On Wed, 2006-01-04 at 15:06, Lou Burnard wrote:
> Peter Flynn wrote:
> > Where is a suitably obvious machine-identifiable place (element or
> > attribute) for the language code of the original document when the
> > current document is a transcription of a translation?
> The ms description module in P5 defines a new element, <textLang> for
> this purpose.
> > All our files have one or more profileDesc/langUsage/language/@id for
> > the language[s] of the current document, which in the case of a
> > transcription of a printed translation from (say) Spanish into English
> > says just "en" because no Spanish words exist in the translation.
> > Should I [ab]use another language/@id or another attribute on the
> > existing language element[s] to identify the original source language,
> > and if so, how best to make it obvious that that is what it refers to?
> I think the only way to do this cleanly with P4 would be to add a <note>
> to the <notesStmt> saying something like "Translated from the Spanish"
> or whatever.
In the P4 interim, though, I need a machine-identifiable location (eg
an element or an attribute) for the language code, to assist in the
machine collation of the corpus.
> > Files also have a sourceDesc/biblStruct for the printed edition from
> > which this transcription was made, and one or more much simpler bibl
> > elements for the MSS used by the author of the printed edition. Is
> > the lang attribute on bibl* elements is intended to specify the language
> > of the bibl* element content,
> Yes it is is.
> or can it be used to denote the language
> > the referenced document is written in?
> No, definitely not it isnt.
I suspected as much.
> A happy newt to all our readers!
And a merry gingrich to you all.