[tei-council] List @type musings

[Martin Holmes]

Hi Fabio,

On 14-08-06 08:05 PM, Fabio Ciotti wrote:
> Sorry if I jump into the debate late, I was in a sort of unwired
> vacation last days...
>
>>> fair point. i might riposte that we need a new attribute on
>>> list (comparable to @place), to contain “numbered” and
>>> “bulleted”, keeping @rend for “shinygoldbullets” or
>>> “romannumbersincurlybraces”
>>
>> And what would that first attribute be holding? the 'type' of
>> list? *ducks in case MartinH throws anything*
>>
>
> mmhh this observation prompts me to say (as I said during the
> discussion in Oxford) that there is more than a simple rendtional
> distinction in the difference between a (anyway) numbered list (a
> sequence) and a (anyway) bulleted one (a set or bag in rdf term)...
> it's a typological difference at a level of abstraction higher than
> that of list types we have as suggested values now. Anyway whats' done
> is done, especially if it is well done :-).

There are definitely different types of list (gloss, litany, syllogism 
etc.), and @type can be used for those. If you believe that the 
renditional aspects of a list (whether sequence matters or not, whether 
that's explicitly marked, etc.) is another sort of typology rather than 
a renditional feature, then you can use @subtype for that; but if you 
believe (as I do) that bullets, inline/block, etc. are all renditional, 
then you can use @rend or @rendition/@style. So I think we've covered 
all the bases now.

> As for the val list on @rend I vote for delaying any action and
> discuss the item more deeply. In my experience @rend has been a sort
> of basket to put any presentational feature an encoder would specify
> in any lingo she likes, and especially since @style (or the more
> complex @rendition) is available for expressing performative and
> formal style expression I think it should be as free as possible. The
> better way would be to give some hints in prose probably where it can
> halp (as in this case).
>
> And +1 for refactoring local @type into model att.typed membership,
> after a scrutiny of each case. Maybe the owner of the ticket could
> assign the scrutiny of a bunch of cases to some sub-owners to help

Paul, would you like to do that? You could raise a new ticket for each 
batch of five elements, and assign each one to one of the new Council 
members. It would be good to build on the little training session we had 
in Oxford so everyone gets comfortable with editing the source.

Cheers,
Martin