Import work from DOI missing contributor data

104 views
Skip to first unread message

Sheila Rabun

unread,
May 5, 2020, 5:08:35 PM5/5/20
to ORCID API Users

When a user imports a work via DOI, how does that work – where is the imported data coming from? for example, on this ORCID record: https://orcid.org/0000-0002-3254-4913 - the imported work actually has multiple authors but the authors/contributors data is not appearing in the work entry in ORCID. Why is that the case?


THanks!
Sheila

Liz Krznarich

unread,
May 5, 2020, 5:58:19 PM5/5/20
to ORCID API Users
Hi Sheila,

The way that import from DOI (or PMID, arXiv ID) works is:

1. User enters an identifier
2. We query relevant the relevant API
3. We populate the Add Works form with data from the response
4. (Optional) User can edit the data to their liking
5. User saves the work, and it's stored with the user as the source

Under the hood, add by DOI is a way to pre-populate the Add Works form (not to import a work directly into the record, as with import wizards).
The reason that there are no contributors is that the Add Works form does not support contributors (except in the citation field), vs the API, which does support contributors. 

Early on in ORCID's history, we allowed users to manually enter contributor names in the Add Works form, but it caused quite a lot of confusion and is really at odds with some of our core goals. That said, we know this is in high demand, and our product team has adding contributors back to the Add Works interface as one of its top goals for 2020.

Cheers,
Liz
---
Liz Krznarich 
Tech Lead, New Projects, ORCID
e.krz...@orcid.org
https://orcid.org/0000-0001-6622-4910

Antonin Delpeuch (lists)

unread,
May 5, 2020, 6:18:18 PM5/5/20
to orcid-a...@googlegroups.com
I am curious about why letting users input contributors metadata is "at
odds with some of [your] core goals"?

I have always found it curious that a service designed to solve the
author ambiguity problem does not actually store author metadata in its
bibliographic references (or at least not via the Add Works form). Why
is it so?

Given a paper, I want to be able to identify its authors (with ORCID
ids)… but for that I need an authors list. You could argue that the list
of authors should in principle be obtained by looking for all ORCID
profiles containing that work. But that is really impractical: in
general not all authors have an ORCID profile, and you would not be able
to retrieve the list of authors in the correct order (which is very
important is some fields).

Most other services (such as Crossref) have got this right: for each
paper, they include a list of authors, with ORCID ids for some of them.

It is great that you are trying to address this issue this year, but I
am hoping to understand one day why this very puzzling design choice was
made in the first place.

Cheers,
Antonin
> --
> You received this message because you are subscribed to the Google
> Groups "ORCID API Users" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to orcid-api-use...@googlegroups.com
> <mailto:orcid-api-use...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/orcid-api-users/14ef7b9c-1509-442c-96e0-8d0cf26ca51a%40googlegroups.com
> <https://groups.google.com/d/msgid/orcid-api-users/14ef7b9c-1509-442c-96e0-8d0cf26ca51a%40googlegroups.com?utm_medium=email&utm_source=footer>.

Liz Krznarich

unread,
May 5, 2020, 6:41:58 PM5/5/20
to Antonin Delpeuch (lists), orcid-a...@googlegroups.com
Hi Antonin, 

Our goal is to enable authors to be uniquely identified (disambiguated) and connected to their contributions (works being one type of contribution). While we agree that a full author list is quite useful, allowing users to input non-disambiguated names as free text is the part that does not serve that goal. 
  
As you have pointed out, when authenticated iDs are collected during the publication process and works are populated by a source such as CrossRef using our API, populating contributors and their iDs in the work metadata added to ORCID makes sense - for the most part we can be relatively sure that the names and ORCID iDs are accurate.

This is not necessarily the case if names and (especially) ORCID iDs are entered manually into the Add Works UI. As a service focused on disambiguating authors and providing an accurate record of their contributions, we must come up with a method of supporting contributors for manually entered works that assures accuracy - which is a non-trivial task!

In the meantime, if you need a complete list of authors for a given publication, the best approach is to retrieve the its source metadata from the relevant API using the publication’s external identifier found in ORCID (which, incidentally, is one possible approach we may consider for displaying contributor info in ORCID).

Cheers,
Liz




To unsubscribe from this group and stop receiving emails from it, send an email to orcid-api-use...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/orcid-api-users/08acbc5f-dc3f-6564-bb3c-7b9c443ba1cc%40antonin.delpeuch.eu.

Antonin Delpeuch (lists)

unread,
May 6, 2020, 2:31:41 AM5/6/20
to orcid-a...@googlegroups.com
Hi Liz,

The problem is that the Add Works dialog is typically used for works
which do not have DOIs or other ids, so as a consumer, fetching the
authors list via some API is not an option.

Authors list are useful even if they are not accompanied by ids, and I
do not see why ORCID's UI would not allow authors to input their
co-authors as accurately as when submitting an article to a publisher.
On the contrary, this would be a great opportunity to encourage them to
match the authors they add to existing ORCID profiles.

The fact that the authors list was input manually on ORCID than fetched
from another service is already reflected by the provenance metadata you
store in works anyway.

Best,
Antonin

That being said
>>> e.krz...@orcid.org <mailto:e.krz...@orcid.org>
>> https://groups.google.com/d/msgid/orcid-api-users/08acbc5f-dc3f-6564-bb3c-7b9c443ba1cc%40antonin.delpeuch.eu.
>
> --
> You received this message because you are subscribed to the Google
> Groups "ORCID API Users" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to orcid-api-use...@googlegroups.com
> <mailto:orcid-api-use...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/orcid-api-users/17486965-C68E-458C-84B6-FFC1CB7B2D0D%40orcid.org
> <https://groups.google.com/d/msgid/orcid-api-users/17486965-C68E-458C-84B6-FFC1CB7B2D0D%40orcid.org?utm_medium=email&utm_source=footer>.

John Aspler

unread,
May 6, 2020, 12:42:42 PM5/6/20
to ORCID API Users
Hi Liz,

I'm also curious about the difference between adding Works via "Add DOI" versus the Crossref Search & Link Wizard. I have a few screenshots below from my own ORCID Record displaying the difference.

1) When I use the Crossref S&L Wizard, it imports a bibtex citation with contributors information:

ORCID Crossref Metadata Search.png



2) However, when adding a DOI manually - even when that DOI can be found in the Crossref Metadata Search Wizard - it imports different / less complete metadata

ORCID Add DOI.png



Is there some way that 'Add DOI' could be made to import the bibtex citation too?

Thanks!

John

Jason Ronallo

unread,
May 6, 2020, 12:57:21 PM5/6/20
to John Aspler, ORCID API Users
Having the DOI import allow for adding more information from the DOI service like Crossref would be nice. Note though that even then some information may be missing. We find that the metadata for quite a few DOIs only include one of the co-authors and on occasion none of the authors are included.

Jason

--
You received this message because you are subscribed to the Google Groups "ORCID API Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to orcid-api-use...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/orcid-api-users/8bc7e9af-f809-4826-a70b-31070d0ea1e9%40googlegroups.com.


--
Jason Ronallo
Department Head, Digital Library Initiatives
North Carolina State University Libraries

All electronic mail messages in connection with State business which are sent to or received by this account are subject to the NC Public Records Law and may be disclosed to third parties.

Arthur Smith

unread,
May 6, 2020, 2:09:49 PM5/6/20
to orcid-a...@googlegroups.com
Having done some work on author disambiguation...

One issue I think is that ORCID doesn't track WHICH author in an author
list is the identified author. Matching names with strings is pretty
nontrivial, and there are many papers with two or more names that could
match the same person (sometimes the string values are identical and
authors only distinguished by their affiliation listing in the author
list). So to properly support this I think an additional "author number"
value needs to be recorded. Getting that right may be complicated though!

   Arthur Smith

Liz Krznarich

unread,
May 6, 2020, 5:03:05 PM5/6/20
to ORCID API Users
Hi John, 

Check out my reply to the original question about the differences between those 2 methods https://groups.google.com/d/msg/orcid-api-users/d_N3c89etcw/l_C81eXOAQAJ
The difference is that the API (which is used by import wizards) supports including contributors but the UI Add Works form does not. Add by ID is essentially a way to pre-populate the UI Add Works form, thus the end result is that contributors are not included, even if they exist in the DOI, ArXiv, PMID, etc metadata.

Cheers,
Liz

Liz Krznarich

unread,
May 6, 2020, 5:30:56 PM5/6/20
to ORCID API Users
All good points on the utility (and some challenges) of contributor lists in ORCID, and thanks for the input. 

While allowing users to manually type in a list of contributors may ultimately be a piece of the solution, ideally the user should not have to re-type this info into ORCID if it exists elsewhere.
The idea of enriching works metadata with additional metadata (contributors and other info) pulled from the external identifier source is being analyzed by our product team for possible future implementation. 

We recognize that this is clearly an important issue to the community and will continue to update you as we make progress - we do have a small team and are in the midst of migrating our UI to a modern framework. We will have more capacity to take on other UI improvements and new features later this year when that migration project wraps up.

Cheers,
Liz
>>>> To view this discussion on the web visit
>>>> https://groups.google.com/d/msgid/orcid-api-users/14ef7b9c-1509-442c-96e0-8d0cf26ca51a%40googlegroups.com
>>>> <https://groups.google.com/d/msgid/orcid-api-users/14ef7b9c-1509-442c-96e0-8d0cf26ca51a%40googlegroups.com?utm_medium=email&utm_source=footer>.
>>> --
>>> You received this message because you are subscribed to the Google
>>> Groups "ORCID API Users" group.
>>> To unsubscribe from this group and stop receiving emails from it, send
>>> To view this discussion on the web visit
>>> https://groups.google.com/d/msgid/orcid-api-users/08acbc5f-dc3f-6564-bb3c-7b9c443ba1cc%40antonin.delpeuch.eu.
>> --
>> You received this message because you are subscribed to the Google
>> Groups "ORCID API Users" group.
>> To unsubscribe from this group and stop receiving emails from it, send
Reply all
Reply to author
Forward
0 new messages