I'm sure I can hear Laurent talking about the disadvantages of implicit
grouping and I can see myself wholeheartedly agree.
I'd like to have a generalised, semantically transparent bracket
element, but I realise that it's not something that meets with
widespread enthusiasm. Still, if one can do the kind of wrapping that
Gabi's colleagues came up with, I'd say it's better than counting kinds
of tokens and drawing the implicit line every second or third of them.
IDs/refs are cute and I like Baroque overall, but why not, at points
where it seems to make some sense, yield to the tree-like, hierarchical
and thus naturally grouping-friendly nature of XML?
Then, what Syd says:
> Any TEI software needs to able to apply the "multiple <publisher>
> elements each followed by info about it" logic for the
> <publicationStmt>, so use the same algorithm for the <bibl>s! :-)
could be perhaps, gradually, turned upside down, in a way. Rather than
extending the logic of implicit grouping, we could/might/please narrow
it down, perhaps?
On 06/16/2011 11:04 PM, Gabriel BODARD wrote:
> <placeName>New York</placeName> <placeName>London</placeName>
> <placeName>Berlin</placeName> <orgName>de Gruyter</orgName>