Following our recent Open Book Metadata webinar, we asked our speakers to respond to the unanswered questions posed by attendees via the webinar Q&A channel. You can find those questions and answers below as well as speaker key takeaways and an edited curated list of resources and comments shared via webinar chat. This may be useful for those who missed the webinar or wish to share information with colleagues.
OASPA is extremely grateful to the speakers for all of the work and time they have given to this webinar – in preparation, during, and after.
Speaker Key Takeaways
From Dominique Babini (CLASCO)
In Latin America:
- Scholarly publishing is mainly scholar-led, non-profit, with no outsourcing to commercial publishers, and no article processing charges (APC) or book processing charges (BPC). Strong tradition of university presses with growing adoption of open access for books
- Open access legislation and policies privilege repositories as the preferred venue for open access to research results. Repositories in Latin America use Dublin-Core open metadata and OAI-PMH protocol for harvesting
- CLACSO´s network 3.200 open access books with Dublin Core open metadata can be harvested (OAI-PMH) from CLACSO´s repository
- Other cases of interest in Latin America are SciELO Books and the The Association of University Presses of Latin America and the Caribbean (EULAC). Both use ONIX open metadata
- Some challenges: advanced training for staff; DOI for individual books and collaborative book chapters; require intermediaries to harvest complete metadata from books; due recognition of books in research assessment procedures; describe peer-review process; develop open access book usage indicators
From Mel Bach (Cambridge University)
- We need to define and agree responsibilities for problematic OA metadata
- We should work towards solutions that prevent problems rather than just resolve them
- Collaboration (publishers, platforms, LMS providers, librarians) is key
- Prioritising OA metadata work is harder for librarians than it should be – the tension between principle and practice
- Like any tool, open metadata can create opportunities for both benefits and harms
- To ensure scholarly freedom, care must be taken to consider the benefits alongside potential ethical risks and disparate impacts that can arise from third party linking and repurposing of open book metadata
- Data collaborative models, like that being piloted in the OA eBook Usage Data Trust, may provide a multi-stakeholder approach to exchanging and stewarding sensitive book metadata alongside open metadata.
From Jennifer Kemp (CrossRef/Metadata2020)
- Metadata matters because it is widely distributed and widely used. Machines are readers too!
- Good metadata is hard. It’s important to recognize the effort it takes to do it well
- Improving the landscape requires broad participation from across the global community and ongoing collaboration
Shared resources and anonymised comments
Questions received via the webinar chat channel (unanswered during the webinar)
Answers from Dominique Babini (DB), Mel Bach (MB), Christina Drummond (CD) and Jennifer Kemp (JK)
Q. How do we join a working group for book metadata?
CD: People interested in linked book usage data are invited to join the OA eBook Usage Data Trust working groups. You can sign up at https://bit.ly/OAEBUGROUPS. You could also consider engaging via Project COUNTER.
Q. Hi, I’m not familiar with the concept of decolonisation presented by Mel Bach. Can you develop a bit?
MB: Following on from what my fellow panellists said live in response at the webinar, I would also say that I am thinking about collection development here. If we want to have diverse collections that are not just wall-to-wall English-language ebooks from the Global North, we need to prioritise collections like CLACSO. Yet how do we prioritise 200 records against 100,000s of other records? That is a good example of the tension, for me, in terms of turning principle into practice.
DB: For us at CLACSO the exchange with Mel Bach and other panelists and participants has helped us better understand the need of finding ways to improve the international visibility of books published in our developing regions. A challenging issue because of costs of translating books and/or its metadata.
Also in reply to this question about decolonization and open scholarly communications, I suggest to look at, among others:
- Bibliodiversity at the Centre: Decolonizing Open Access
- Confronting Epistemological Racism, Decolonizing Scholarly Knowledge: Race and Gender in Applied Linguistics
- Creating a More Inclusive Future for Scholarly Communications: ACRL’s New Research Agenda for Scholarly Communications and the Research Environment
- Creating a More Inclusive Future for Scholarly Communications
- Introduction: The Politics of Open Access — Decolonizing Research or Corporate Capture?
- Contextualizing Openness: Situating Open Science
- ACRL-SPARC Forum: What we learned about community alignment and equity for emerging scholarly infrastructure
- Decolonizing scholarly data and publishing infrastructures
- Open Access, the Global South and the Politics of Knowledge Production and Circulation: An Open Insights interview with Leslie Chan
Q. Did I get it right? Metadata with some other than CC-0 licence?
DB: Maybe it refers to my presentation where I mention that in repositories in Latin America we follow the guidelines of DRIVER https://zenodo.org/record/1299203#.YDvHqGhKjZA which require informing the CC licence used for the digital object (in this case the CC applied to the content of the book) but we do not have a specific requirement for CC used for the metadata,which could be a different CC licence.
Q. Is there a place for shared information about metadata for open access ebook collections? Or plans for such an endeavor? Our consortial members are interested in adding OA ebooks to our shared catalog and we have in place a workflow to gather information about the collection and if/how metadata records are provided. The quality of metadata affects whether or not we add collections. It would be helpful to have and share such information for anyone wanting to consider adding collections rather than everyone seperately looking it up.
CD: This is one of the services the OA eBook Usage Data Trust is exploring. We’ve heard through our use case development process that this is a widespread need. Every organization is expending resources on “data wrangling” and there is interest in finding sustainable ways to achieve economies of scale and reduce the institutitional burden. We are actively piloting open source infrastructure to research technical feasibility as we develop community-based sustainability and governance models. Personally, I think this an emerging piece of core scholarly communications infrastructure: a system level solution to aggregate and return feeds of public metadata alongside access-controlled platfrom or service data about OA ebook related usage, engagement, and impact. Happy to follow up offline.
Q. Can any of you say something about multilingualism in metadata? Do you know a system or program that allows enter metadata in one language and the search engines recognize it in another language?
DB: This question is giving a fantastic idea to experiment in a project and see the results!
Q. Is there a current library organization(s) that are able to move this conversation forward? IFLA? What is your response to OCLC being involved with this?
DB: maybe DOAB could move this conversation forward.
Q. NISO would be a good intermediary for the discussions that you want Dominique. Many of the big publishers and content providers are members of NISO as are a growing number of libraries.
CD: The OA eBook Usage Data Trust effort is engaging NISO to explore recommended practices around the aggregation and display of usage data.
Q. How we can Open Access eBook metadata more rich and useful for our readers?
JK: From the Crossref perspective, it may be helpful to know that our metadata users frequently request abstracts, references, license and funding information and affiliations for all authors but we encourage book publishers to supply as much metadata as possible. Our Participation Reports highlight some key metadata fields.
DB: In Latin America OA books in repositories follow the Driver guidelines for metadata and other OA book initiatives use other metadata formats, eg. ONIX.
Q. In relation to subject headings, which standards do data trusts advocate for? There are multiple standards so one ends up adding different standards into one database.
CD: Our technical standards and norms working group is where this type of conversation would take place.
Q. How about the idea of Linked Data for Open Access Book metadata? Library catalogers are experimenting more with something like Wikidata, which might give some solution for multilungual issues?
CD: Data linking is at the core of the data trust approach, as we look to achieve economies of scale to foster sustainability through common core infrastructure that supports analytical innovation on the edges.
Q. The users, the readers, what would be their comments on the relationship with the metadata?
DB: The users of the OA books metadata are so diverse: students/professors/researchers/other social actors searching in person in libraries, those users searching in the web, bookstores needing ONIX metadata for purchasing the printed version of academic books, etc.
Q. Do libraries – as paying customers – either individually or collectively, have any influence over the ”enrichment” applied by the systems they purchase/make use of? Wouldn’t this be expected to some extent, from a service point of view?
MB: Yes, I think that is how this should work. During the webinar, a librarian who is a rep on my LMS provider’s community zone group got in touch to suggest she and I talk about the problems I had mentioned, which I am looking forward to. It could be that there are nuances I am not aware of but also I am hopeful that the LMS provider can offer solutions. Her response to your question was captured here too, as follows. Re: collaboration between LMS suppliers and libraries – Ex Libris have a CZMG to advise on enrichment. It’s not perfect, as highlighted by Cambridge blog but it’s better than it was 4 years ago. I assume that other systems providers have similar groups?
DB: As OA book publisher we would like to see suppliers include the link to our Dublin Core metadata in the metadata they supply to libraries
Chat log resources
Q. ScienceOpen is aggregating metadata, mostly via citations of DOIs for books and chapters from Crossref – we are now up to over 3 million records. One of the issues that we encounter is multiple DOIs for a single title. Is there a better solution than the co-access option at Crossref?
JK: At Crossref, we are investigating a replacement for the Co-access option. Stay tuned!
Q. Enriched metadata is a legitimate cause, but what about when databases and repositories do not support them? We as a publisher are again and again meeting with databases that do not care about the format nor what is listed in the metadata files. I have been told we can enrich all we want but they can’t display it. I have been told that we can put DOI in ONIX metadata file as a main identifier even though ONIX as a format is based on ISBN as a main identifier. The problems in practice are numeorus.
JK: Very good question and something that came up in Metadata 2020 research. I don’t have an easy answer but I think it certainly speaks to the need to better understand stakeholder needs and advocate for systems and processes that support fuller metadata.
CD: I think this is one of the areas where a neutral multi-stakeholder backbone organization can help as this is a global problem that spans government, industry, and academia. This It really is a collective action challenge that crosses public/private, industry silos and disciplinary verticals. I find myself thinking of parallels with internet standards organizations.
Q. I’m concerned about digital preservation… Can we please comment?
JK: I am too so thanks for asking! At Crossref, members can include information on where book content is archived in the metadata. We also archive the metadata itself with CLOCKSS and Portico.
Q. Christina Drummond said something like “Stakeholders have their own lexicon, use and need of metadata – at the end of the day we need interoperability on a regional basis and what that means in an international context.” Could she explain/elaborate on the regional/international a little more? Is this referring to language barriers? How specific/generic are the metadata needed by the different stakeholders?
CD: While others are better positioned to speak to global inter-operability across character sets, I was specifically thinking about policy dimensions at the intersection of national data policy, international affairs, open access, and big data. As open metadata is linked and can tell stories of influence and impact, if open it will be repurposed and leveraged in different contexts.
Q. The whole of open infrastructure is a collective action problem – given the huge diversity and needs raised here, how can we bring all this energy together?
JK: This is such a key question in my view! I would love to get suggestions and talk with others on continuing these conversations.
CD: Whole-heartedly agree – efforts like Invest in Open Infrastructure (IOI) and SCOSS have related roles here. I think we need to be specific on what we’re trying to accomplish to find the right place to host interrelated conversations. Personally, I would love to help develop a collective action agenda/roadmap that shows how efforts intersect to drive us forward towards system-wide impacts.
Also the list of reading material shared by Dominique Babini in answer to the question above.
Note that previous OASPA webinar details and recordings can be found here.