search

globalwordnet / schemas

WordNet-LMF formats
https://globalwordnet.github.io/schemas/
19 stars 11 forks source link

Release 1.1 #38

Closed jmccrae closed 3 years ago

jmccrae commented 3 years ago

This the pull request for the 1.1 Release Candidate of the GWA Schemas. Let me know if I have missed anything

fcbond commented 3 years ago

Hi,

I still need to add the new relations. I will try to do that today.

goodmami commented 3 years ago

Weren't we going to change the URL of the dc namespace? (see https://github.com/globalwordnet/schemas/issues/24#issuecomment-670516816)

goodmami commented 3 years ago

Weren't we going to change the URL of the dc namespace? (see #24 (comment))

For this, see #39.

goodmami commented 3 years ago

If it's not too late, why is the id attribute of the new <SyntacticBehaviour> elements #IMPLIED? Without it, the senses have nothing to refer to in their subcat attribute, and the syntactic behaviour entity would be useless. Can't this be #REQUIRED?

jmccrae commented 3 years ago

The id attribute is not required so that we have backwards compatibility with the previous schema.

goodmami commented 3 years ago

Ah yes, compatibility. So is the intention (that cannot be spelled out in DTD) then that the attribute should be required when the element appears under <Lexicon> and forbidden when the element appears under <LexicalEntry>? That is, when we eventually shed ourselves of the constraint of compatibility, it will become required? In any case, I guess for now I cannot rely on the presence of the id attribute.

jmccrae commented 3 years ago

Yes, it can't really be enforced by the DTD. I would also note that a syntactic behaviour underneath the lexicon with no ID is obviously useless, so it should be clear to data providers that an ID is needed and data consumers could just discard this.