[tei-council] <punctuation>

[Kevin Hawkins]

Ah, someone else trying to sneak in changes past the deadline!  ;-)

On 6/20/13 9:40 AM, Rebecca Welzenbach wrote:
> 1) Originally, we agreed that this new element should belong to
> model.encodingDescPart
> (http://www.tei-c.org/release/doc/tei-p5-doc/en/html/ref-model.encodingDescPart.html).
> However, it looks to me now like it fits much more sensibly in
> model.editorialDeclPart
> (http://www.tei-c.org/release/doc/tei-p5-doc/en/html/ref-model.editorialDeclPart.html),
> along with <quotation>, <hyphenation>, <normalization>, etc.
>
> Documenting the treatment of punctuation clearly falls under providing
> "details of editorial principles and practices applied during the
> encoding of a text." And I don't think we want <punctuation> floating
> around <encodingDesc> outside of <editorialDecl>.
>
> Do you agree that <punctuation> should instead be a member of
> model.editorialDeclPart, or is there good reason to stick with the
> original proposal to make it a member of model.encodingDescPart?

As I've just posted on the ticket, I believe that's a mistake in the 
minutes and our summary.

> 2) As well as allowing prose content to describe the project's policy
> for handling punctuation marks, we agreed to add some relevant
> attributes. I propose the following:
>
> @marks : Just like on <quotation>, this would be an optional attribute
> of the dataype data.enumerated, with permitted values of "none",
> "some", or "all" to indicate whether punctuation marks from the source
> text have been retained in the TEI document.

Seems right, except that we should probably create an attribute class 
for @marks since it's now identical on these two elements.  If you 
manage to implement the rest of this before the deadline, I guess just 
create a separate ticket for it.  If not, maybe it can all be done at 
once in time for the following release.

> @location: this would be an optional attribute, datatype
> data.enumerated, with permitted values of "internal" or "external" to
> indicate whether retained punctuation marks are captured as content
> within adjacent elements or placed outside of them.
>
> My model here was the use of @location on <variantEncoding>, although
> the usage of "internal" and "external" is not quite consistent between
> these two.

Instead of "within adjacent elements", I prefer "within adjacent tags" 
to make it clear that we're not talking about another opening and 
closing tag pair that happens to be nearby.  (

Other things equal, I prefer that we avoid having attributes with the 
same name that work differently on different elements.  But I don't have 
a better name for what we need on <punctuation> than @location.

--K.