search

DARIAH-ERIC / lexicalresources

Data space of the DARIAH Lexical Resources Working Group
https://dariah-eric.github.io/lexicalresources/
BSD 2-Clause "Simplified" License
18 stars 24 forks source link

Adjust spec for entry to reflect the narrative #11

Closed ttasovac closed 6 years ago

ttasovac commented 6 years ago

Fix schema spec for entry:

laurentromary commented 6 years ago

Do you mean removing fs completely from TEI Lex 0?

ttasovac commented 6 years ago

Yes — we don't have a single example with it. I would remove it completely until we explicitly decide that we do need it and that we will be recommending the use of feature structures for this or that.

xlhrld commented 6 years ago

I'd second Toma here. <fs> is such a general meachanism that you can apply it to mark-up anything (including defs, headwords, grammatical information and all the rest). We should focus on the specific elements for a basic lexicographical encoding such as TEI Lex-0.

laurentromary commented 6 years ago

It is impossible to remove <def> and <lbl> from the content model of <entry> because this occurs through their membership to model.entryPart.tom which is also use in <sense>. This would required some class surgery and a ticket on the TEI git

ttasovac commented 6 years ago

I see. But shouldn't we then completely recreate the content model <entry> by hand, i.e. remove model.entryPart.top from <entry> (mode="replace"?)and add individual elements allowed (cit, dictScrap, entry, etym, form, gramGrp, usg, xr)?

We can't really leave def and lbl dangling there... because our narrative section and our schema would be saying different things.

laurentromary commented 6 years ago

In that case. I would take the time (probably a couple of minutes, but I did to have fresh mind) to implement something which would foresee a future proposal for the guidelines. We need at least 2 distinct classes for sense and entry

ttasovac commented 6 years ago

well... your train ride tomorrow would be perfect for this type of thinking... :)

laurentromary commented 6 years ago

:-)

Le 3 sept. 2018 à 18:27, Toma Tasovac notifications@github.com a écrit :

well... your train ride tomorrow would be perfect for this type of thinking... :)

— You are receiving this because you were assigned. Reply to this email directly, view it on GitHub https://github.com/DARIAH-ERIC/lexicalresources/issues/11#issuecomment-418156938, or mute the thread https://github.com/notifications/unsubscribe-auth/AE_Q74Rcx_6An35e08zAiLZXMD2Blnpbks5uXVh3gaJpZM4WWm1w.

Laurent Romary Inria, team ALMAnaCH laurent.romary@inria.fr

laurentromary commented 6 years ago

The solution is clear: introduce model.sensePart and use this class instead of model.entryPart.top for the content model of .... sense. We can then do some surgery on model.entryPart.top without any worry. We should file a TEI ticket at once and implement it ourselves.

laurentromary commented 6 years ago

Sorry for the lack of modesty, but I am so proud of the result achieved. It works!