Apparently this isn’t as straightforward as the DSpace documentation might suggest, and further problems may arise when installing patches/plugins etc as most are designed with a PostgreSQL database in mind. This is a problem as it negates one of the main reasons for choosing DSpace, that of being able to use and feed back into community developments.

The web team are think-tanking this to work out a solution. We’d like to be able to extend our repository to also work with a publications database, so that’s an additional issue to take into consideration.

Any comments or suggestions appreciated..

Snapped at home – you can just about make out a pheasant strolling through the brambles.. This little guy has really been making himself at home around our place lately. Caught him strutting along the patio the other morning. I wouldn’t be bothered except pheasants are so pea-brained! As long as he doesn’t try any ‘why did the pheasant cross the road’ type stuff because pheasants don’t seem to be very successful at that…

I’ve just read Richard Jones’ and Fereshteh Afshari’s research paper on their work developing an integrated repository at Imperial College in London. This is very much the direction we’re moving in, so it’s great to read about how they’ve developed their systems. They’re also DSpace users, with a personal information portal gathering research expertise and are working on tying the two together. Recommended reading, and many thanks to the authors for sharing their experiences.

ps. -5 celcius when we left the house this morning – weee!

Okay, off to a steady start – over ten items in the repository. Hopefully more to come soon. I’ve got a spreadsheet of possible authors from Web of Science and BMC. I’m checking these against the PIP, our research expertise portal. On the side I’m contacting heads of research centres and institutes. The School of Management has a great list of research publications for their staff.
Ignore what you read – so far this part has been easier than the technical set up.

You might think that an embargo function for items that can’t be made publicly available would be pretty standard in a digital library software – think publisher embargoes, theses embargoes, etc. All standard practice. Does an embargo function come built in with DSpace though? Not that I can see, and I can only find this information on Embargo on Bitstream V2 on the DSpace wiki. Very disappointing. I’m happy to be proved wrong, especially if it means there’s one less piece of code to tweak.

The RSP team held a DSpace technical day here in Bath on Monday 12th November. It was a great chance to catch up with a few people from the RSP summer school from back in June, and to ask a few questions about DSpace. Interesting points
- There’s a patch for DSpace 1.4 (included in 1.5) for IP authentication – this would be really useful for UG dissertations or masters dissertations – could these be available on campus only?
- The number that displays beside a collection title showing the number of collection items – these are called ’strengths’ (??!) Should be a simple matter of switching this on..
- Keep log files separately – probably a standard technical procedure – saves disk space in case of repeating errors
- It’s possible to preset ‘Bath’ as the institution for theses.
- Register with ROAR (done). Southampton peeps also wrote the irstats package – sounds good, particularly as it recognises robot/spider visits to a site..

Statue near Iford Manor, near Bath – great walk between here and Farleigh Hungerford castle (remains of..!)

Well I’ll be reporting live (almost) from the DSpace Technical Day the RSP are holding here on Monday 12 November. They’ll be running a basic session in the morning, and advanced tutorial in the morning, with a surgery running parallel, presumably where you can drop in to see Dr DSpace.
Meanwhile, back at the ranch hopefully we’re near the end for the submission form. Surely you can concatenate fields in DSpace into the identifier.citation field? Fingers crossed. I’m sure this is what Hasselt University did with their super-deluxe input form..

I think I could probably be a case study in a book called Metadata For The Terrified. Or Metadata For Dummies.
Over the last week or two I’ve been coming to terms with how the DSpace metadata works and relates to the input submissions form – the metadata is pretty good, I like how DSpace lets you add fields using a web form, but the submissions page is pretty dismal, IMHO. Using an idea from Mahendra in UKOLN, I’ve written out a spreadsheet listing which fields are needed for which item and come to the resounding conclusion that we need to customise the submission form to deposit by item type. A big ask, apparently, although some good examples out there, such as Hasselt University, who presented on submission based on document type at the DSpace User Group meeting in Rome in October.
Yes, I know, ePrints does this nicely. Point taken.

It’s the little things that sometimes take ages.. like deciding how to describe the communities and sub-communities for our repository installation. We’re a University, so the faculties and schools are easy enough, and I can put the research centres and institutes together but what about ‘the leftovers’? What is the collective noun for the administrative services and support units? Everyone I spoke with had a different opinion.. Well I’ve finally found a page on the University website that covers ‘Administrative and Central Services’, so that’s what it is.. for now..

I’ve also spent bits of the last day or so tidying up our deposit licence and notice and takedown agreement. It’s coming along, and I’m writing up a checklist of things to consider before depositing – this is what I have so far:

1 Are you the owner of the copyright for this item? If not, do you have your co-author’s permission to deposit? Asking them to agree to the Deposit Licence is sufficient, but as co-owners of the copyright you must have their permission. (Charles Oppenheim made a good suggestion in response to a question on this today on a mailing list – agree to allow authors to self-archive before publishing..)

2 If the item is a journal article, does the journal publisher permit self-archiving? Have you checked RoMEO? Alternatively, have you negotiated the right to self-archive a copy of your work? There is more information available on the OPuS wiki (or there will be!).

3 If the item was sponsored by industry, do you have permission to make a copy of the item publicly available? You may wish to embargo the item to meet sponsor requirements, or check with your supervisor or Research Support Unit to make sure you meet contractual obligations.

4 Do you intend to publish this content commercially for the purposes of making a profit? If so you may wish to reconsider your deposit. The Bath OPuS is designed to make material widely accessible, enhancing impact, not necessarily income.

5 Is the item in a file format that is readily recognisable? Unusual file formats may create preservation difficulties and it is wise to discuss this with the repository manager.

What else needs to go in? Hmm.

The Repository Support Project has just advertised a suite of events for repository practitioners in the UK over the next few months. Topics for the Professional Briefing and Networking event series include: Repository developments; Metadata issues; Stakeholder roles & perceptions and more to be announced.
Additionally, here at Bath there will be a DSpace Technical day on November 12, 2007. This provides DSpace users a similar event to that held recently for ePrints, where there will be the opportunity to attend a DSpace Surgery and ask installation specific questions. Alongside this are basic and advanced tutorials. I’m really pleased to see the RSP getting these events up and running, and I’ve just discovered their website is becoming more fleshed out with information useful for repository start-ups.