tei-council mailing list
>>> tei-council at lists.village.Virginia.EDU
>>> http://lists.village.Virginia.EDU/mailman/listinfo/tei-council
>> _______________________________________________
>> tei-council mailing list
>> tei-council at lists.village.Virginia.EDU
>> http://lists.village.Virginia.EDU/mailman/listinfo/tei-council
>>
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> tei-council mailing list
> tei-council at lists.village.Virginia.EDU
> http://lists.village.Virginia.EDU/mailman/listinfo/tei-council


-- 
Dr James Cummings, Research Technologies Service, University of Oxford
James dot Cummings at oucs dot ox dot ac dot uk