Notes from project team Skype call on 21st March 2011

Last week the project team held one of their regular (approximately monthly) Skype calls, to discuss progress to date and follow up on current work areas. The minutes make for interesting reading, since we covered quite a few different topics, ranging from our new workgroups approach for the final stages of the project to how much great discussion has been going on within the newly energised Google group. There was also quite a lot of talk about what real-life issues are arising now that some people have started their advocacy campaigns (for instance around the RoMEO colours) - and that we must keep a good log of what happens during the campaigns so we can report back on it at the end of the project.


posted by: Lizzie Dipple

Repositories in a new CRIS landscape - some discussion points from the RSP Winter School

Six weeks on from the RSP Winter School and time for some reflection on a few Reposit-relevant points we discussed at this very useful event.

1. Workflow. Our current Symplectic to Eprints workflow (Leeds) 'surrenders' control of metadata to the central research management system. Surrender may not be the right word, but we're used to having complete control within EPrints so it feels that way. It’s certainly an important workflow question to consider – and not all institutions have come to the same conclusion. If metadata is incorrect, is it fixed in Symplectic or fixed in EPrints? We fix in Symplectic. This means the Symplectic and EPrints records are in synch, but it means repository staff need to interact with two systems. I’m not suggesting this as an ideal model but it’s our reality at the moment. Where does your ‘master’ record live? Is it/ can it be locked down?
   
   Symplectic is now our only deposit route for White Rose Research Online content for University of Leeds. Will it continue to be? Are there other places we need to capture deposits?
   
   Overall, as CRIS/repository models roll out, I’d really like to see some more discussion around the pros and cons of different workflows and how life can be made easier for both depositors and repository staff. One comment that stuck with me was the high expectations researchers have (well, I suppose we all have) that systems should be simple and easy to use. Obvious – but difficult to achieve. And how do we (should we?) avoid systems which are simpler for the depositor but result in extra work for someone else further along the workflow (library/repository staff)?
   
   
2. CRIS/Repository landscape. We discussed the importance of being aware of – or even pre-empting – the CRIS discussion at your institution. If you’re not involved already, does this mean the discussion has started without you? .. or should you be starting the discussion?
   
   We’ve discussed for many a long year how institutional repositories fit with other repositories – arXiv; SSRN; RePec; UKPMC; ESRC etc – do we push data to them or pull data from them or both? Is the mechanism for this located within the CRIS, the repository, somewhere else? I was very happy to see progress with Repository Junction; we really need something to help rationalise multiple location deposit.
   
   
3. Relationship with Research Office. Knowing a/the key person can make all the difference. If you’re not already talking to your research office, start! We have several areas of shared interest even if we don’t always talk the same language.
   
   
4. Repository visibility vs invisibility. Not a new question, but should the repository be high profile and obvious – or so well integrated depositors don’t even know it’s there? One questions was raised – if everything is deposited via a CRIS won’t the CRIS get credit for all the behind the scenes work the library and repository staff may do? I don’t think this is an issue – or is it? Seems to me there’s a quid pro quo here – the Library is a trusted, established service – and I think Library involvement increases the credibility of CRIS type systems with researchers. (Discuss?!) Of course, this only works if the depositors know there’s a Library role – be it quality control, copyright advice, copyright checking, version advice, file format advice, preservation, user testing. From the Library perspective we need to understand our new role but we also need to publicise this – not something we’re always good at.
   

Symplectic Elements to Eprints connector in practice

At the moment, the EPrints connector for University of Leeds is configured to handle two different deposit scenarios:

A. Straightforward deposit
1. The author (or their ‘proxy’ depositor) adds a file to a publication entry in Symplectic Elements using the Full Text tab
2. The author (or their ‘proxy’ depositor) applies a deposit licence to record using the Full Text tab
3. The record appears in the EPrints Review area where it is processed by staff (metadata check, copyright check)
4. All being well, the metadata and file are made openly available (metadata only if we can’t clear copyright or a copyright permission is pending)

B. Deposit without licence
Depositors may not be happy to click the deposit licence straight away – they may want the Library to check the copyright status of the item first.
1. The author (or their ‘proxy’ depositor) adds a file to a publication entry in Symplectic Elements using the Full Text tab
2. The record appears in the EPrints Manage Deposits area where it is processed by staff (metadata check, copyright check)
3. Once the record has been processed and copyright clarified, the depositor is sent the URL of the relevant page in Symplectic Elements to apply the licence and/or asked to deposit a different version of the file in question e.g. the Accepted Version.
4. The record moves from EPrints Manage Deposits to EPrints Review once the licence is applied. The EPrint is then made live by library staff (with open full text if possible, if not, as metadata only)

Most deposits so far have followed one of these two scenarios.

However, we also have unexpected variations in depositor behaviour.

C. Metadata only
In theory, metadata only records do not feed from Symplectic Elements to EPrints (in our configuration). In practice, some depositors unexpectedly applied the deposit licence using the Symplectic Elements ‘Full Text’ tab even though they hadn’t uploaded a file. At the EPrints end, we get a metadata only record and scratch our heads wondering where the full text is. It’s a perfectly reasonable behaviour by the depositors. We’re not sure whether they’re just testing out the system or assuming that if they click the licence library staff will somehow get and disseminate the relevant file for them. Further investigation is required.

D. Files uploaded and deleted
1. The author (or their ‘proxy’ depositor) adds a file to a publication entry in Symplectic Elements using the Full Text tab
2. The author (or their ‘proxy’ depositor) then deletes the file in Symplectic Elements using the Full Text tab!
By this point, whether or not the licence has been applied, the record and attached files have already been sent to EPrints. In this case, we have a clue to what has happened because the attached file is ‘soft’ deleted in EPrints – i.e. it’s still attached the EPrint record but has the description ‘deleted’ and is not publicly visible. We are aware of one depositor who realised he’d uploaded the ‘wrong’ file – hence the deletion. However, a number of records show this pattern – are authors regularly uploading the wrong file or is there another reason for the behaviour? For example, are they put off by the deposit licence? Do they feel they need to delete their uploaded files for copyright/version reasons when they notice the publication policy information (from RoMEO) in Symplectic Elements? Further investigation needed.

We will keep monitoring behaviour and seek further feedback from depositors. On balance we should probably make it much more difficult to delete files from Symplectic Elements – or at least make it very clear to depositors that the record is transferred to the repository the moment they upload a file in Symplectic Elements.

Any views or comments on our workflows or on user behaviour would be most welcome.

JISC Repository Deposit Programme Meeting - 1st March

Last week RePosit was represented at the JISC Repository Deposit Programme meeting in Birmingham. Under the supervision of Balvair Notay, JISC manager for the Deposit projects, the following projects provided updates on their progress:

• DepositMO
• DURA
• Kultivate
• Open Access Repository Repository Junction (OA-RJ)
• Sword v2
• Sonex

As well as providing an opportunity for the projects to learn more about the work of each other, we were all also asked to explore three key questions and to report back, both at the meeting and via blog posts. So, below, are the three points, and my thoughts on them for RePosit!

What actions are required for the success of your project? What are the indicators?

These are based around the core deliverable of advocacy, ensuring the right level of engagement, to the right audience at the right time - moving from the planning of advocacy to the actual delivery and the use of materials developed so far. The delivery of the technical solutions for each partner institution also remains key. The indicators would be increased deposit and a correlation with the release of the embedded deposit tool. More broadly, an increased understanding of deposit, and the reasons for it, would also be an indicator of success.

What strategies are required to spread good practice?

Our strategies include the provision of advovacy materials (including institutional strategies), engagement with other projects and organisations (such as Kultivate and RSP), the provision of this very blog and ongoing testing and evaluation of project outputs.

How are you going to feed data for the evaluation?

RePosit is perhaps fortunate among the projects in that it can gather quantitative data via the monthly repository statistics collection, and can monitor this in relation to the release of embedded deposit tools, and the effect this may have on deposits. However, it is also important to collect qualitative data, via surveys and other similar activity, as a change of culture, of mindset, among academics is also an important output. The project will also make the data available to Evidence Base.

In summary, the programme meeting provided an excellent opportunity to meet the other projects, share good practice and mull over common issues.

For my part, I managed to avoid being photographed, but no such luck for Richard! He can be seen assisting Kultivate with their afternoon report in a very informative blog from the SONEX project.

Posted by: Ian Tilsed

Notes from project team Skype call on 28th Feb 2011

Last Monday we had our latest project Skype call, covering a number of outstanding topics for discussion, and here are the minutes. In brief: we pretty much finalised the advocacy materials designs from QMUL Creative Services; we discussed the necessity for images to be available for multiple re-use after the project ends; we tried to pin down our next steps for the survey (although since then we've had the complication of hearing about the RSP-created countrywide survey about open access); we talked further about our project dissemination event and the possibility of doing something jointly with RSP; and finally we came up with a plan for how to approach our newly created expert panel in the Google group discussion forum.


posted by: Lizzie Dipple