DOI and Handle in future Dataverse versions?

220 views
Skip to first unread message

ofu...@gmail.com

unread,
Jan 19, 2017, 5:57:22 AM1/19/17
to Dataverse Users Community
Hi,
We have datasets with only Handle after migration from Dataverse 3.x  to 4.x. We would like to register/generate DOI for these datasets. Is there still any plan in future to extend support for both DOI and Handle in coming Dataverse versions?

Obi

Philip Durbin

unread,
Jan 19, 2017, 8:43:11 AM1/19/17
to dataverse...@googlegroups.com
Hi, Obi! I want to make sure I understand your goal. In your installation of Dataverse you have a mix of datasets where some have DOIs and some have Handles (so do we in the Harvard Dataverse, but like in your installation, all new datasets get a DOI). Here's an example of one of your datasets with a Handle: https://opendata.uit.no/dataset.xhtml?persistentId=hdl:10037.1/10121

Are you saying that you'd like to have a DOI for that dataset? You want the dataset to have both a Handle and a DOI? I assume you'd want the Handle to continue to work/resolve, especially if it's been referenced in a published article. But you want the dataset to have a DOI too? There's a field called "Other ID" you could populate. It's defined as "Another unique identifier that identifies this Dataset (e.g., producer's or another repository's number)." However, the citation only can have a single persistent identifier, either a DOI or a Handle*.

I hope this helps!

Phil

* The persistent id is stored in the database at the dataset level across multiple fields (protocol, authority, identifier) for anyone who is interested: http://phoenix.dataverse.org/schemaspy/latest/tables/dataset.html

--
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-community+unsub...@googlegroups.com.
To post to this group, send email to dataverse-community@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dataverse-community/13f0cae3-b8e7-462a-a681-56a5a6be2719%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.



--

ofu...@gmail.com

unread,
Jan 23, 2017, 7:59:00 AM1/23/17
to Dataverse Users Community, philip...@harvard.edu
Hi Philip,
Yes, we want datasets with Handle to have DOI too. I have seen "Other ID".  But is there any script to automatically generate DOI value on "Other ID" ?


Obi


torsdag 19. januar 2017 14.43.11 UTC+1 skrev Philip Durbin følgende:
Hi, Obi! I want to make sure I understand your goal. In your installation of Dataverse you have a mix of datasets where some have DOIs and some have Handles (so do we in the Harvard Dataverse, but like in your installation, all new datasets get a DOI). Here's an example of one of your datasets with a Handle: https://opendata.uit.no/dataset.xhtml?persistentId=hdl:10037.1/10121

Are you saying that you'd like to have a DOI for that dataset? You want the dataset to have both a Handle and a DOI? I assume you'd want the Handle to continue to work/resolve, especially if it's been referenced in a published article. But you want the dataset to have a DOI too? There's a field called "Other ID" you could populate. It's defined as "Another unique identifier that identifies this Dataset (e.g., producer's or another repository's number)." However, the citation only can have a single persistent identifier, either a DOI or a Handle*.

I hope this helps!

Phil

* The persistent id is stored in the database at the dataset level across multiple fields (protocol, authority, identifier) for anyone who is interested: http://phoenix.dataverse.org/schemaspy/latest/tables/dataset.html
On Thu, Jan 19, 2017 at 5:57 AM, <ofu...@gmail.com> wrote:
Hi,
We have datasets with only Handle after migration from Dataverse 3.x  to 4.x. We would like to register/generate DOI for these datasets. Is there still any plan in future to extend support for both DOI and Handle in coming Dataverse versions?

Obi

--
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-community+unsub...@googlegroups.com.
To post to this group, send email to dataverse...@googlegroups.com.

Philip Durbin

unread,
Jan 24, 2017, 10:28:26 AM1/24/17
to dataverse...@googlegroups.com
Well, you could script the addition of any metadata field to a dataset using the "native" API. Here's an example script: https://github.com/IQSS/dataverse/blob/v4.6/scripts/search/tests/edit-dataset-finch1 . The approach would be to download the dataset as JSON, make changes to that JSON, and then re-upload the JSON to Dataverse. The example script above is a simple replace of the title, but again, you could add more JSON to add any metadata field. Here's an example of how "otherId" looks as JSON (from http://dx.doi.org/10.7910/DVN/GSVMEI ):

$ curl -s https://dataverse.harvard.edu/api/datasets/:persistentId?persistentId=doi:10.7910/DVN/GSVMEI | jq '.data.latestVersion.metadataBlocks.citation.fields[] | select(.typeName=="otherId")'

{
  "typeName": "otherId",
  "multiple": true,
  "typeClass": "compound",
  "value": [
    {
      "otherIdAgency": {
        "typeName": "otherIdAgency",
        "multiple": false,
        "typeClass": "primitive",
        "value": "mEDRA"
      },
      "otherIdValue": {
        "typeName": "otherIdValue",
        "multiple": false,
        "typeClass": "primitive",
        "value": "http://dx.medra.org/10.7336/academicus.2017.15.11"
      }
    }
  ]
}

I hope this helps!

Phil

p.s. This reminds me of https://github.com/IQSS/dataverse/issues/2110 which is where I went for the jq syntax.


On Mon, Jan 23, 2017 at 7:59 AM, <ofu...@gmail.com> wrote:
Hi Philip,
Yes, we want datasets with Handle to have DOI too. I have seen "Other ID".  But is there any script to automatically generate DOI value on "Other ID" ?


Obi

torsdag 19. januar 2017 14.43.11 UTC+1 skrev Philip Durbin følgende:
Hi, Obi! I want to make sure I understand your goal. In your installation of Dataverse you have a mix of datasets where some have DOIs and some have Handles (so do we in the Harvard Dataverse, but like in your installation, all new datasets get a DOI). Here's an example of one of your datasets with a Handle: https://opendata.uit.no/dataset.xhtml?persistentId=hdl:10037.1/10121

Are you saying that you'd like to have a DOI for that dataset? You want the dataset to have both a Handle and a DOI? I assume you'd want the Handle to continue to work/resolve, especially if it's been referenced in a published article. But you want the dataset to have a DOI too? There's a field called "Other ID" you could populate. It's defined as "Another unique identifier that identifies this Dataset (e.g., producer's or another repository's number)." However, the citation only can have a single persistent identifier, either a DOI or a Handle*.

I hope this helps!

Phil

* The persistent id is stored in the database at the dataset level across multiple fields (protocol, authority, identifier) for anyone who is interested: http://phoenix.dataverse.org/schemaspy/latest/tables/dataset.html
On Thu, Jan 19, 2017 at 5:57 AM, <ofu...@gmail.com> wrote:
Hi,
We have datasets with only Handle after migration from Dataverse 3.x  to 4.x. We would like to register/generate DOI for these datasets. Is there still any plan in future to extend support for both DOI and Handle in coming Dataverse versions?

Obi

--
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-community+unsubscribe...@googlegroups.com.

--
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-community+unsub...@googlegroups.com.
To post to this group, send email to dataverse-community@googlegroups.com.
 To view this discussion on the web visit https://groups.google.com/d/msgid/dataverse-community/21862ecd-db7a-4913-b66c-25ed8af99d1b%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

Amber Leahey

unread,
Jan 25, 2017, 10:49:49 AM1/25/17
to Dataverse Users Community, philip...@harvard.edu
Sorry for jumping on the thread, we have the same kind of situation right now but it isn't ideal. Specifically we'd like to have the choice to assign a handle to dataset that we don't have "authority" over, or, there are multiple copies of that data across various access repositories (e.g. licensed data for example). So we need support for both handle and doi as a PID in the same installation, but only need one PID per dataset. 

What do you think will be possible in the future for supporting the choice between doi and handle upon upload in the same system? Is this something you've thought about or need as well?

Best, 
Amber 
To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-community+unsub...@googlegroups.com.

--
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-community+unsub...@googlegroups.com.
To post to this group, send email to dataverse...@googlegroups.com.

Durand, Gustavo

unread,
Jan 25, 2017, 11:17:58 AM1/25/17
to dataverse...@googlegroups.com, Durbin, Philip
Hi all, 

I know we have discussed allowed support of multiple persistent id providers in one installation, but it is not currently on the roadmap.

Amber, in your case, I'd be curious to know, is it something you might be able to solve with harvesting? That is, do you need a full copy of the data (i.e. files) on your Dataverse, or could it just harvest the info from one of the other sources?

Gustavo

On Wed, Jan 25, 2017 at 10:49 AM, Amber Leahey <amber...@gmail.com> wrote:
Sorry for jumping on the thread, we have the same kind of situation right now but it isn't ideal. Specifically we'd like to have the choice to assign a handle to dataset that we don't have "authority" over, or, there are multiple copies of that data across various access repositories (e.g. licensed data for example). So we need support for both handle and doi as a PID in the same installation, but only need one PID per dataset. 

What do you think will be possible in the future for supporting the choice between doi and handle upon upload in the same system? Is this something you've thought about or need as well?

Best, 
Amber 

On Tuesday, 24 January 2017 10:28:26 UTC-5, Philip Durbin wrote:

--
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-community+unsub...@googlegroups.com.
To post to this group, send email to dataverse-community@googlegroups.com.

Amber Leahey

unread,
Jan 25, 2017, 4:59:33 PM1/25/17
to Dataverse Users Community, philip...@harvard.edu
Well it is a bit more complicated since we want to be the ones to provide access to these licensed content on our own platform for our own users. In some cases there may not be an alternative and some data providers seem slow to adopt DOIs not to mention accessing the data is inconsistent across providers, some don't even have a mechanism. 

Coordinating this with other repositories and data providers is the ideal solution, agreed, but this will take a lot of time and effort and collaboration, especially to ensure the same level of access and service around delivering that data to end-users. For these licensed datasets a DOI might not be appropriate, perhaps a Handle might be better for our access copies. Does that make sense?

I think it is something to consider especially as some libraries begin to use Dataverse for delivering a variety of data, including licensed content for example. Adhering to DOI best practices is a bit difficult at the moment...

Thanks, 
Amber

To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-community+unsub...@googlegroups.com.

--
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-community+unsub...@googlegroups.com.
To post to this group, send email to dataverse...@googlegroups.com.

--
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-community+unsub...@googlegroups.com.
To post to this group, send email to dataverse...@googlegroups.com.

Eugene Barsky

unread,
Jan 26, 2017, 9:30:25 AM1/26/17
to Dataverse Users Community, philip...@harvard.edu, Amber Leahey
Folks:

You may already know, but here @UBC Dataverse we have achieved this flexibility of assigning multiple persistent identifiers. 

For licensed data sets, we mint handles that are built-in into the Dataverse software.

For open access research data sets we mint DOIs (e.g. https://dx.doi.org/10.14288/1.0314339), however, we do not mint them in the native Dataverse interface but in our discovery platform - Open Collections - https://open.library.ubc.ca/, which collates four UBC repositories into one interactive interface with a robust API.

This allows us to be quite flexible with persistent ids. We could also issue ARKs, PURLs, etc as needed. 

Happy to answer any questions.

Eugene


To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-community+unsub...@googlegroups.com.

--
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-community+unsub...@googlegroups.com.
To post to this group, send email to dataverse...@googlegroups.com.

--
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-community+unsub...@googlegroups.com.
To post to this group, send email to dataverse...@googlegroups.com.

Philip Durbin

unread,
Jan 26, 2017, 9:47:19 AM1/26/17
to dataverse...@googlegroups.com
Eugene, this is a very interesting example. Thanks! I see you're still using DVN 3 and can only assume you're waiting for Handle support in Dataverse 4 ( https://github.com/IQSS/dataverse/issues/2437 ) before you upgrade to Dataverse 4 so that you can continue to mint Handles using Dataverse. I'm assuming you don't have a need to mint both Handles and DOIs from Dataverse. I don't think there's an issue tracking this but there is a more specific issue called "Support Registering Multiple Prefixes for DataCite DOIs In an Installation" at https://github.com/IQSS/dataverse/issues/2943 that was opened following discussion with Alan Darnell from Scholars Portal at https://groups.google.com/d/msg/dataverse-community/7-5NsSA6iGg/7kG4RR_VEQAJ (Amber, I'm wondering if there's a GitHub issue that captures Scholars Portal's latest thinking in this area). Sorry to ramble. I feel like there are various user stories here and I'm not sure they're each being clearly tracked in GitHub issues. Maybe they are and I'm just not sure which issue is which. :)

Phil

To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-community+unsubscribe...@googlegroups.com.

--
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-community+unsubscribe...@googlegroups.com.

To post to this group, send email to dataverse...@googlegroups.com.

--
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-community+unsubscribe...@googlegroups.com.

To post to this group, send email to dataverse...@googlegroups.com.

--
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-community+unsub...@googlegroups.com.
To post to this group, send email to dataverse-community@googlegroups.com.
 To view this discussion on the web visit https://groups.google.com/d/msgid/dataverse-community/126ed831-61e6-4aa0-b9b5-a24f01693d7a%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

Eugene Barsky

unread,
Jan 26, 2017, 10:30:16 AM1/26/17
to dataverse...@googlegroups.com
Phil:

We support Scholars Portal interest in assigning multiple prefixes, as we also support a few institutions. right now, we assign UBC DOIs to all data sets. 

Definitely, an interesting area to explore. And yes, we are still on 3.6 for a number of reasons, handles included :)

E.




To post to this group, send email to dataverse-community@googlegroups.com.

--
You received this message because you are subscribed to a topic in the Google Groups "Dataverse Users Community" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/dataverse-community/E_nxXqH68NE/unsubscribe.
To unsubscribe from this group and all its topics, send an email to dataverse-community+unsub...@googlegroups.com.

To post to this group, send email to dataverse-community@googlegroups.com.

Eugene Barsky

unread,
Jan 26, 2017, 2:47:28 PM1/26/17
to Dataverse Users Community, philip...@harvard.edu
BTW, our Open Access code is open access and is available here - https://github.com/ubc-library/open-collections-discovery-ui

It is also being adopted for the Canadian National Federated Research Data Repository as a search interface.

E.

On Thu, Jan 26, 2017 at 9:08 AM, Amber Leahey <amber....@utoronto.ca> wrote:

Thanks, Eugene.

 

This is great to hear, and hopefully we can offer a similar type of service within Dataverse in the future. We don’t control the DOI minting now that it is incorporated into the Dataverse system with version 4, as you know.

 

Some option for handles or DOIs, at the dataset level, would improve the system I think.

 

Any other thoughts?

 

Best,

Amber

Philip Durbin

unread,
Feb 3, 2017, 6:01:09 PM2/3/17
to dataverse...@googlegroups.com
Amber, thanks for opening https://github.com/IQSS/dataverse/issues/3623

Phil

To post to this group, send email to dataverse-community@googlegroups.com.

ofu...@gmail.com

unread,
Feb 13, 2017, 5:03:51 AM2/13/17
to Dataverse Users Community, philip...@harvard.edu
Thanks Phil.

- Obi
To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-community+unsub...@googlegroups.com.

--
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-community+unsub...@googlegroups.com.
To post to this group, send email to dataverse...@googlegroups.com.

Philipp at UiT

unread,
Oct 12, 2017, 6:44:57 AM10/12/17
to Dataverse Users Community
I would like to return to the original question posed by my colleague Obi:

Many datasets in our Dataverse installation were created before Dataverse switched over to DOI. Those datasets therefore use handle as their PID. We would like to mint DOIs for those datasets. As Phil has suggested we could place a DOI in the metadata field "Other ID", which is defined as "Another unique identifier that identifies this Dataset (e.g., producer's or another repository's number)." Phil also provided a script for how this field could be populated with DOIs.

But, our main problem/question is how to *mint* DOIs for existing handle datasets, thus datasets that already are created (and most of them also published) in a Dataverse installation, and that use handle as their PID. DataCite Metadata Store provides a service called Register new Dataset (cf. https://mds.datacite.org/datasets?form) Has anyone been able to mint DOIs for existing handle dataset created in Dataverse using this service?

Best,

Jonathan Crabtree

unread,
Oct 12, 2017, 10:23:48 AM10/12/17
to dataverse...@googlegroups.com

We have the same issue here at Odum. Right now we continue to pay our Handle subscription in addition to EZiD

 

I guess when we migrate to Datacite would be a good opportunity to solve this.

 

Jon

 

 

From: <dataverse...@googlegroups.com> on behalf of Philipp at UiT <uit.p...@gmail.com>
Reply-To: <dataverse...@googlegroups.com>
Date: Thursday, October 12, 2017 at 6:44 AM
To: Dataverse Users Community <dataverse...@googlegroups.com>
Subject: Re: [Dataverse-Users] DOI and Handle in future Dataverse versions?

 

I would like to return to the original question posed by my colleague Obi:

To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-commu...@googlegroups.com.

--
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.


For more options, visit https://groups.google.com/d/optout.

--

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 post to this group, send email to dataverse...@googlegroups.com.

Eugene Barsky

unread,
Oct 12, 2017, 11:39:13 AM10/12/17
to Dataverse Users Community
Hello 
​
Philipp:

Here in UBC, we have been minting DOIs on top of handles since the very beginning. We have been using Datacite DOIs to do that programmatically. We use a discovery layer - Open Collections (home developed in UBC) to mint DOIs to all UBC digital assets (DSpace, CONTENTdm, Dataverse and AtoM), more than 250,000 items in total. 

I feel that Datacite solution will be the best fit for your problem.

Eugene




To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-community+unsubscribe...@googlegroups.com.

--
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-community+unsubscribe...@googlegroups.com.

To post to this group, send email to dataverse...@googlegroups.com.

--
You received this message because you are subscribed to a topic in the Google Groups "Dataverse Users Community" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/dataverse-community/E_nxXqH68NE/unsubscribe.
To unsubscribe from this group and all its topics, send an email to dataverse-community+unsub...@googlegroups.com.
To post to this group, send email to dataverse-community@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dataverse-community/63973de7-a6d5-4dd9-8266-8d79669c30dc%40googlegroups.com.

Eugene Barsky

unread,
Oct 12, 2017, 11:46:21 AM10/12/17
to Dataverse Users Community
I should also say that we point our DOIs in Open Collections (here is an example - https://dx.doi.org/10.14288/1.0319861) to the handles in our Dataverse. We love this solution as it allows us a flexible minting technique, where we do not issue DOIs to the licensed datasets we maintain, but only to open research datasets. Our Dataverse is mostly populated by licensed data. 

E.

To unsubscribe from this group and all its topics, send an email to dataverse-community+unsubscribe...@googlegroups.com.

To post to this group, send email to dataverse-community@googlegroups.com.

danny...@g.harvard.edu

unread,
Oct 12, 2017, 3:05:33 PM10/12/17