Mike Nason PRO
Open Scholarship and Publishing Librarian @ UNB Libraries // Metadata Nag and DOI Wrangler @ PKP // General Loudmouth and Malcontent
You've asked me many things
Mike Nason
Open Scholarship & Publishing Librarian
UNB Libraries
(2024 CARL Repositories Engagement Group Call)
just off the top here, i want to thank folks for asking all these questions. Let's keep things conversational as we go, also.
I've broken things into a few sub-categories.
Categories & Themes
- Repositories and Open Scholarly Infrastructure
- Discoverability
- Scholaris
- PIDs
- Potpourri
- ¯\_(°⊱,°)_/¯
Let's work our way through the questions in these chunks. Emily is going to keep us on track if stuff gets too far afield.
Let's get poppin'
Broadly...
"What do you think is the easiest to accomplish, big win, we can achieve in the next couple of years in this domain?"
For me, this comes down to three/four things that are already on our plate.
- A responsible and intentional application of persistent identifiers.
- Leveraging open infrastructure.
- More collective effort to help with indexing (especially in OpenAire).
- Mutual support and shared expertise.
I think this stuff is "easy" if we focus on starting with newly ingested content and worry less, for now, about how to address enormous, existing collections.
"What do you think is the easiest to accomplish, big win, we can achieve in the next couple of years in this domain?"
I think if we focus on appropriate/sustainable use of PIDs and leverage OpenAire indexing, we're improving discoverability enormously and also highlighting the connections between works, funders, and institutions that make up narratives of research.
"easy" is obviously relative, but I think this is well within reach.
Scholaris
"What's your impression of the Scholaris Project this far?"
I'm optimistic!
Scholaris
"As an expert [you are too kind] who has been involved in multiple repository migrations, what advice can you give to the Scholaris project that will soon be undertaking a massive migration when they begin their work."
Fortunately, I'm lucky to be on both the Shared Repository Infrastructure Advisory Committee (SRIAC) and the Scholaris Metadata and Discovery Experts Group (as a co-chair).
I am blessed with opportunities to be pushy. But, it's not just Scholaris and Scholar's Portal who need advice here...
THis is an enormous opportunity to clean up messes, reset some expectations, and establish some sensible boundaries around repository management.
THis is Also an enormous opportunity to engage more directly with open source software as a community, and potentially contribute to "making the sausage", so to speak.
Repository Management
"We are moving to Scholaris and need to decide how much we want to accept in our repository (ie, should we aim for all faculty publications even just the metadata if they are closed, or only if they need the repository to fulfill grant requirements) have best practices in this area changed in the last few years?"
Generally, my opinion is that a repository that's just offering up a different version of a paywall is an astounding waste of time/effort.
Repository Management
"Can you tell us more about your work with your metadata schema? How did you make certain decisions regarding including DC and OAIRE fields in your schema? Do you think DC will map cleanly to DataCite fields?"
Repository Management
"Any best practices for high-level structure/collections by school?"
Repository Management
"PIDs. PIDs? PIDs!"
This is more of a question than a comment, but ok...
Repository Management
"What approach should we take with minting things in our repositories with DOIs? Should everything get a DOI?"
Please, please, please do not assign a DOI to everything.
Does everything need a DOI?
Are you going to maintain them?
HANDLES ARE TOTALLY FINE.
Discoverability
"How can we best communicate to faculty the impact that depositing their articles into the institutional repository has?"
Patience? Keep your expectations low.
Discoverability
"Further, how can we incorporate repository profiles/uploading their publications to the repository into the wider project of helping faculty demonstrate their research impact?"
Firstly, I really do not care about "helping faculty demonstrate their research impact." What I do care about is saving researchers time and improving access to – and discoverability of – their works.
Let's get stuff in ORCID, plz. At a minimum.
Let's talk about it! And entities!
Discoverability
"Are there best practices for metadata re-use (pulling and pushing) to other systems, such as a RIMS. Are there integrations with DSpace and RIMS systems so that we are entering data only once and re-using it in our other systems? How to best take advantage of them? Aside from tech requirements, are there policy requirements? Other?
PIDs should be doing this work for us, assuming we're using them, and assuming the RMS can use them, and assuming publishers provided that metadata, and assuming a PID was registered for a work in the first place, and assuming authors have allowed for access to an ORCID record, and assuming that author has configured ORCID to ingest metadata from DOI registration agencies... and and and...
Discoverability
"Is the Canadian landscape sophisticated enough to do the work that is required for discoverability via repositories (eg. are there enough people working on this in Canadian Universities to make progress)"
I love this question. Yes. Maybe. I think so. Kind of.
I don't think sophistication is the issue so much as cultural and political buy-in. But, there are reasons to be optimistic!
There are also many legitimate challenges!
BONUS ROUND QUESTIONS
"WorldCat Digital Collection Gateway : Is it possible for DCG not to push certain “dc.contributor” fields. For example, dc.contributor.affiliation causes problems because it is pushed into the author field in the bibliographic records of our search tool (OCLC WorldShare)."
dc.contributor.affiliation is an application of a custom qualifier. In dublin core there is an expectation that all metadata ascribes to the "dumb down" rule, wherein any metadata in a qualifier has to also match the expectations of the element itself. In this case, an affiliation is not a contributor. DC doesn't really have a place to store metadata about authors beyond their name. HOWEVER...
BONUS ROUND QUESTIONS
"Why are articles that are in our institutional repository not flagged as open access by Google Scholar, especially in the author profile? Can we do something about it?"
I'll be honest, I try to think about Google Scholar as little as possible. Almost all of their indexing is based not on available metadata in open infrastructure (like DOIs) but, instead, on platform-dependant META tags.
All I have for you in this space is guesses. You may have luck trying to reach their contacts. All answers would depend on what platform you use, and how you store/reveal access metadata.
BONUS ROUND QUESTIONS
"I am working for a university adjacent non-profit and we want to create a document repository for university members to be able to access the resources in. What are the steps we would need to take to do this?"
This is kind of an enormous question but I, if access management isn't an issue and everything can just be public, I might recommend just making a collection in Zenodo and saving yourself a lot of time and money.
I'd like to thank everyone for having literally any faith at all in my ability to answer these questions. It's nice!
AMA
By Mike Nason
AMA
CARL AMA
- 160
