Combination of DOIs (for datasets) and Handles (for files) at same time?

[Volodymyr Kushnarenko]

Dear Dataverse developers team,

could it be possible to configure Dataverse so, that for datasets you get DOIs and for each file inside the dataset you get a Handle. The idea is to reference datasets with DOIs (because of metadata on DataCite, to have a good findability) and all files inside the dataset to reference with Handles (to reduce costs, because a large amount of files can increase subscription price in case of DOIs).

I think, currently is possible to use either DOI or Handle, combination of both types as described above is not possible. But I'm also not sure, if such a feature will be used by the majority of the community (definitely you already had such a discussion). And how difficult could be an implementation (a logic complexity will be definitely increased...), to open an issue for that.

I have this question, because I don't know, how differs a findability in case of DOIs and Handles, if additionally stored metadata fields on the DataCite (DOI) really increase it compare to Handles without metadata. E.g. when user tries to find a "title" via Google, would the DataCite "title" metadata field (DOI) really help in this case or not? If findability is same, then for a big amount of files we can simply use Handles-only solution without extra complexity.

Many thanks for your response in advance,

Vladimir

[Philip Durbin]

Hi Vladimir,

As you suspect, this is not currently possible. You are very welcome to create a GitHub issue for this (please reply on the list with a link if you do).

As for how much work it would be, it's hard to say. As you know, currently you can only configure one PID provider at a time. When we estimate issues we use trivial, small, medium, and large. I'd probably vote large on this one, mostly due to uncertainty. Others may feel differently.

With regard to findability, I feel like DOIs are more findable than Handles because they're supported by DataCite, including their Make Data Count initiative (which Dataverse supports). But I could be wrong about that. I'm happy to hear what others think.

I hope this helps,

Phil

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-commu...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dataverse-community/644bf687-8ba2-4ad5-b727-0a5f0ee2068en%40googlegroups.com.

--

Philip Durbin
Software Developer for http://dataverse.org
http://www.iq.harvard.edu/people/philip-durbin