Would linking via "ana" attributes help in situations where there is a single instance in the copy text
applicable to several instances of roles?
<roleDesc xml:id="fiendzz"> ... </roleDesc>
<role ana="fiendzz" xml:id="duo1" corresp="duo2">...</role>
<role ana="fiendzz" xml:id="duo2" corresp="duo1">...</role>
Likewise the role could be linked to the <head> in the <castGroup>. In P4 "ana" can take more than one
target (separated by white space). "The ana attribute may be specified for any element. Its effect is to
associate the element with one or more others representing an analysis or interpretation of it. Its target
should be one of the elements described in the section"
I know I am pushing attributes again. It appears to be an under-utilized axis accesible to XSLT processing.
> Very interesting. My own feeling is that in both cases (the "friends of
> Mathias" and the "twin brothers"), they are clearly role descriptions
> and would be recognized and encoded as such instantly if they were
> presented in the context of an ordinary cast item; the following
> encoding wouldn't raise any eyebrows:
> <castItem><role>Walter</role> <roleDesc>friend of
> <castItem><role>Hans</role> <roleDesc>friend of
> and in fact if in an introductory TEI class you pointed to the following
> text and asked "so, what is 'friend of Mathias'?"
> Walter, friend of Mathias
> you would very much hope for a chorus of students answering "a <roleDesc>!"
> So <head> and <trailer> equally puzzle me. From a processing standpoint,
> I would like to imagine that all the role descriptions might be encoded
> alike so that I know where I can grab them. I guess one might answer
> that "within <castGroup>, if there's a <head> or a <trailer>, that's
> where the role description is" but this seems unsatisfying somehow. And
> what if one really did find a text with something resembling a true head
> or trailer inside a cast group?
> many thanks--Julia
> Michael Beddow wrote:
> > I suspect that the impression of abusiveness comes as much from the
> > nomenclature as anything else, which I imagine was probably inspired (along
> > with <head> which can precede the <castItems>) by the div-like nature of
> > this structure. If we look at one of the examples in the body text of P5
> > (10.1.4) , we see
> > <castGroup rend="braced">
> > <head>friends of Mathias</head>
> > <castItem>
> > <role>Walter</role>
> > <actor>Mr Frank Hall</actor>
> > </castItem>
> > <castItem>
> > <role>Hans</role>
> > <actor>Mr F.W. Irish</actor>
> > </castItem>
> > </castGroup>
> > The implication of the rend value of "braced" applied to the whole group is
> > that the text of the <head> should be rendered in such a way that it visibly
> > applies to all the contained <castItem>s. And presumably the same would hold
> > good about the rendering of any <trailer> that might be added after the
> > <castItem>s, which would take care of the general alignment problem of the
> > element ensemble, while remaining agnostic about specific line-breaks
> > within any of the text nodes. My point re nomenclature is that semantically
> > (though not in this case particular case syntactically) "friends of Mathias"
> > could be construed as a role description, in much the same way as "Twin
> > brothers..." could also be interpreted as such. But the Guidelines here show
> > that conjoined role description assigned to a <head> And if that is
> > legitimate (despite it not being all that much like the content of <head>s
> > we find elsewhere) I think there's a parallel case for assigning the "Twin
> > Brothers..." to a <trailer>, possibly with some mention, indeed maybe an
> > additional example, in the Guidelines, to indicate that it is indeed
> > available for such conjoint labelling, all governed by the value of "rend"
> > on the enclosing grouping element.
> > Michael Beddow
Francois Lachance, Scholar-at-large
Everyone is a little bit crazy; everyone at some time has a learning disability;
No one is ever a little bit positive.