Open agarny opened 3 years ago
Only suggestion might be repository
rather than PMR workspace
.
@tgbugs - are these the datacite relationship types that you mentioned are part of the SDS 2.0: https://support.datacite.org/docs/relationtype_for_citation. And if so, these terms would be used to identify the relationship between the dataset and the resource being linked to? (rather than the current set which are more like mime-types)
Wouldn't repository
(or Repository
) be better suited to reference a whole repository as such (e.g., https://models.physiomeproject.org/)?
a workspace is a standalone (git) repository - but you could also imagine a repository on, for example, github or biomodels. With this approach there is no technical restriction to only having models from PMR...but there is likely a curation restriction to having models from known reliable and permanent sources.
Oh, ok, repository as in Git repository, as opposed to Physiome Model Repository. Fair enough.
(I have updated my simulation dataset and now use Repository
rather than PMR workspace
for https://models.physiomeproject.org/workspace/486.)
@nickerso yes, those are the relation types I was referring to. They function as you describe. There are also resource types and resource type general which function like the mimetypes that are used in the manifest file additional types
column.
Background: we have a (simulation) dataset that contains some additional URLs (stored in
dataset_description.xlsx
using theAdditional Links
field). To distinguish those URLs, we use the corresponding information stored in theLink Description
field, i.e.Request: the above might be fine, but the information available in the
Link Description
field should be formalised and we would like to suggest the following terms:PMR workspace
;CellML file
;SED-ML file
;COMBINE archive
; andRDF file
.@nickerso, would you agree with those terms?