Add support for wikidata HAL_ID P7864

[Futur3r]

It adds the key HALID in the extra field from wikidata items with P7864.

[AbeJellinek]

@dstillman @adam3smith: Should HAL Archives Ouvertes be a search translator and support searching by these? We could also add support to Add by Identifier. They look reasonably detectable, although I haven't found documentation - a prefix (usually, but not always, hal), a dash, a numeric ID, and optionally v followed by a version number (like arXiv).

[adam3smith]

@fflamerie @nchachereau -- I hope you don't mind me tagging you here: see the question about HAL identifiers above. Are these sufficiently widely used in France to make sense to add to search by identifier? Are you aware of any documentation on them and is there a list of prefixes uses beyond hal-?

[fflamerie]

I don't have a strong opinion about the usage question. @symac have you got one?
I am not aware of any other documentation than the API one : https://api.archives-ouvertes.fr/docs/search/?schema=field-types#field-types. It does not include a list of the prefixes used beyond -hal. I can't find this list on HAL general documentation, I guess it should be available upon request from the CCSD @rtournoy.

[rtournoy]

For HAL we have 32 prefixes for identifiers:
anses-
artxibo-
bioemco-
cea-
cirad-
dumas-
edutice-
emse-
ensl-
hal-
halshs-
hceres-
hprints-
ijn_
in2p3-
ineris-
inserm-
insu-
ird-
irsn-
lirmm-
mem_
meteo-
mnhn-
obspm-
pastel-
pasteur-
sfo-
sic_
ssa-
tel-
ujm-

New portail are using the default prefix hal- so no more prefixes will be created in the future.

[rtournoy]

You might also like this regex https://github.com/inveniosoftware/idutils/blob/master/idutils/__init__.py#L238 that was added to Invenio

[Futur3r]

Are these sufficiently widely used in France to make sense to add to search by identifier?

The french state is pushing scientists to open science, and publishing on HAL is a part of it.
There is HAL referents in universities that do training and are pushing PhD students to consider put their publication on HAL.
It is a great alternative to ResearchGate for example.
So it is already used, and used more and more.

[dstillman]

It's not about use of HAL itself, though. It's about whether HAL identifiers are sufficiently used such that someone would think or expect to use a HAL ID in Add Item by Identifier in Zotero.

[Futur3r]

Ah, sorry.
On that note, I guess no, they are not used. Maybe it's because there is no service to (easily) query a paper via it's HAL ID already, but if it's possible via Zotero, there will be at least one.
AND, as a contributor of wikidata, I contribute systematically HAL IDs to Wikidata when they are available (that's why I made those PR). And a lot of publications on HAL doesn't have a DOI, so the only stable identifier is their HAL ID.
Adding the functionality would probably reinforce the use of HAL and so, the use of those IDs.

Also, now PhD thesis are published on HAL and I think it's the only place to make them publicly available (with the tel- prefix at https://theses.hal.science/). So when a thesis is available on HAL, it's HAL ID might be it's only "open access stable ID".

[adam3smith]

And a lot of publications on HAL doesn't have a DOI, so the only stable identifier is their HAL ID. Adding the functionality would probably reinforce the use of HAL and so, the use of those IDs.

Also, now PhD thesis are published on HAL and I think it's the only place to make them publicly available (with the tel- prefix at https://theses.hal.science/). So when a thesis is available on HAL, it's HAL ID might be it's only "open access stable ID".

I don't have a strong opinion, but I think that's a pretty good case to add them. Allowing retrieval by ID where not other PIDs are available seems like a good use, especially since we already have a regex we can use and the costs are thus minimal.

[rtournoy]

Disclaimer: I am not impartial. Sometimes the identifier is the only way to find an open access version of a document that would otherwise only be available with a DOI behind a paywall. Those identifiers are actually used outside HAL for instance with Unpaywall, RePEc, OpenAIRE and many other open science tools. Also, at least in France, universities and other public institutions are training researchers to use both Zotero and HAL.

[dstillman]

Allowing retrieval by ID where not other PIDs are available seems like a good use, especially since we already have a regex we can use and the costs are thus minimal.

The main cost is just that trying to communicate what can be entered makes the input dialogs (Zotero, web library, ZoteroBib) too unwieldy. So we can add this, but we'll need to find another way to list some of the more esoteric identifiers. E.g., in Zotero, "Enter DOIs, ISBNs, or other identifiers to add to your library", where "other identifiers" is underlined and hovering over it shows a popup with a list of all the formats. We can do something similar in web library. Not sure what we would do on ZoteroBib, but we're already missing ADS Bibcodes there.

[adam3smith]

E.g., in Zotero, "Enter DOIs, ISBNs, or other identifiers to add to your library", where "other identifiers" is underlined and hovering over it shows a popup with a list of all the formats.

Agree, yes, that'd be the way to go. I'd assume we get some more. E.g. we already have an ERIC ID search translator although that's not currently available via the add by identifier dialog (but could/should easily be).