search

bdarcus / csln

Reimagining CSL
Mozilla Public License 2.0
12 stars 0 forks source link

Refactor subtitles and SimpleName Contributor #91

Closed bdarcus closed 1 year ago

bdarcus commented 1 year ago

Allows:

title:
  main: A title
  sub: A subtitle # can also be a vector of strings
author:
  name: United Nations
title: Some string title # titles can also be simple strings
publisher: # as with all contributors, this can also be a vector of more than one
  name: ABC Books
  location: New York

I may also want to add:

access:
  url: ...
  date: 2020-10-11

Close #90

Any feedback on this @denismaier, before I merge it?

denismaier commented 1 year ago

I'd say this looks reasonable. What else would you add under access? DOIs for sure, PMIDs also. Other things?

bdarcus commented 1 year ago

I'd say this looks reasonable. What else would you add under access? DOIs for sure, PMIDs also. Other things?

I was thinking could also put stuff for physical archives?

denismaier commented 1 year ago

I was thinking could also put stuff for physical archives?

Ok. But maybe this should work recursively then. (A document in an archive may have a physical location, a location in the archive, and a DOI)

denismaier commented 1 year ago

Oh, and DOIs are top level in your example bib.

bdarcus commented 1 year ago

maybe this should work recursively then. (A document in an archive may have a physical location, a location in the archive, and a DOI)

Good point; but maybe not "recursively"; just:

access:
  - physicalLocation: foo
  - doi: bar

Aside: thinking again about FRBR, but wanting to avoid that complexity :-)

I'm not sure, however, if a DOI actually belongs there.

It's a globally-unique identifier, that happens to have a mechanism to resolve to a web resource.

The only reason we do the access date of online resources is because the URIs are more ephemeral?

bdarcus commented 1 year ago

Oh, and DOIs are top level in your example bib.

Right; I had in mind the identifiers to be top-level.

But maybe this needs more thought; I hadn't actually yet added the identifiers (so now am confused why the example chicago file validates!).

As a general matter, I'm trying to design this so it can be as simple as possible, but still work well. That includes allowing simple strings on titles, contributors, and dates, but also more structure where needed.

bdarcus commented 1 year ago

I merged this; will defer the identifiers/access to a separate PR. If you have any other thoughts, feel free to add them here in the meantime.