I say cataloguing, you say metadata - let's sort the whole thing out

This was a talk and demonstration organised by the Office of Scholarly Communication (OSC) and the Open Access team, presented by Danny Kingsley (Head of Scholarly Communication), Sarah Middle and Philippa Grimstone (Repository Assistant).
Sarah Middle discussed the changes taking place in the institutional repository. Presently known as Dspace, the repository is going to be upgraded and renamed Apollo.

The repository structure is based around communities, which is a coverall term to describe all the departments, faculties and other institutions within the university. Within each community there may be sub-communities (for example for specific groups within a department), and three collections: scholarly works, research data, and theses. Currently the repository is browsable, and searchable by authors and titles, and better subject keyword searching is something being worked on, but it is also well-indexed by Google.
At the moment, it is compulsory for students to provide a print copy of their PhD thesis, but an electronic copy is optional (unless stated otherwise by funder of the PhD). An electronic copy can be self-deposited or done through the OSC.

There are sometimes issues preventing an author from submitting an electronic copy, such as licence agreements and 3rd party copyrights (e.g. in having used images or quotes). However, it is possible to put an embargo of up to 2 years delay on open access which allows time to sort these issues out.
Sarah outlined some projects and future plans for the repository. These include:
‘Unlocking Theses’
When someone requests a digital copy of a thesis, they pay for a copy which can be held in ‘dark space’ in the repository until the author approves it to be made available online (35% positive response rate).
‘Hero Theses’
The OA Team are proactively digitising the theses of famous alumni of the university and holding them in dark space until the author approves its availability (49% positive response rate) – the intention is to encourage people to follow in their heroes’ footsteps and make their theses available online.
Future plans:

• Creating a “Request a Copy” button
• Providing a facility for author to claim works and consent to make available online
• E-theses pilot (taking place in selected departments only)

Philippa Grimstone began with a description of what was meant by ‘Open Access’, and the types of Open Access that exist.

	Green OA	Gold OA	Hybrid OA
Access through	Institutional Repository	Publishers' websites	Both IR and publisher
Cost	Free	Not free	Not free
Limits	Embargos (delays) and restrictions	Available immediately in final published form	May be embargos or restrictions applying

Journals can be fully open access, partial open access, or completely closed access.

In order for research to be eligible for the post-2014 Research Excellence Framework, it must comply with the policy of whichever group funded the research in the first place. HEFCE is the biggest funder, and say that peer-reviewed articles must be deposited in an IR within 3 months of its acceptance by a journal. There are other funders too, such as RCUK, COAF and Wellcome Trust, all with different policies.
Philippa then took the audience through the process of depositing in the IR, which is made up of 4 simple steps:

1. Submit author’s accepted article to www.openaccess.cam.ac.uk
2. Zendesk® ticket is generated
3. OA team picks up the ticket and deals with the submission
4. A record is created of the article within the IR

1. Submission
Which version of an article is acceptable?
Pre-print >> Goes for peer review >> Accepted for publication >> Final version is published
An article can't be accepted until it has reached the third stage (accepted for publication).
2. Ticket generation
Zendesk is the management software used by the OA team, which can be used for communication, will store information, and can generate reports.
3. Processing submission
The OA team deal with answering questions like:

• what is the funder’s policy on OA?
• what is the publisher’s policy on OA?
• what type of OA is used?
• will fees apply?
• is this the correct version of the article?

4. Repository record creation
This is done by the OA team based on the Dublin Core Metadata Initiative. There are still problems to iron out regarding author authority headings. In the future it is hoped that Dspace will be able to create automatic citations too.
Once the record is created, it is uploaded to a collection/community and granted a cc-by licence (the default, though others may apply). The author is then informed that their work is available.

Datasets can also be uploaded like articles, but there are more issues inherent:

• datasets aren’t published like articles, so may be unclear/incomplete
• there is no standard format for datasets or for metadata
• there may be proprietary software involved
• they may be large files (there is a fee for files larger than 1GB)
• they may be zipped files
• how much data constitutes ‘enough’? (e.g. hundreds of hours of viedo footage, or several samples?)
• there may be ethical concerns with making data available

Danny Kingsley rounded off the session with a a Q&A. She mentioned that there are lots of rumours and worries amongst researchers who may be reluctant to put their work online, but perhaps the majority of this is misinformation which can be dispelled by encouraging researchers to talk to the OSC.
Q. Are articles placed in the ‘dark space’ OA-compliant?
A. Yes. While not presently OA, there is a record within the IR of that article.
Q. Are there exceptions to this seemingly straightforward process?
A. Yes, and the OA team do their best to account for everything in each case.
Q. Will the approved manuscript (pre-publication) be retained in the IR even if the final version becomes available?
A. Yes, as this is the version owned by the university, but when the final version comes in, it will be appended as ‘additional information’ to the record of the final version rather than being a separate record.
Q. How do we find out if something is being made available on Open Access?
A. We need to be logged out of the university domain, then try to access the article.
Q. Why can’t we use our archive, why do we have to use the IR? (Faculties with own repositories)
A. You can still, at the moment, and the OSC is looking at ways of porting the data across, before ultimately supplanting the faculty repositories.
Q. How much does Open Access cost?
A. Millions. Gold OA can often cost £1500-£5000, with some major publishers charging £7000. This is in addition to the subscription cost (this is known as ‘double-dipping’, where the publishers charge the university twice – once to publish, once to access) and it is estimated that publishers like Elsevier and Wiley are scalping an extra £2m each out of the university in this way.
What next?
Having attended this talk, I looked at what the museum had deposited in the university IR: two newsletters. I’m sure there must be lots being published, as the staff here are incredibly scholarly and constantly researching things, so the next step is to find out what is happening to this research, and whether it can be shared with the university and beyond.

Update:

I reported back to my line manager and discussed the workshop, and found out some additional things to think about. The first is that museum staff are academic-related, not academic, therefore not subject to the same terms as academics when it comes to things like REF. This also probably has knock-on effects for things like funding, which doesn't pay for specific research so policies for OA may not come into play the same way. The other thing was that a lot of staff publish on academia.edu (not linked here), which is a venture-capital for-profit organisation, and they need to be aware that the data is mined for commercial purposes. It also doesn't count as OA for compliance purposes because it's behind a membership wall.