Sure, if I wanted to be explicit in my example I might use @corresp as you suggest pointing between the two. I tend to just use unspecified loose correlation by ID numbers myself. So if I have Volpone001-facs and Volpone001-text I'd document that in the project notes and write the processing to assume that every $ID-text had a corresponding $ID-facs or something similar to that. But actually pointing at it with corresp would be much clearer (and also enables the ability for you to keep facsimiles and texts in separate documents, at separate URIs if you wanted to be all distributed about it).
Thank you, James. What you suggest was actually my first thought, but then I thought it couldn’t possibly be that simple. So you have a sequence of facsimile elements, each of them mapped to a text child of group through some explicit linking, e.g.@corresp. If you want to treat the text as separate documents (Volpone, Catiline etc. in the Jonson folio), you just combine the relevant facsimile with the relevant and do something about the header.
As long as there is some explicit link between the facsimile and corresponding text elements, the use of surfaceGrp doesn’t really buy you anything. Or does it?
If I'm understanding correctly, you wanting to group surfaces relating to a single text, why not just repeat <facsimile>, one corresponding to each <text>? If it is that you want to group sets of facsimile elements (for each <group>) then I don't think it is worth the hassle personally.
I'd just do something like:
But I may be misunderstanding your need?
From: TEI (Text Encoding Initiative) public discussion list <[log in to unmask]> on behalf of Martin Mueller <[log in to unmask]>
Sent: 17 March 2018 23:58:08
To: [log in to unmask]
Subject: a question about surfaceGrp
Is there a good way of grouping <facsimile> elements so that each <text> childe of a <group> element has its own <facsimile> section?
I have the following use case in mind. In our EarlyPrint project we provide enriched versions of TCP texts, some of them with <facsimile> elements that map to image sets currently in the Internet Archive. You can find examples at https://texts.earlyprint.org.
Some TCP texts use the <group> element. Typical examples are the Ben Jonson folio (A04632) or the three volumes of Hakluyt (A02495). We have mounted these texts on our site, but we also make them available for downloading so that folks can use them for their various projects. There are reasons why one would want to keep all the texts of the group together. There are also reasons why you would want to split them. We split the Jonson plays because the predecessor of this site was a drama site. A user who waits for the very long three volumes of Hakluyt to load into eXist might have preferred a decision to split the text into volumes. There is the additional factor that the Internet Archive has images for the first two volumes, but not the third. And so on.
So it would be nice to have a solution that lets you split or combine texts with minimal effort. If the facsimile section is one undifferentiated stream of surface children from 1 to more than 1000, it takes some (and probably manual) effort to figure out the start and end points of surfaces for each <text> element.
Is <surfaceGrp> an appropriate element for establishing an explicit relationship between a <text>child of <group> and its corresponding image set? The use case given in the Guidelines talks about recto and verso. If the answer to my question is ‘yes’, it might be a good idea to mention a use case of this type so that it becomes apparent to naďve users like myself to see right away that <surfaceGrp> can be used for any number of purposes ranging from the micro structure of a page (or even part of it) to the macrostructure of works and volumes.