Add ORCID icon to contributors grid row for authenticated ORCID ids #116
Comments
This is interesting. Currently we display an ORCID icon whenever an ORCID URL has been entered for a contributor. Removing them from the frontend display will likely effect a lot of journals who won't know about the authentication, and won't have any way of getting some contributors (who in many cases are not users in the system) to authenticate. Emailing a contributor who has not registered a user account would be a breach of GDPR. Has there been any discussion about what happens to existing ORCID ids that have not been authenticated? If we are moving to a model where ORCID ids have to be entered and authenticated by the owner, that will probably disrupt our contributors-not-connected-to-a-user approach. And in that case we'll need to track status in the contributors grid -- similar to how Primary Contact and In Browse Lists is handled. I'd like to get some more feedback on this and how we expect to manage this change. |
|
If we are moving to a model where ORCID ids have to be entered and authenticated by the owner, that will probably disrupt our contributors-not-connected-to-a-user approach. I see benefits in having contributors/authors associated with user accounts. One of the great things about ORCID is that it adds a layer of authentication, preventing fraud. When you allow someone to add ORCID IDs for anyone, that authentication is lost. By only allowing the owner of that ORCID ID to be one associating it with their account, you are certifying that that person really is who they are claiming to be. I might be mistaken but I think ORCID doesn't recommend anybody else including ORCID IDs except the researcher owner of that ORCID ID. Other benefits of having contributors/authors associated with user accounts: ability to track down submissions by co-authors; allowing co-authors to check on their submission status. |
|
@NateWr , you wrote:
Is this true? Because that's what I have implemented for the OrcidProfile Plugin in the member-api Branch. The idea was that the submitting author or an editor could enter the contributor's email and request the ORCID authorization by ticking a checkbox in the metadata form. |
My understanding is that technically, yes, it is a breach. But there has to be a carve-out of some kind for one-off emails. Think about a forgotten password form, which allows you to enter any email address in and an email is sent off. So there must be an exception allowed for cases where one person asks for another person to be emailed. What that exception is, and how we can use it properly, I'm not sure. |
|
Thinking about this, as this would be part of the publishing process, there has to be a prior agreement between the journal and the authors, so that the e-mail address can be used for communication purposes. This would only be a problem if an author uses the submission form and enters the other contributors e-mails without checking with them. |
|
We should be treating ORCIDs as an opportunity to counterbalance our author/user account split: when both a user account and author record share the same ORCID, that's currently the only reliable way of knowing they're the same person. I do worry about a shift in requirements for authentication basically sidelining all the ORCID data that's been entered until this point. But I see what ORCID's trying to achieve and think we should help facilitate that as much as possible. The ORCID guidelines restrict the use of the |
|
Just a short note from an editor's side regarding this:
Although it is best practice to contact all co-authors and have them agree to a publication, it is sometimes not possible for the submitting author to track down everyone who has contributed to a paper. Think about the student, who did measurements several years ago and has long left the lab. The submitting author then either enter their own email address along with the co-author's name or they omit the names on the submission form but still have them on the document. The editorial staff has then to enter the names into OJS afterwards, but without valid email address. |
@asmecher Yes, this approach also came to my mind, while developing. I would prefer this as well.
@heike-r |
|
@isgrim
I think the current system is the most convinient one and if ORCID is not necessary on submission, everything should be fine. However, I agree with @alexxxmendonca comment on the
If a submission in progress could be seen (but not altered) by co-authors, who have an account at the journal too, it would be nice (maybe an additional role 'co-author'?). The submitting author would have to indicate upon submission, that this specific account is his/her co-author. Since the unique identifier is the OJS username, but the submitting author most probably does not know that one, he/she would have to resort to another, publicly accessible unique identifier, which is associated with an OJS account. That could be a validated ORCID. |
How about a red "x" for not validated and a green "check" for validated? |
A different badge for non-authenticated iDs was also raised in the comments of the ORCID for Repositories discussion document, but to date ORCID does not have one. Perhaps something to be raised / suggested in the ORCID iDeas Forum?
ORCID recommends the green iD icon for validated/authenticated iDs. Some systems use the red X for those iDs which haven't been authenticated -- but only internally, for example to an author in their personal profile or a journal admin in the metadata profile, as an indication that the author should authenticate their ORCID iD. When displaying iDs non-authenticated publicly, we've previously suggested displaying the identifier only, hyperlinked, with an [unauthenticated] notice next to the identifier and "authenticated=false" in relevant metadata. |
|
Right. I meant internally indeed.
I see. That works both internally and externally. |
|
@nils-stefan-weiher @withanage I just came upon this thread because I was not sure what the intended behavior of the ORCID profile plugin should be: I noticed that, in our case (using OJS 3..3.0.8, ORCID Profile 1.1.3.4), the plugin also displays already existing but not authorized ORCID IDs with the green badge on the article page: Is this the intended behavior? |
|
Hi Felix, |
|
Thanks @withanage! |
Hi everyone,
during the work for pkp/pkp-lib#2818 and testing the new additions @alainna suggested, that it would be nice for editors to see at a glance which Authors have an authenticated ORCID id stored.
I talked with @NateWr in the pkp-dev Slack channel about this. He suggested something like this:
What would be the condition for the display to come?
As per the ORCID display guidelines (https://orcid.org/trademark-and-id-display-guidelines#02-b) the icon should only be displayed for authenticated ORCID ids. I could check for the plugin specfic Author setting "orcidAccessToken", which is only stored after successful authorization by the owner of the ORCID id.
Edit: Added link to ORCID display guidelines
The text was updated successfully, but these errors were encountered: