upcoming changes to DataCite's policies for metadata

[Ryan Scherle]

DataCite is planning to implement more formal requirements for the DOIs we register, including minimum standards for metadata. See the forwarded message below for more details.

In general, more formal requirements will be good for Dryad. However, there is a slight complication. For data packages that are in "publication blackout" status, we want to register a DOI without the real metadata. These data packages need to have the full metadata hidden until the associated article is published. One possible solution is for us to submit dummy metadata values. Most importantly, we would need dummy values for the DataCite Creator field and Title field. 

Some possible values...

Creator: Dryad submitter (name temporarily restricted)

Title: Dryad data package <<insert DOI here>>

Does this make sense? Sound like a good idea? Elena suggested that it may be ok for us to publish the actual creator names, since there would be no way to link this information to a particular journal until the article was published. I'm erring on the side of caution, because I know how sensitive some of the journals are about releasing metadata early.

-- Ryan

Begin forwarded message:

From: Joan Starr <Joan....@UCOP.EDU>

Date: February 16, 2012 2:22:15 PM EST

To: EZI...@LISTSERV.UCOP.EDU

Subject: [EZID] upcoming changes to DataCite's policies afftecting EZID Service Guidelines

Reply-To: Joan Starr <Joan....@UCOP.EDU>

All,

 

In the next couple of months, DataCite will be adopting some new policies that will apply to any DOI registrations made through EZID. I will update the EZID Service Guidelines when these changes go into effect, but I wanted to give you advance notice so that you can ask any questions. I serve on the DataCite Business Policy Working Group, and I can bring your concerns forward.

 

Below are the key provisions that are likely to be of interest to you. Please note that the exact wording of these is still under development. I am not quoting a document, but rather I am trying to give you the general direction of the changes.

 

·         a landing page will be required if the dataset (or other object) being registered is restricted; the policy will provide guidelines as to what information the landing page must provide;

·         the DataCite required metadata must be supplied with registration;

·         agreement that objects being assigned DOIs are stored and managed such that persistent access to them can be provided (commitment to persistence);

·         and an understanding that the metadata deposited with DataCite is held under the terms of Creative Commons 0 (CC0).

 

Regarding the second item (requiring citation metadata with registration), I’d like to say a couple more things. First, providing metadata is very much in the best interests of increasing access and exposure—and the more the better. In addition to the required metadata, we encourage the provision of an abstract and also an indication of the resource type. The new user interface we are working on will make the provision of both of these pieces of information easy and straightforward.

 

Second, if you are interested in obtaining a DOI early in the research workflow—before metadata is available—EZID offers the option of creating the identifier but not registering it. For a good description of how this works, see the API documentation: http://n2t.net/ezid/doc/apidoc.html#identifier-status. This functionality is also available in the user interface.

 

As I mentioned, I’m interested to hear any questions or comments you have about these topics. The specifics will become available soon, but if you have concerns, please voice them now.

 

Thanks,

 

Joan

 

Joan Starr

EZID Service Manager and

Manager, Strategic and Project Planning

California Digital Library

University of California Office of the President

wk: 510-987-0469

twitter: joan_starr

 

[Ryan Scherle]

Mark,

We had considered this approach in our design for the publication blackout, but we decided that it was important for the DOIs to actually resolve. If we didn't register DOIs for items in the blackout, the article would be published with a non-functional data DOI. We are currently unable to get immediate notification when an article is published, which means there would always be a period during which the DOI did not resolve. The negative impact of Dryad DOIs that don't resolve (even temporarily) would be much worse than DOIs with less-than-perfect metadata.

--- Ryan

On Feb 17, 2012, at 11:25 AM, Mark Diggory wrote:

Ryan,

At this time we reserve (mint) the DOI in DSpace without registering it in datacite during the submission and workflow stages. It could be to case that it not be registered into datacite until the end of the publication blackout period as well. We would just extend the logic that uses local resolution in the user interface instead of proxy resolution if the item is in blackout.  When the embargo blackout ends, we can then complete the registration of the DOI and add the metadata.

Thoughts,

Mark

--
You received this message because you are subscribed to the Google Groups "dryad-dev" group.
To post to this group, send email to drya...@googlegroups.com.
To unsubscribe from this group, send email to dryad-dev+...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/dryad-dev?hl=en.

--

Mark Diggory (Schedule a Meeting) 2888 Loker Avenue East, Suite 305, Carlsbad, CA. 92010 Esperantolaan 4, Heverlee 3001, Belgium http://www.atmire.com

[Carl Boettiger]

Ryan,

What you propose makes sense to me.

Would it be possible to give the author the choice of what metadata is
visible at the same time that they choose to embargo or not?

At least there should be a clear message to the author indicating what
metadata will and will not be visible if they are embargoing the data.

Cheers,

Carl

> --
> You received this message because you are subscribed to the Google Groups
> "dryad-dev" group.
> To post to this group, send email to drya...@googlegroups.com.
> To unsubscribe from this group, send email to
> dryad-dev+...@googlegroups.com.
> For more options, visit this group at
> http://groups.google.com/group/dryad-dev?hl=en.

--
Carl Boettiger
UC Davis
http://www.carlboettiger.info/

[Vision, Todd J]

Ryan,

I agree that dummy metadata seems like a necessary evil for journals with blackouts.

Another issue is the one about abstracts. I don't think we are really in a position to say that these can be distributed under CCZero - their copyright status is vague. We should probably be adding a question to our integration questionnaire requesting pre-approval to distribute the abstracts to DataCite under those terms.

Todd

>> Mark Diggory (Schedule a Meeting)
>> 2888 Loker Avenue East, Suite 305, Carlsbad, CA. 92010
>> Esperantolaan 4, Heverlee 3001, Belgium
>> http://www.atmire.com
>>
>>
>
>

[Elena Feinstein]

Todd,

I'm in agreement about the dummy metadata.

As for the abstracts, we should consider whether to send them to DataCite at all, since they're not part of the required DataCite metadata. The rights issues surrounding article abstracts is something Peggy and I discussed with Kevin Smith when we were drafting the Terms of Use. Kevin's well-informed opinion is that the way we're using the abstracts would almost certainly fall under fair use, but we will never get a publisher to sign off on releasing them from copyright, and that we might be better off not raising the subject with them.

Elena

[Hilmar Lapp]

On Feb 17, 2012, at 12:16 PM, Carl Boettiger wrote:

> At least there should be a clear message to the author indicating what
> metadata will and will not be visible if they are embargoing the data.

It's not about those records for which the data are embargoed, if I understand correctly, but rather those that are embargoed in whole, i.e., for which the metadata are not be visible before a certain date as per policy of the journal.

-hilmar
--
===========================================================
: Hilmar Lapp -:- Durham, NC -:- informatics.nescent.org :
===========================================================

[Elena Feinstein]

Carl,

I agree that we should make it clearer to Dryad submitters exactly which parts of their submission (the metadata description, the data files) will be public and when.

Publication blackout involves hiding the entire Dryad metadata record until the associated article is published, and is something that's handled on a journal-by-journal basis and isn't a choice the submitting author can make. The current practice is that all "non-integrated" journals receive publication blackout, along with integrated journals that have requested this.

The embargoes preventing download of the actual data files, on the other hand, are selected by the submitting author. The only wrinkle to this is that integrated journals may choose to disallow embargoes for their authors, in accordance with their data archiving policies and requirements.

I hope that makes sense.

Elena

Ryan,

What you propose makes sense to me.

Would it be possible to give the author the choice of what metadata is
visible at the same time that they choose to embargo or not?

At least there should be a clear message to the author indicating what
metadata will and will not be visible if they are embargoing the data.

Cheers,

Carl

[Greenberg, Jane]

I am not sure I can add anything more to the conversation here, but agree w/Elena’s comment.

 

Elena and Peggy shared their communications with me too, and, we had previous situation at SILS, which led involving university lawyers b/c of a faculty members’s research using a significant number of abstracts from a system w/out seeking permission.  It is too sticky to deal with.  The only suggestion, and I think Todd notes this later, is to ask publishers to sign off on the use in joining Dryad, but this could also be a show-stopper.

 

Best wishes, jane