Thanks Peter and Lou for input on this. I think we can say now that: - All transformations prior to the output of RNG or HTML at the end of the chain should preserve all language versions of gloss/desc/remarks, even those which are technically obsolete per their @versionDate attributes. - A final rendering process should use elements in the specified docLang, EVEN THOSE WHICH ARE OUTDATED, in preference to English, but should fall back to English where nothing is available in the target language. - If the chain originates with P5, then English is special, because it's the canonical language of the base schema. But I like Peter's suggestion that a language could/should be declared as canonical for a schema, and where no element exists in that language, and error could be raised. If that sounds reasonable to everyone, then it answers all the questions I had; if Syd can't think of anything we've missed, then perhaps the topic doesn't need to be discussed at the F2F after all. Cheers, Martin On 2022-03-31 02:44, Peter Stadler wrote:

Hi Martin and Council,

here’s my attempt to answer these questions. My general assumption is that: a) ODD being used elsewhere than TEI, there need not be an English fallback for every desc or gloss in every source of ODD chaining b) you’re asking about ODD chaining in general, and there’s no special case of TEI chaining

Ad 1) Correct me if I’m wrong, but @docLang is only needed for the output formats (e.g. HTML, RelaxNG), not for the "intermediate“ compiled/processed ODD. So I’d say to push *all* language versions of descs and glosses (and examples) to every subsequent processed ODD in the chain. The application of the docLang parameter is deferred until the (final) processing of schemas and documentation. Ad 2) Good one! Yet that’s sort of a semantic inconsistency (of schemas and documentation being out of sync) that we are hardly able to catch with syntactic tests. Given my assumption that there might be other (main) languages involved outside TEI ODD, we can’t even rely on tests that check @versionDate against the canonical English desc or gloss. A possible solution that jumps to my mind is the introduction of a canonical (default) language that needs to be declared in every ODD (or on every *Spec?) Given such a language, we could enforce its existence and check (the versionDate of) other languages against it Ad 3) Yes and no. The current hard coded fallback „en“ is a legacy and a mere hack. Yet it’s probably more important for the data format to declare a default (see ad 2). Having the user specify a sequence of languages is more an optional thing to me.

Best Peter

...

Am 29.03.2022 um 23:12 schrieb Martin Holmes :

Hi there,

I have the discussion on languages and ODD chaining on my schedule for 8am PST on Saturday. Helena, Syd and I put together three unresolved questions for that session:

https://wiki.tei-c.org/index.php?title=New_ODD_processing#unresolved_questio...

and I think if Council is able to give clear answers to those questions, we'll be in good shape when we come to work on ODD chaining in the XSLT rewrite.

Cheers, Martin

On 2022-03-29 13:58, Scholger, Martina (martina.scholger@uni-graz.at) wrote:

...

Dear Council, This is the agenda for our virtual F2F meeting on Friday and Saturday (April 1 and 2): https://docs.google.com/document/d/1WhGAcLK0H_xNNPoEECBk9f0eEn-qhSiWfqnhaikS... https://docs.google.com/document/d/1WhGAcLK0H_xNNPoEECBk9f0eEn-qhSiWfqnhaikS... Please feel free to add more topics to the agenda. Meeting times: *Friday, Apr 1: * North American break-outs The North American group needs to decide whether to start at 12 PDT / 15 EDT or at 13 PDT / 16 EDT (see poll on slack). *Saturday, Apr 2:* European break-outs 10:00–13:00 CEST Full Council 06:00–10:00 PDT 09:00–13:00 EDT 15:00–19:00 CEST I’ll add the table with links to the Guidelines/Stylesheets issues tomorrow. Best, Martina Dr. Martina Scholger Centre for Information Modeling Austrian Centre for Digital Humanities University of Graz A-8010 Graz | Elisabethstraße 59/III +43 316 380 2291 http://informationsmodellierung.uni-graz.at http://informationsmodellierung.uni-graz.at/| http://gams.uni-graz.at http://gams.uni-graz.at/ Chair of the TEI Technical Council | http://www.tei-c.org/ http://www.tei-c.org/ Institut für Dokumentologie und Editorik e.V. | https://www.i-d-e.de/ https://www.i-d-e.de/ _______________________________________________ Tei-council mailing list Tei-council@lists.tei-c.org http://lists.lists.tei-c.org/mailman/listinfo/tei-council

-- ------------------------------------------ Martin Holmes UVic Humanities Computing and Media Centre

I acknowledge and respect the lək̓ʷəŋən peoples on whose traditional territory the university stands and the Songhees, Esquimalt and WSÁNEĆ peoples whose historical relationships with the land continue to this day. _______________________________________________ Tei-council mailing list Tei-council@lists.tei-c.org http://lists.lists.tei-c.org/mailman/listinfo/tei-council

-- ------------------------------------------ Martin Holmes UVic Humanities Computing and Media Centre I acknowledge and respect the lək̓ʷəŋən peoples on whose traditional territory the university stands and the Songhees, Esquimalt and WSÁNEĆ peoples whose historical relationships with the land continue to this day.