Closed sroertgen closed 2 months ago
I tried it with dc:description
(as it is mentioned in https://metadaten.community/t/lrmi-vokabulare-als-json-ld/59/3?u=acka47 that DC vocabs used it) which does not work. Please add this as well. (I just now added it to the original issue description.)
Although I have removed dc:description
from hcrt.ttl
it still won't build: https://test.skohub.io/build/?id=3e4fc380-6e5d-4137-b917-df2b7a82cb58
Is it because I use dc:elements
with a URI? I don't know.
Is it because I use
dc:elements
with a URI? I don't know.
No, this wasn't the problem, see https://test.skohub.io/build/?id=d824cf8f-d9d5-4dab-9659-0408aae249bf
It looks like there is a problem with the HCRT file at https://github.com/acka47/testing-skohub-vocabs/blob/master/hcrt.ttl. I just copied it over from the AMB org and added dc:title
and dc:creator
, so I guess this isn't working yet.
The error was that the SEO component did not receive the new title options and therefore could not build the <meta>
elements in the head
. I did not notice because that was not yet tested, but I added tests now.
I also added dc:description
for describing the concept scheme.
Maybe just another thing to shortly discuss. How do you suggest the new attributes to be displayed in the JSON? Currently it is:
"dctitle": {
"de": "Hochschulcampus Ressourcentypen",
"en": "Higher Education Resource Types",
"nl": "Brontypen voor het hoger onderwijs",
"uk": "Типи ресурсів вищої освіти"
},
"dcdescription": {
"de": "Eine Wertelliste für Typen von Lernressourcen (Learning Resource Type), entstanden im Kontext des Metadatenschemas \"LOM for Higher Education OER Repositories\" (https://w3id.org/dini-ag-kim/hs-oer-lom-profil/latest/)."
}
and in the context:
"dctitle": {
"@id": "http://purl.org/dc/elements/1.1/title",
"@container": "@language"
},
"dcdescription": {
"@id": "http://purl.org/dc/elements/1.1/description",
"@container": "@language"
},
which is all valid, but we did not talk about it. Is this ok for you? @acka47
How do you suggest the new attributes to be displayed in the JSON?
Do we have one uniform context for all concept schemes right now? Then I am for using Compact IRIs which works like prefixes in turtle:
"dc:title": {
"de": "Hochschulcampus Ressourcentypen",
"en": "Higher Education Resource Types",
"nl": "Brontypen voor het hoger onderwijs",
"uk": "Типи ресурсів вищої освіти"
},
"dc:description": {
"de": "Eine Wertelliste für Typen von Lernressourcen (Learning Resource Type), entstanden im Kontext des Metadatenschemas \"LOM for Higher Education OER Repositories\" (https://w3id.org/dini-ag-kim/hs-oer-lom-profil/latest/)."
}
and in the context:
{
"dc":"http://purl.org/dc/elements/1.1/",
"dct":"http://purl.org/dc/terms/",
"dc:title":{
"@id":"http://purl.org/dc/elements/1.1/title",
"@container":"@language"
},
"dc:description":{
"@id":"http://purl.org/dc/elements/1.1/description",
"@container":"@language"
}
}
An even nicer approach might be to generate the context separately for each vocab, we could map description
, title
, creator
etc. to dc
or dct
depending on what the published vocab uses.
Now everything is working fine and we'll only have to update the context as discussed above.
@acka47 Implemented changes as requested. And deployed to dev
for testing
The JSON looks good now but I encountered a problem with the HTML build. It is completing but the site does not work in the browser:
Added attributes as proposed in #284