Shouldn't ACTOR.languages be a list of CODE_PHRASE instead of DV_TEXT?
Description
Environment
Activity
Pablo Pazos October 14, 2022 at 6:27 PM
Sorry, just realized changing from 1..* to 0..* is not actually a breaking change since it’s a wider constraint.
About the new spec, that would be wonderful, since “demographic” is kind of short when managing other things like AGENTs (from software systems, to imaginology modalities and lab analyzers).
Related to my other concerns about deprecation of specs and cross-spec baseline, this could be a good opportunity to start doing that (1. deprecating demographic, 2. mapping which versions of which specs play well together and which versions of which specs are a “baseline” release of the whole openEHR spec family).
Thomas Beale October 14, 2022 at 10:08 AM
@Pablo Pazos I agree in fact, my proposal (yet to be discussed by SEC) would be to create a new ‘Entity’ specification, covering demographics and resources. This contains a lot of breaking changes to the demographics spec, and has a wider scope, hence being a new spec.
Pablo Pazos October 13, 2022 at 8:46 PM
Yes, also to change the identities from 1..* to 0..* will be a breaking change, but I think releasing a major version of the demographics model after wouldn’t hurt. I’m testing the model and schemas for demographic RM instances, also generating archetypes and OPTs (some by hand), and taking notes on the possible improvements to the model. This is similar of what I did for AQL which ended in several edits to the specs, so I hope we can (paraphrasing the past US president…) make the demographic specs great again!
Thomas Beale October 13, 2022 at 4:08 PM
BTW, I don’t really disagree with it being a CODE_PHRASE; it’s just that it would be a breaking change.
Pablo Pazos October 13, 2022 at 3:38 PM
@Thomas Beale I see ISO639-2 having those local variations
cpe |
| Creoles and pidgins, English based | créoles et pidgins basés sur l'anglais | Kreolisch-Englisch (Andere) |
https://www.loc.gov/standards/iso639-2/php/code_list.php
https://en.wikipedia.org/wiki/List_of_ISO_639-2_codes
If we leave DV_TEXT there, it can say Spanish or Español, or any translation of the world Spanish on any language, which doesn’t seem very useful. With CODE_PHRASE, it could be a standard code, or even a local code, if the country defines it’s own codes for languages or local variations.
In other parts of the RM when we have a language, like ENTRY.language, it is CODE_PHRASE. Why using DV_TEXT to something that should be coded?