Tag Archives: institutional repositories

Checking out the Elsevier / U of Florida pilot

One of the papers at Open Repositories 2017 I couldn’t attend was:

Maximize the Visibility and Impact of Open Access and other Articles through integration of Publisher APIs
Letitia Mukherjee (Elsevier), Robert Phillips (University of Florida)

The University of Florida searched for solutions to expand access to university-authored journal articles thru institutional repository. UFL and Elsevier collaborated to automatically feed journal platform data and links to the IR through free APIs. The project enabled UFL to support university authors/researchers and compliance with US public access policies.

I wrote most of this blog post based on what I heard about the presentation at conference, and my own investigations a couple of days later (ie a month ago); I’ve made some small edits and am posting this now after seeing the presentation recording on YouTube.

I first read about this project a year ago in an Inside Higher Ed article (in which Alicia Wise is quoted with an infuriating “The nice thing about this pilot is it opens up the repository”. No, it doesn’t open the repository. The repository was already open. It also doesn’t open up Elsevier content, which remains completely closed) and in a more sceptical blog post (which describes it as turning the repository into “a de facto discovery layer”. From what I can tell, this is being extraordinarily generous: as a discovery layer it doesn’t even make a particularly good Amazon affiliate programme, because Amazon at least pays you a few cents for the privilege of linking to them.)

Before going further I want to make it clear that any and all scathing comments I make in this post are reflective of my opinions about Elsevier stinginess, not about the repository or its staff who are clearly just doing what Elsevier allows them to do. Also I’m writing about the system as it is right now (Phase I). [Phase II was briefly discussed starting about 18:55 in the video and in Q&A at the end of the presentation.]

While still at conference, I heard that Robust Discussion was had following the presentation (and this is captured in the video too). Among other questions, an audience member asked if Elsevier would offer all subscribers the ability to download final accepted manuscripts via API for example (21:59). The eventual answer (after some confusion and clarification) seems to be that it’s not currently available to all subscribers as they’re creating author manuscripts specifically for the pilot and need to work out whether this is scalable (24:44). [This raises the question to me of why. Why not just use the actual author manuscript instead of converting the author manuscript into the publisher manuscript and then apparently converting it back?]

In any case, when I asked the same question at the vendor stall, I was told that if they provided the pdf to repositories, they wouldn’t be able to track usage of it. The vendor also asked me why we’d want to. I talked about preservation, primarily because I foolishly assumed that the system they’ve got with Florida actually worked as advertised to provide ‘public access’ but a couple of days later, somewhat recovered from the exhaustion of conference, I had second thoughts. Because of course the other things that we want are full-text searching and access via Google Scholar. Also access for the general public, not just our own university. Also, well, access at all. I thought this went without saying until I actually began to test how it works in practice.

So University of Floriday’s repository is IR@UF. I ran a general search for {Elsevier} and turned up 32,987 results. I chose an early result that wasn’t from the Lancet because the Lancet is a special snowflake: “(1 1 0) and (1 0 0) Sidewall-oriented FinFETs: A performance and reliability investigation”.  The result is plastered honestly with “Publisher version: Check access”.

Is it open access? I clicked on the title. Elsevier has made much of “embedding” the content in the repository. I think this is in fact intended for phase II but they’d managed to give the impression that it was already in place so at this point I expected to be taken to a repository page with a PDF embedded in an iframe or possibly some unholy Flash content. Instead, I was taken straight to the item pay-to-download page on Elsevier. Further exploration uncovered no additional ways to access the article. So there’s no access to the public: it’s not open access and it does absolutely nothing to support “compliance with US public access policies”.

Is it easily accessible to institution members? If I was a UFL student or staff member who happened to be off-campus (say, at a conference, or researching from home) there’s no visible way to login to access the article. I assume UFL has IP access to content in which case it’d work on campus or through a VPN, but that’s it.

Is it findable through full-text search? I dug up access through my own library to download the pdf so I could select a phrase early on in the full-text that didn’t appear in the title or abstract. But doing a full-text search in IR@UF for {“nMOS FinFET devices”} resulted in “Your search returned no results“.

(Just to be sure the full-text search was working, I also tried it with a phrase from the title, {“Sidewall-oriented FinFETs”}, which did bring up the desired article. The link from this result is broken, though, which is presumably a bug in the implementation of the scheme, since links for non-Elsevier results on similar full-text searches are fine.)

Is it findable via Google Scholar? Scholar lists 6 records for the article, none of which are via IR@UF. Not, at this point, that there’s any advantage to seeing the IR@UF version anyway, but the pilot is certainly not driving traffic to the repository.

Is it a discovery layer? Even aside from the lack of full-text search and the inability to get access off-campus, it only works for ScienceDirect articles by UFL authors, so no.

If I had to come up with an analogy for what it is and does, I guess I’d say it’s a bit like a public-facing RIMS or CRIS, except those would include more data sources and more reporting functionality.

So to answer the question as I could have if I’d realised how limited this functionality is: why do institutional repositories want to have the full text?

  • to make it discoverable via full-text searching
  • to provide easy access for our own institution’s members
  • to provide open access for the rest of the world
    • thereby increasing its impact (including but not limited to that measured in citations and altmetrics)
  • to ensure it’s preserved and accessible for the centuries to come
  • to bring traffic to our own repository and the rest of its valuable collections; and
  • to track usage.
    UFL’s repository can do this last one. Sort of. It’s got a page for “Views” (hits) and “Visits” (unique visitors) . But it doesn’t tell us how many of these visitors actually succeeded in accessing the full-text. My suspicion is that this number would be much lower.

Phase II, if it works as advertised, may address some of these issues, but I’m not sure how many. I feel we’re getting conflicting messages of how it will actually function and at this point am not inclined to believe anything until I see it in action. For now it’s the same as any other vapourware.

Beyond Repositories: Problem-solving-oriented #or2017

Beyond Repositories: From Resource-oriented towards Problem-solving-oriented by Dr Xiaolin Zhang, National Science Library, Chinese Academy of Sciences

With the ubiquitous deployment of digital ecosystems, developing repositories to meet next generation needs and functions become an imperative and increasingly active efforts. However, a paradigmatic shift may be needed to prepare repositories to go outside the resource-orientation box, as JISC report “The future of data-driven decision making” puts it, “[I]t is not sufficient simply to focus on exposing, collecting, storing, and sharing data in the raw. It is what you do with it (and when) that counts”.

The presentation first discusses the emerging digital ecosystems in research, learning, publishing, smart campus/cities, knowledge analytics, etc. where traditional content/repositories are just a small part of stories.

Then an exploration is made about making repositories embedded into, integrated with, and proactively contributing to user problem-solving workflows in digital ecosystems such as scholar hub, research informatics, open science, learning analytics, research management, and other situations.

Further effort is attempted to understand (admittedly preliminarily) strategies for repositories to be transformed into part of problem-solving-oriented services, including, but not limited to, 1) enhancing the interoperability to be re-usable to third part “users”, 2) developing repositories into smart content with application contexts, and 3) developing smart contextualization capabilities to better serve multiple, varied, and dynamically integrating problem-solving processes.

[I’ve previously blogged a keynote by Dr Zhang at THETA 2015.] He has a new perspective since moving jobs two years ago.

104 research institutes, 55,000 researchers. Various repositories eg NSFC Repository for Basic Research, CALIS IR portal of 40+ universities. Research data sharing platform, and Chinese Academy of Science distributed research data management and integrative service platform.

  1. Changes in the digital ecosystems
    • Steady progress of repositories but numbers don’t tell the story – better to look at how users use it. Most still collection based and local applications are the main service. What if we move away from repository-based approach. Imagine new scenarios out in society. What do they need?
    • All media and content can be data (including processes, relations, IoT devices, tweets). Can be smart – and semantic publishing will be the new normal. Knowlege as a Service.
    • Transformation from subscription to open access. Born digital = born linkable.
    • eScience is the knowledge system – opening up data-intensive scientific discovery. Not just about access, it’s a different way of doing science
    • Open Science again more than open access, but open evaluation, open process, open collaboration. (Displays open science taxonomy). Even social science now incorporating computational methods.
    • eLearning creating a new knowledge ecosystem. Things changing quickly. In the classroom everyone (200 students) uploading content and system going down even though made plans for it only 2 years ago. Flipped classrooms where students do work before the class in digitally collaborative environments, multimedia-rich laboratories so students can interact with each other. Requires intelligent campus and services. eStudent Center where student’s whole learning life is together to be analysed; university center can look at trends etc
    • Knowledge analytics – converging data science, computer science, information science. Open source tools for data visualisation and analysis. Data analytics can become new infrastructure
    • Moving into the Machine Learning Age? 7.5 million university graduates every year in China
  2. Explorations to re-orient repositories
    • Towards working labs: Elsevier Knowledge Platform; WDCM
    • From resources to problem solving, eg digital healthcare needing knowledge from literature but also from wearables and other devices; eg intelligent cities with data, linking, analysis, to answer questions.
  3. Challenges in re-developing repositories
    • Re-purpose and reposition repositories? but outside the scholarly communication environment? Eg using big data in smart cities – scholarly knowledge plays a huge role here. Eg learning analytics where we combine data on students (grades, interactions on Moodle).
    • Cycle: environmental scanning -> idea/design/testing -> R&D -> data management -> Data analysis -> dissemination -> preservation/reuse -> evaluation -> environmental scanning
    • Interoperability cf W3C recommendations
    • Identify/select/developed/integrate value-added services (not all work together, but some aren’t meant to). How to turn content into computable data? how to develop rich and smart media resources? eg How to turn powerpoints into actionable data?
    • Working on automatic translation, domain interaction dynamics, scientometrics tools, social network metrics, automatic thesaurus/k-graph development. Hard for students to select a topic when there’s open-source tools already out there about it! Calculations and results become objects to be reused.
    • Representing knowledge with knowledge graphs. which can enable intelligent applications. Text analytics, RDF data management. eg SpringerNature SciGraph – turning all papers into semantic network of knowledge.
    • Too many vocabularies! Some used by many people, some very common (eg Schema.org ) and general – but also very specific ones eg neuron ontology; Internet of Things developing their own. Ontology mapping tools? Cross-language linking of knowledge graphs and smart data eg Chinese/English Wikipedia pages.
    • What about when live machines join the integration and we put our data into real-life processes? Geospatial/temporal/event/methods/workflows-identifiable.

Are these real life scenarios really relevant to our repositories? If not we’ve got a problem! Is what we’re doing now getting us into these scenarios? Are we talking/collaborating with people in these scenarios? They’re not necessarily going to approach us! Time for us to think and act before it’s too late.

Ideas Challenge presentations #or2017

Challenge to solve an existing problem with emerging technologies.

Data Pickle

Research wanted to upload data but didn’t know how to wrap it up. So cf ThisToThat for gluing thing A to thing B. Let’s make this for data.

Package Shapefiles for Preservation. click “PICKLE!” and it recommends a) the best practice and b) the minimum requirements.

Crowdsourced but curated information for various options.

Technology handshake to achieve Australasia PMC

Right now have EuropePMC and CanadaPMC (child nodes of US PubMed Central which has 27million references). So create AustralasiaPMC so PMC can link to OA articles. Can populate PMC with clumsy markup so need clever handshake technology  to make full-text available in children and parent nodes simultaneously.


Museum is an interface for scientific information to general public. But takes too long for simplified explanations of science (from eg journals) to general public, and journalists don’t always guard scientific integrity.

Want to do a better job of spreading info through social media. Natural language processing to create automated simplified summaries from technical abstracts; push notification to simple.wikipedia.org proposal pages so they can create or add to articles; Google translate for other languages.

Put it all together and you get communication immediately after acceptance, being picked up correctly by major news outlets.

(In Q&A: hard to contextualise. Audience notes researchers want to say ‘further researcher needed while lay people want to know what the answer is.)


The technology we’ll use in future repositories has already been written – GitHub is full of work in progress – some people know about it but not all of us. Pull code automatically from everywhere, put it together, throw data in, see if it works.

Plan A – artificial intelligence – most advanced AI right now is self-driving car, so jump in front of one with the repository and the car can evaluate it and then run you over.

Plan B – use humans

(In Q&A: Kim Shepherd suggests when on GitHub and look at number of forks on projects – what percentage might be active, what percentage should we have merged in.)

Global Connections

Deep learning for repository deposit – use existing repository PDFs and metadata to train AI to a) create structured metadata for unstructured content (ie articles), find relevant articles, add structured metadata.

Slice ‘n’ Dice: API-X + XProc-Z

XProc-Z is a simple web server framework HTTP request -> -> -> HTTP response (especially useful for proxies)

API-X for plumbing together microservices.

GET request for info on resource – API-X intercepts/proxies, tweaks, and makes request to server, retrieves result, wraps in a header, tweaks and returns to user.

Don’t need to develop code, just write a text file in XProc language so you can test out what it looks like and you don’t need to wait for repository support. Signposting; generating IIIF manifests; add OAuth authenticating; adding CSS.

Brisbane Declaration ON the Elimination Of Keywords (B-DONEOK)

Keywords can’t express the complexity of language the way full-text can. We spend time doing it anyway. So let’s stop. Instead just use sophisticated full-text search and indexing. SIgn on to the declaration at http://bit.ly/2u4KjMm

(In Q&A audience asks if there’s evidence keywords aren’t useful; team asks in return if there’s evidence keywords are useful.)


Institutional Publications Repositories and beyond #or2017


Curating, But Still Not Mediating by Jim Ottaviani, Amy Neeser

aka “don’t let anyone get away with 6accdœ13eff7i3l9n4o4qrr4s8t12ux” (Isaac Newton establishing priority on calculus in code)

Chinese proverb: “The best time to plant a tree is 20 years ago; the second best time is now.”

Curation starts immediately: don’t wait or people will forget. You have to open every data file and check it’s usable. They assume if it’s intended for humans it’s a “document” but if it’s intended for machines it’s “data”.

Acknowledge/thank the deposit (signing your name so they know you’re human not bot). Then you can ask for a README.txt or offer to help write it.

Home and Away: Exploring the use of metrics in Australia and the UK, with a focus on impact by Jo Lambert, Robin Burgess

Sydney measures metrics through: Atmire tools embedded in repository; PlumX integration altmetrics; ERA requirements; CAUL stats; exploring UK methods. Employing FAIR principles. Researchers provide context for impact.

JISC OA services support through article lifecycle of submission (SHERPA/RoMEO etc), Acceptance (Monitor UK, Jisc collections, OpenDOAR), Publication, etc. Stats collected via aggregation then available in COUNTER format; raw download data from UK IRs using DSpace, Fedora, PURE, ePrints…. Collaboration is important – working with OpenAIRE, concept of creating other IRUS instances eg IRUS-ANZ?

Set up Australian Repository Working Group. Looking at standards and collaboration. “We dream the same dream, we want the same thing” – Belinda Carlisle

Uniform metadata for Finnish repositories by Jyrki Ilva, Esa-Pekka Keskitalo, Päivi Rosenström, Tanja Vienonen, Samu Viita

Open Scientific Publishing project (Tajua). 60 orgs in Finland have an IR, mostly DSpace – some are shared so total of 17 IRs.

Challenges: heterogeneous metadata practices, ad hoc solutions, no general metadata guidelines so repository managers have to fend for themselves.

80 experts got together, formed a smaller working group including National Library experts. Goal-oriented approach to develop a “good-enough” metadata format, semantics prioritised over correct Dublin Core. Compiled most used metadata fields and suggesting fields closer to standard DC.

Spreadsheets collected into Google Drive, meetings held online and in person – then done! Final version published on National Libraries public wiki fields. 62 core dc fields, and 11 extras if needed. 6 fields labelled important: title, author, date, persistent ID, rights (pref Creative Commons). Guidelines at: https://www.kiwi.fi/x/94R7B

Isomorphic Pressures on Instutional Repositories in Japan by Jennifer Beamer

Comparing US and Japanese repositories as interested in situation as institutions interact with repositories. In Japan repositories exploding from 2015-17 and wants to know why. More of a national push in Japan (whereas in the US it’s more grassroots). Previous work not looking at research in Japan especially on big picture scale.

Isomorphism – regulatory/coercive pressures; cognitive/mimetic pressures; normative pressures.

Collecting data from OpenDOAR and ROARMAP – content analysis of themes, mandates, core beliefs. Then interviews with SPARC in Tokyo, librarians, faculties. National Institute for Informatics has a shared cloud server with IR architecture so limited resources  not a barrier. OA policies have started very recently, but librarians play a major role in getting deposits even though only in that role briefly and assist directly. You don’t have to be a PhD to work in faculty so tenure and promotion completely different – publishing isn’t connected to tenure.

The role of the repository in increasing the reach and influence of research by Belinda Tiffen, Kate Byrne

(Acknowledge work of Catherine Williams). Repository enables reporting and assessment but also shopfront to sell research to the world. These roles don’t always sit well together – hard to explain to researchers why we want two versions of their papers.

What role does repository play in sharing research? Data from last year: 2609 UTS publications from Scopus. 33% also in repository. Looked at Almetrics for engagement. 1000 (of the 2609) have an Altmetric score. 47% in both Scopus and repository have an Altmetric score. 63% also in other repositories have an Altmetric score. But only 34% of outputs only in Scopus have an Altmetric score.

What will UTS do with this data? Have OA policy (since 2014) which has increased IR content but still only 35%. In 2015 rolled out new user interface. Active training to get authors to deposit. Want to find out which interventions are having an impact.

(In Q&A: redesigned theme done by part-time graphic designer in library, and in-house DSpace developer.

Scholarly Identity and Author Rights: guiding scholars as they make choices with their scholarly identities in a messy world by Jen Green

Project to managing scholarly/research identity. Schol comm team with wider working group and work with research community to focus outreach efforts. Workshop attendees mostly faculty and postgrads – but this changed when they started talking about online identity.

Workshops on ORCID – in the absence of a repository seemed a good place to start. Short pop-up sessions with ice-cream worked well, chatted, they created an ORCID.

Thought they needed help creating ORCIDs. Learned they needed that plus managing professional identity online and helping their student too. Scholars have limited time and want to spend it on own goals which may not match institution’s.

“Your Research Identity” – covered Twitter, Facebook, etc – so they’d know their online identity exists whether they manage it or not, and here are tools to manage it. Started with Google search on their names, discussed results. When results come up with other people with same name bring up ORCID. Suggested creating one place everything else can link back to (eg her own website).

Outcomes: after this workshop, workshops began to fill up. Once accidentally sent invite to whole campus and 30 seats filled in 10 minutes. Audience didn’t know existed mostly support staff.

(In Q&A: many faculty had never googled themselves, or didn’t know different results with different IP addresses.)

The University of the Philippines Baguio Faculty Research Database: starting a university repository by Cristina Borja Villanueva, Jay Mendoza Mapalo

Cordillera region home to country’s second largest concentration of indigenous people with 7 major ethnolinguistic groups. At Uof Philippines Baguio research is a priority. and library needs to collect and make outputs available to wider community.

Faculty Research Database started in June 2012, launched 2013, 500+ entries to document and disseminate outputs, increase citations, and advance knowledge. Use Joomla. Search author (by dropdown menu). Results page shows number of visits for each item – stats available to show most viewed. Item page may have full-text or may say available on request from author.

Has accomplished availability objectives. Hope to continue improve repository.

Crosswalks, mapping tables, and normalisation rules: when we don’t even share the same vocabulary for authority control by Deborah Fitchett

That’s me! So I didn’t summarise; instead see my full slides and notes.

Integrating DSpace #or2017


Harvesting a Rich Crop: Research Publications and Cultural Collections in DSpace by Andrew Veal, Jane Miller

Currently DSpace v3.2, Repository Tools 1.7.1; upgrading to DSpace v5.6, RT 1.7.4

Wanted independent identity for each major collection area especially research publications and cultural collections; and to avoid weirdly mixed search results – so decided on a multi-tenancy approach. Four repositories on four domains. So could make customisations appropriate to specific collections.

  • research publications (via Elements and self-deposit for theses)
  • cultural collections (digitised; populated by OAI from archives collection and by bulk ingests via csv)
    • 77,000 records: pdf, images, architectural drawings, complete books, audio, video which requires specific display options. Collections based on ownership/subject. Files stored in external archive with metadata stored in DSpace and linking back to file; thumbnail generated on the file.
    • AusStage pilot project – relational index (contributors, productions) linked with digital assets (reviews, photos, video). So eg an event record has a “digital assets” link which brings back a search based on an id shared by related records.
    • Created custom “melbourne.sorting.key” field to enable different sort orders eg for maps where date of accession is irrelevant.
  • coursework resources (eg past exams; architectural drawings for a specific course) – no sitemap or OAI feed
  • admin collections (for ERA)

Couldn’t have done it without service provider (Atmire). Have done lots of business analysis to say what they want, for Atmire to set up. Downside of success is now stakeholders thinks it’s easy to fix anything!


  • develop gallery/lightbox interface
  • upgrade to 5.6; improve Google Scholar exposure
  • OAI harvesting of additional cultural collections
  • look at thesis preservation via Archivematica

DSpace in the centre by Peter Matthew Sutton-Long

Acknowledges Dr Agustina Martinez-Garcia who did much of the integrations work

[Follows up a bit on Arthur Smith’s presentation earlier so I won’t repeat too much background from there.] Before integration, had separate systems for OA publication and research dataset submissions, e-thesis submissions, Apollo repository, CRIS system for REF. This meant a lot of copy-pasting for admins from the manual submission form into repository submission for. And researchers had to enter data in CRIS (Elements) as well as submitting for repository! Also hard to report on eg author collaborations.

Approved June 2016 to integrate things to meet OA requirements, monitor compliance, help researchers share data, allow electronic deposit of theses, integrate systems with community-drive standards for the dissemination of research activities data.

Item deposited in Elements to repository via repository tools connector (though not all files are passed through). An e-theses system feeds into the repository too. Zendesk is also integrated – any deposit creates a Zendesk ticket, which can be used for communication with researchers.

Researchers can work with a single system. They can add grants and links to publications, link to their ORCID profiles (though they don’t seem to want to), obtain DOIs for every dataset and publication (so some people submit old data just to get this DOI; or submit data early, or submit a placeholder to get a DOI they can cite in their article).

Fewer systems for team to access and manage, enhanced internal workflows.

In future want to integrate VIVO.

DSpace for Cultural Heritage: adding support for images visualization,audio/video streaming and enhancing the data model by Andrea Bollini, Claudio Cortese, Giuseppe Digilio, Riccardo Fazio, Emilia Groppo, Susanna Mornati, Luigi Andrea Pascarelli, Matteo Perelli

DSpace-GLAM built by 4Science as an extension to DSpace, which started from discussions around challenges faced by digital humanities. Have to deal with different typologies, formats, structures, scales – and that’s only the first level of complexity. In addition, most data are created/collected by people (not instruments) so affected by personality, place, time, and may be fragmentary, biased. Has to be analysed with contextual information.

How to do this in a digital library management system? Need tools for:

  • modelling, visualising, analysing – quantitatively and qualitatively, and collaboratively
  • highlighting relationships between data
  • explaining interpretations
  • entering the workflow/network scholars are working in

DSpace-GLAM built on top of DSpace and DSpace-CRIS.

  • Flexible/extensible data model – persons, families, events, places, concepts. When you create a “creator-of” relationship, it automatically creates the inverse “created-by” relationship. Can be extended to work with special metadata standards. By setting these up you can see relationships between people, events, etc.
  • with various add-ons
    • IIIF compliant image viewer addon with presentation API, image API, search API, authentication API coming soon. Gives a “See online” option (instead of just downloading) which shows the image, or PDF, or… in an integrated Universal Viewer player: a smooth interaction with the image, alongside metadata about the object, and linking with the OCR/transcription (including right-to-left writing systems). Sharing and reusing with proper attribution.
    • Audio/video streaming with open source stack: transcoding, adaptive streaming, mpeg-dash standard. DASH standard protocol lets you share video along with access to server to provide zoom to make sure the content stays in the digital library so complete access to stats and ensure people see ownership.
    • Visualising and analysing datasets by integrating with CKAN to use grids, graphs, maps.

Extending DSpace #or2017


Archiving Sensitive Data by Bram Luyten, Tom Desair

Unfortunately not all repository is equally open – different risks and considerations.

Have set up metadata-based access control. DSpace authorisation works okay for dozens of groups, but doesn’t scale well to an entire country where each person is an authorisation group – as they needed. Needed an exact match between a social security number/email address on the eperson and on the item metadata.

  • Advantages: Scales up massively – no identifiable limits on number of ePeople/items, no identifiable effect on performance. Can be managed outside of DSpace so both people and items can be sourced externally.
  • Disadvantages: your metadata becomes even more sensitive. And the rights to modify metadata also gives you the rights to edit authorisations.

Strategies for dealing with sensitive data:

  • Consider probability and impact of each risk eg
    • Data breach
      • Impact – higher if you’re dealing with sensitive data
      • Probability – lower if it’s harder for people to access your system and security updates are frequent
    • Losing all data
      • Impact – high if dealing with data that only exists in one place
      • Probability – depends on how you define “losing” and “all”. Different scenarios have different probabilities

Code on https://github.com/milieuinfo/dspace54-atmire/ (but documentation in Dutch)

(In Q&A: DSpace used basically as back-end, users would only access it through another front-end.)

Full integration of Piwik analytics platform into DSpace by Jozef Misutka

User statistics really important if defined reasonably and interpreted correctly. Difference between lines in your access logs, item views, bitstream views (which includes community logos), excluding bots, identifying repeat visitors.

DSpace stats

  • internal based on logs
  • SOLR (time, ip, useragent, epersonid, (isBot), geo, etc etc) – specifics for workflow events and search queries; visits vs views confusion
  • ElasticSearch time, ip, user agent, geo, (isBot), dso) deprecated in v6
  • Google Analytics, plugins

DSpace are good for local purposes but for seeing user behaviour lacks a lot of functionality

Had various requirements – reasonable, comparable statistics; separate stats for machine interfaces; ability to analyse user behaviour, search engine reports; export and display for 3rd parties; custom reports for specific time/user slices.

Used Piwik Open Analytics Platform (similar to Google Analytics but you own it – of course means one more web app to maintain). Integrate via DSpace REST API or via lindat piwik proxy. Users can sign up to monthly reports (but needs more work, and users need to understand the definitions more.)

In DSpace, when user visits a webpage, java code is executed on the server, which triggers a statistics update. But with Piwik, the java code is executed and returns html with images/javascript and when that’s executed it triggers statistics update – potentially including a lot more information; this excludes most bots who don’t execute js or view images.

If you change how you count stats, include a transition period where you report both the old method and the new method.

Code at: https://github.com/ufal/clarin-dspace

Beyond simple usage of handles – PIDs in Dspace by Ondřej Košarko

Handles often used as a metadata field for the reader to use to refer back to the item, or for relations between items. But while a human can click on the handle link, but a machine doesn’t know what it’s for or what’s on the other end of it.

Ideally a PID is all you need: You can ask ‘the authority’ for information of not just where to go but what’s on the other end.

Handles can do more things:

  • split up the id and location of a resource
  • content negotiation – different resource representation based on Accept header (which is passed by hdl.handle.net proxy)
  • template/parts handles – register one base handle but use it with extensions eg http://hdl.handle.net/11372/LRT-22822@XML or http://hdl.handle.net/11372/LRT-22822@format=cmdi – can refer to different bitstreams, or to different points in audio/video
  • get metadata based on handle without going right to the landing page – eg to get info in json, or generate citations, or show a generated landing page on error, or…

The Request a Copy Button: Hype and Reality by Steve Donaldson, Rui Guo, Kate Miller, Andrea Schweer

Trying to keep IR as close to 100% full-text as possible; DSPace 5 XMLUI, Mirage 2, now fed from CRIS Elements.

Request a copy button designed to give researchers an alternative way to access restricted (eg embargoed) content. Reader clicks button, initiates request to author; if author grants request, files are emailed to requester. Idea is that authors sharing work on one-to-one is covered by a) tradition of sharing and b) fair dealing under copyright.

Why? Embargoes auto-lift but no indication showing when that’d happen. Working on indicating auto-lift date but wanted to embrace other ways. Did need to make some tweaks to default functionality.

Out-of-the-box two variants: a) request author directly (but all submissions from Elements!) or b) email to admin who can review and overwrite author email address and and and … They used this latter ‘helpdesk’ variant but cuts out steps (and copies in admins).

So admins can use local knowledge of who to contact, intercept spam requests, aren’t responsible for granting access (to avoid legal issues).

Other tweaks:  to show file release date; wording tweaks; some behaviour tweaks (don’t ask if they want all files if there’s only 1; don’t ask the author to make the file OA because in our case the embargo can’t be overridden).

Went live – and lots of very dodgy requests – even for items that didn’t exist. So put in tweaks to cut down on spam requests to ensure requests actually come via the form, not a crawler. Respond appropriately to nonsense requests (for non-existing/deleting files); avoid oversharing eg files of withdrawn items or files not in the ORIGINAL bundle.

Went live-live. Informed of requests as made and approved, and added counts to admin statistics (requests made/approved/rejected).

Live a year

  • 9% of publications (49 items) have the button
  • 18 requests made – mostly local subject matter, majority from personal e-mail addresses
  • 8 approved, 0 denied (presumably ignored – though once author misread email and manually sent the item) – seemed like personal messages may have had a higher chance of success. One item was requested twice

So hasn’t revolutionised service but food for thought.

  • Add reminders for outstanding requests
  • Find out why authors didn’t grant access, maybe redesign email
  • Extend feature to facilitate control over sensitive items
  • Extend to restricted theses (currently completely hidden)

One good comment from academics “great service, really useful”.

(ANU also implemented – some love it, some hate it – if hate it change email address to repository admin.)

(Haven’t yet added it to metadata-only files – as it’d mean the author would have to find the file. OTOH would be a great prompt ‘Oh I should find the file’. Another uni has done this but had to disable for some types due to a flood of requests.)

Repository admin and integration #or2017


Auditing your digital repository(ies): the U-M Library migration experience by Kat Hagedorn

Have over 275 collections to handle: needed to audit prior to migrating. Also needed to know how the repositories interact and how the systems work.

Determined minimum and maximum factors for the audit, and divided into qualitative and quantitative. Quantitative easy to fill in, harder to figure out relevance.

Ran a pilot – including some problematic ones. Finding:

  • Even “number of objects” needs conversation in some collections. May need carefully thought out words – do you count the object or the record?
  • “Collection staleness” not ‘last updated’ but ‘last used’. And ‘update dates’ changed once to date of migration…
  • Some information about collections is only in email. Data often unclear even when locatable.
  • Technical issues – broken script meant collection usage appeared to be nil. Options for format types didn’t originally include XML.
  • Sometimes a stakeholder has been non-responsive in providing better images.

Mind the gap! Reflections on the state of repository data harvesting by Simeon Warner

A long time when 10GB was a lot… OAI-PMH was formed. It works, scales, is easy, is widely deployed. Harvested into aggregators, discovery layers etc. But not RESTful, clunky, focused on metadata and pull-based.

So we hate it, but don’t know what to do instead!

New approach has to meet existing use cases; support content as well as metadata, scale better, follow standards, make developers happy. Need to be able to push for more frequent updates.

Wants to use ResourceSync – ANSI/NISO Z39.99-2017 – has WebSub (was PubSubHubbub) companion standard.

CORE is looking at replacing OAI-PMH with ResourceSync. Work with Hyku & DPLA. Samvera (was Hydra) building native ResourceSync support).

The community should agree on ResourceSync as a new shared approach. Have to support it as primary harvesting support, OAI-PMH as secondary for transition.

(In Q&A: Haven’t yet talked to discovery layers – need to get consensus in community first. Trove have switched some to SiteMaps (basic level of ResourceSync).

Audience suggestion to create user stories of migration.)

Does Curation Impact the Quality of Metadata in Data Repositories? by Amy Elizabeth Neeser, Linda Newman

Various research questions about metadata options and curation. Compared 4 institutions. Looked at most recent 20 datasets in each. Each institution’s metadata is very different, eg ‘author’ vs ‘creator’; some automatically generated (which were discounting). Finding:

  • Sizable variation of metadata ‘options’ per institution
  • Choice to curate doesn’t necessarily guarantee more metadata. More than minimum is available regardless
  • Documentation is far less common in self-submission repositories, usually only a readme
  • Institutions who curate can ensure that each dataset gets a DOI, others currently leave the choice to the user. (May just be related to policy though)
  • Not sure whether placement of input form or curation is the bigger factor in number of keywords

(In Q&A: As a community we think curation is good but want some proof of this to justify all the hours! Also as a result of the study have made changes to own practices eg better input forms.)

Leading the Charge: Supporting Staff and Repository Development at the University of Glasgow by Susan Ashworth

What does repository success look like? Enlighten is a recognised brand that covers all the services. Multiplying repositories as users keep asking for them. Recognition of value of data – populates web pages, research evaluation, benchmarking, KPIs.

How did they get there? Early and ongoing engagement with deans of research and research admins. Surface data publically on researcher profiles. Say yes (and panic afterwards) eg improving reporting from the repository. Adapt quickly to external forces (funding and govt requirements).

What does this mean for libraries? Cross-library services, appointing new staff. Have 6 OA staff in various teams. Developing new skills in data management, licensing, metrics, etc. Lots to do leading to lots of opportunities for staff. Staff can easily see the contribution they make to institution. Clear when service has to deliver on high expectations.

UK-wide adoption of “UK Scholarly Communications License” also used/adapted from Harvard – where unis retain some control over outputs to make them available at point of publication instead of embargos. [Seems to be adapted from CC-BY??]

(In Q&A: 70 UK institutions discussing adoptions of this license – may be declarations in Open Access Week. Some pushback from publishers but Harvard have been able to work with it for 10 years!)

Towards an understanding of Open Access impact: beyond academia by Dr Pauline Zardo, Associate Professor Nicolas Suzor

Most research into OA usefulness is focused on use for researchers. Worked in govt and wanted to use research evidence but couldn’t get access. Did PhD on how to help govt use research evidence in decision-making. Increasingly important in context of govt impact assessments eg REF, PBRF, ERA. Access (and knowing it exists!) is the biggest, structural barrier to using academic information.

“The Conversation” website aims to help researchers communicate research in a way that’s easy for lay people to understand. Free to read and share under Creative Commons. 4 sectors reading it: research and academia; teaching and education; govt and policy; health and medicine together represented 50% of survey participants. Value academic expertise, research finding, clarity of writing, no commercial agenda, editorial independence. Discuss it with friends or colleagues afterwards, many share on social media. Used in discussions and debate; may change behaviour; some use it to inform decision-making (or to support an existing decision….)

(In Q&A: To do more research with pop-up surveys on downloads from IRs to find out why people are using the content.)

Batch processes and outreach for faculty work by Colleen Elizabeth Lyon

UT at Austin – research-intensive. IR on DSpace. Big campus, competing priorities = lots of missing content. Wanted to increase access to content and improve outreach skills using existing repository staff (2 full time plus a few grad students to upload content).

Use CC licenses and publisher policies to identify which publishers allow it without having to ask faculty permissions, and create automated process:

Export content from WoS to Endnote -> csv -> Google Sheets > cf SherpaRomeo via API -> download articles -> use SAFCreator to get into right format for batch import to DSpace (followed by stuff for usage reports for faculty). Results:

  • Filtering and deduping took more time than expected.
  • Faculty didn’t respond to notifications – not sure if they just ignored, didn’t think it needed a response – but at least no complaints!
  • Added almost 2500 items between other projects.

Under the DuraSpace Umbrella: A Framework for Open Repository Project Support by Carol Minton Morris, Valorie Hollister, Debra Hanken Kurtz, Andrew Woods, David Wilcox

DuraSpace is a not-for-profit org so mission to support open tech projects to provide long-term durable access and discovery. Fosters projects eg DSpace, Fedora, Vivo. Offers services eg tech/business expertise, membership/governance framework, infrastructure, marketing/comms. Affiliate project Samvera.

Ecosystem is larger so want to expand support to ensure community/financial/technical sustainability. Criteria include: philosophical alignment; strategic importance to community; financially viable; technical pieces in place. So if you know of a project needing support, contact them.

A Simple Method for Exposing Repository Content on Institutional Websites by Gonzalo Luján Villarreal, Paula Salamone Lacunza, María Marta Vila, Marisa Raquel De Giusti, Ezequiel Manzur

2 institutions with active IRs but questionable web dissemination practices – lack of interest/staff/time/money to maintain web presence for staff profiles.

  • Improved existing sites with Choique CMS, WordPress, Joomla
  • Designed and developed new sites with WordPress multisite – hosted if research centre didn’t have their own site
  • Gave advice about web publishing
  • Use IRs to boost websites by using OpenSearch to retrieve contents, then software library to fetch/filter/organise/deliver and share results, then created software addons for each CMS
    • Easy configuration
    • flexible usage eg “all research centre publications” or “theses from the last 5 years” or “all researcher A’s content”

Results: 7 new websites published, 14 in development. Researchers want to deposit in IRs to keep the website updated. Dev work to continue eg flushing cache; multi-repository retrieval?

Research and non-publications repositories, Open Science #or2017


OpenAIRE-Connect: Open Science as a Service for repositories and research communities by Paolo Manghi, Pedro Principe, Anthony Ross-Hellauer, Natalia Manola

Project 2017-19 with 11 partners (technical, research communities, content providers) to extend technological services and networking bridges – creating open science services and building communities. Want to support reuse/reproducibility and transparent evaluation around research literature and research data, during the scientific process and in publishing artefacts and packages of artefacts.

Barriers – repositories lack support (eg integration, links between repositories). OpenAIRE want to facilitate new vision so providing “Open Science as a Service” – research community dashboard with variety of functions and catch-all broker service.

RDM skills training at the University of Oslo by Elin Stangeland

Researchers using random storage solutions and don’t really know what they’re doing. Need to improve their skills. Have been setting up training for various groups in organisation. Software Carpentry for young researchers to make their work more productive and reliable. 2-day workshops which are discipline-specific and well-attended. Now running their own instructor training which allows expanding service. Author carpentry, data carpentry, etc.

Training for research support staff who are first port of call on data management plans, data protection, basic data management. Recently made mandatory by Dept of GeoSciences to attend DMP training.

Expanding library carpentry to national level.

IIIF Community Activities and Open Invitation by Sheila Rabun

Global community that develops shared APIs for web-based image delivery; implements that in software; to expose interoperabie image content.

Many image repositories are effectively silos. IIIF APIs allows a layer that lets servers talk to each other and allow easier management and better functionality for end-users. Lots of image servers and clients around now so you can mix-and-match your front and back-ends. Can have deep zoom; compare images and more.

Everything created by global community so always looking for more participants. Community groups, technical specification groups eg extending to AV resources, discovery, text granularity (in text annotations). Also a consortium to provide leadership and communication channels.

Data Management and Archival Needs of the Patagonian Right Whale Program Data by Harish Maringanti, Daureen Nesdill, Victoria Rowntree

Importance of curating legacy datasets. World’s longest continuous study of large whale species: 47 years and counting of data. Two problems:

  • to identify whales – found the callosities of right whales were unique (number, position, shape) and pattern remained same despite slight change over time. So can take aerial photos when they surface. Data analysed with complicated computer system and compared with existing photos.
  • to gather data over a period of times – where to find whales regularly. Discovered whales gather in three places: 1) mothers and calves; 2) males and females; 3) free-for-all.

Collection has tens of thousands b&w negatives; color slides; analysis notebooks; field notes; Access 1996 database records; sightings maps.

Challenges: heterogeneity of data; metadata – including how much can be displayed publically; outdated databases.

Why should libraries care? We can provide continuity beyond life of individual researchers. Legacy data is as important as current data in biodiversity type fields and generally isn’t digitised yet.

Repository driven by the data journal: real practices from China Scientific Data by Lili Zhang, Jianhui Li, Yanfei Hou

China Scientific Data is a multidisciplinary journal publishing data papers – raw data and derived datasets. Submission (of paper and dataset), review (paper review and curation check), peer review, editorial voting.

How to publish:

  • Massive data? – on-demand sample data publication: can’t publish the whole set, but publishes a sample (typical, minimum sized) to announce the dataset’s existence
  • Complex data? – publish data and supplementary materials together eg background info, software, vocabulary, code, etc. Eg selected font collections for minority languages
  • Dynamic data? – eg when updating with new data using same methodology and data quality control. Could publish as new paper but it’s duplicative so published instead as another version with same DOI. Can be good for your citations!

Encourage authors to store data in their repository so its long-term availability is more reliable.

RDM and the IR: Don’t Reuse and Recycle – Reimplement by Dermot Frost, Rebecca Grant

We all have IRs and they’re designed for PDF publications. Research Data Management is largely driven by funder mandates; some disciplines are very good at it, some less so (eg historians claiming “I have no data” – having just finished a large project including land ownership registries from 17th century, georectified etc!)

FAIR (findable, accessible, interoperable, reusable) data concept (primarily machine-oriented ie findable by machines). IRs can’t do this well enough. Technically uploading a zip file is FAIR but time-costly to user.

Instead should find a domain-specific repository (and read the terms and conditions carefully especially around preservation!) Or implement your own institutional data repository (but different scale of data storage can take serious engineering efforts). Follow the Research Data Alliance.

Developing a university wide integrated Data Management Planning system by Rebecca Deuble, Andrew Janke, Helen Morgan, Nigel Ward

Need to help researcher across the life-cycle. UofQueensland identifying opportunity to support researchers around funding/journal requirements. Used DMP Online but poor uptake due to lack of mandate. UQ Research Data Manager system:

  • Record developed by research – active record (not plan though includes project info) which can change over course of project. Simple dynamic form, tailored to researchers, with guidance for each field.
  • Storage auto-allocated by storage providers for working data – given a mapped drive accessible by national collaborators (hopefully international soon) using code provided in completing the form.
  • [Working on this part] Publish and manage selected data to managed collection (UQ eSpace). Currently manual process filling in form with metadata fields in eSpace. Potential to transfer metadata from RDM system to eSpace.
  • Developing procedures to support the system.

Benefits include uni oversight of research in progress, researcher-centric, improves impact/citation, provides access of data to public.

Preserving and reusing high-energy-physics data analyses by Sünje Dallmeier-Tiessen, Robin Lynnette Dasler, Pamfilos Fokianos, Jiří Kunčar, Artemis Lavasa, Annemarie Mattmann, Diego Rodríguez Rodríguez, Tibor Šimko, Anna Trzcinska, Ioannis Tsanaktsidis

Data very valuable – data published even 15 years after funding stopped, and although always building new and bigger colliders, data is still relevant even decades after collected.

Projects involve 3000 people, including high turnover of young researchers. CERN need to capture everything need to understand and rerun an analysis years later – data, software, environment, workflow, context, documentation.

  • Invenio (JSON schema with lots of domain-specific fields) to describe analysis
  • Capture and preserve analysis elements
  • Reusing – need to reinstantiate the environment and execute the analysis on the cloud.

REANA = REusable ANAlyses supports collaboration and multiple scenarios.

Electronic Poster Display #or2017

There were lots of fantastic posters, these are just the ones I wanted to refer back to as they sparked thoughts I want to followup on. In no particular order:

  • Governmental Educational Repository in Health – they have 7000+ open access learning objects. [We have 175. Which isn’t nothing. But actually what I’m still mostly interested in is whether anyone’s ever going to develop an aggregator for OA learning objects….]

  • Extending the value of the institutional repository with metrics integration – they’ve got individual researcher profiles showing metrics. [We’ve got some of this in Elements. To get the rest though would require coding, and dealing with authentication to keep it private to the researcher. I recently wrote an authentication module for a hand-coded php/sql app using EZproxy which I could adapt to something like this – or any other homegrown personalisation effort.]

  • COAR Resource Type Controlled Vocabulary: Dspace Prototype implementation – [I saw (and gave feedback on) a draft of this a while back; should have a look at the latest version (v1.1) and check how it maps (or doesn’t) to PBRF types]

  • The PLACE Toolkit: exposing geospatial ready digital collections – [what value would there be, in our own collections, of adding time/location metadata to content to enable eg map/timeline exploration? (and therefore would it outweigh the cost?)]

  • COR(E)CID: Analysing the use of unique author identifiers in repositories via CORE to support the uptake of ORCID iDs – this gives repositories a dashboard to check how many ORCIDs are in their repository. [I wasn’t clear though on whether it’s available for public use or requires a sign-up. Further investigation shows the CORE Repository Dashboard does require registration and is specifically for repositories submitting data to CORE, which makes sense.]

  • International Image Interoperability Framework (IIIF) – [this is beyond my expertise but I want to check it’s on the radar of our non-research-output repository vendor]

  • If you digitise them they will come: creating a discoverable and accessible thesis collection – U of Tasmania made their theses open access retrospectively if at least 10years old, with a disclaimer. [I’ve heard of a number of universities doing similarly; we’ve been more conservative, only making them available to staff and students unless we can secure permission. I’d like to push for the more open model.]

  • Strategies for increasing the amount of open access content in your repository – one tip they suggest is to set up a ScienceDirect email alert for ‘accepted manuscripts’ at your institution. When you get the email, download it immediately before it gets replaced by the ScienceDirect-branded ‘in press’ version. [This. Is. Genius.]

  • Enabling collaborative review with the DSpace configurable workflow – [I did some javascript hacking of the workflow, to sort the items by age and allow other sorting, but there’s very limited information still.] This poster shows improvements like displaying extra metadata fields (eg item type – author/publisher/year might be useful for us), adding statuses (eg questions for the researcher), and adding other notes. [This is Relevant To Our Interests.]