LiveBlog: Pecha Kucha Session 1

Ian Stuart is chairing our first Pecha Kucha session. As usual you can vote for your favourite – and no eating the chocolate coins we’re using to vote!

In this session we will be hearing from:

• Adrian Stevenson (UKOLN) – Linked Open Copac Archives Hub (LOCAH) project – use of Timemap for visualising linked data

I work on LOCAH which is part of #jiscexpo. The Archives Hub is a an aggregatiuon of archival descriptions, Copac is similar for library catalogies. We are exposing linked data for both of these and also creating a prototype visualisation – lots more on the blog. And see also linked data design issue resources.

We are linking to VIAF, to DBPedia, to the BBC etc. It’s important to the geotemporal side of stuff we are doing. We have linekd around name, location, people, subjects.

Archives Hub is already live: http://data.achiveshub.ac.uk/  – get your data here, it’s free CC0 data. There’s a SPRQL end point there as well as a more basic browser view. We are doing the same for Copac, it’s coming soon adn should be released about the beginning of September.

Visualisation prototype – several use cases – using tools like Simile, Many Eyes, Google Chart. So you can see timelines, maps etc.

We have a new project the Linking Lives Project – based around individuals much like BBC artist pages.

Key benefits of Linked Data? Has the potential to be a universal API in a way – you shouldn’t need to hand craft things though there is a challenge about matching things, particularly placenames etc. It’s quite a think to do this matching stuff. And sustainability wise the links to others work makes you vulnerable to losing your work. This guy caled Ed Summers put out the Library of Congress Subject Headings and it disappeared (though now resolved). Data modelling can be tricky especially can be complex. Licensing important but less of an issue these days. But linked data can make your repository work harder, find new channels in your data, and expose potentially hidden collections.

 

• Sheila Fraser (EDINA) – Using OpenURL Activity Data

I have a challenge of talking about middleware and data, what kind of analogy can I use. I started with pizza (I’m a fan) as it gave me an idea. When I’m hungry I go to the web, I put in what I want, they figure out how to get that pizza made and at my door and that works really well. The OpenURL router is just like pizza. You go to it, it finds the right copy and brings you the full text back. Like my pizza supplier the OpenURL router stores some data so that I can get what I want as quickly as possible.

The log of all this data is hugely valuable but looked like a bit of mess – we have been turning our data into something much more usable. And to use that, for instance, to create recommendations for future reading. We did some work with our logs to prove that the data could help us find citations and further reading.demos.openurl.ac.uk

What else can we find out? We can see that the busy periods of the year seem to be around the exam periods. Saturday is the quietest day of the week but then there is a weird peak of usage. We can have a good look through to find times when it may be appropriate to do system maintenance work for instance.

So

ODC PL – links to lots of others experiments

• Jodie Double (Univ. Leeds) – RePosit

Last year Sarah Malloy talked about the beginning of this project. See also our blog on the project: http://jiscreposit.blogspot.com.

We are looking at how advocacy can effect engagement with repositories through connections to repositories, There are 5 HEI partners and 1 commercial partners. We al have a research management system and separate implementations with different numbers of repositories connected to our CRIS. We had to talk to a lot of stakeholders – that takes time so we are a little behind where we hoped to be but we do have more of our objectives achieved. We have a Google Group set up that you are welcome to join if you want to discuss CRIS issues.

These are community resources that we want others to use – materials that you can get out there. If life different with a connector? Not really but we connect in more places with more people. As faculty team went out to talk about the system we did find that easier login helps. we’ve stolen tag lines to help sell our repository. Open Access is the same no matter what the connector – whether a CRIS or your own stand alone system. Have the numbers increased? Those we did advocacy with did get better numbers. But we don’t really have carrot or stick – mandates would be important here. We have a survey out. Now that we offer a £100 Amazon voucher on our survey we suddenly have 300 responses – bribary works.

Please comment and join in discussion about our repositories.

Q&A

Q1 – JAckie ? from the Repositories RP) You said numbers had not increased – was that of full text?

A1) Yes.

Q2 – Kevin Ashley, DCC) You were talking about excessive matches – the scale you were getting seemed reasonable to handle (not thousands). There are a fair few online services that use just the same level of match but they are still useful as humans can handle that short list of choices.

A2) Well four matches means a manual check. We’ve done some cheating here – we’ve taken that as a starting point. It still felt like a challenge. In some cases those matches may be more important/more difficult to sort between than others

Q3 – Peter Murray Rust) Glad to see that the author data is open. What percentage of the other data is open?

A3) It’s a small subset. We’d like the whole archive to be open. We have a stylesheet that we’d like to apply  but we think we should be able to make the whole thing CC0. Copac is the same actually. Initially RLUK were quite uncomfortable about us making their data open but they are fine with it now, it seems to have momentum now.

Q4) In the OpenURL Data how many years or months does it cover?

A4) It covers everything from the 1st April this year. We spent a lot of time looking at data privacy concerns. It’s hidden underneath, we needed permission for that data so we had that from the 1st April and that will be made available on an ongoing basis.

Q5 – Balvier Notay, JISC) Interesting use of SHERPA ROMEO in the symplectic system?

A5) It’s been an incredible journey actually. We get the added value that wasn’t in the system before. Email reminders to faculty saying ‘by the way you have a paper to deposit”.

A5) Richard, Symplectic: what’s been really interesting is that you can prove to repository managers what their potential deposits could be – realistics expectations of what could be collected in an ideal world.

• Marie-Therese Gramstadt (VADS) – ‘Kultivating’ artistic research deposit

Kultivate is a JISC funded project under the Reposit strand. It came from the Kultur II Group which is a group of researchers adn repostory managers working around the arts. Once of the key issues is what is artistic research? It is quite a young subject. Another issue is describing the artistic researcher – e.g. the anonymous Carrot Collective. One of the barriers to deposit is terminology. When I say repository to people say “suppository”. And what is the alternative? Maybe archive but there are connotations around control and authority that can be negative.

But the repository can enable the collection of ephemera around exhibitions – such as an invitation. Some of the problems about artistic research is that it is an ongoing proces, they want to be able to edit. A Royal College of Art case study has looked at simplifying the process. There are also administrative barriers at present – that can be overcome – such as needing separate logins for repositories.

EPrints projects container – a wrapper for project items. Important as artistic research is often complex with multiple objects and as part of a project type process. Some depositors refused to include their content until the repository was customised for their artistic research.

One of the tips we picked up was to have high profile champions to your repository. The Goldsmiths case study also suggested tailoring advocacy for each departments. Researchers are not clear what the repository will be used for. We know it is for the REF, but some wondered if it was for performance review. Making links to personal researcher websites is also important to getting researchers to understand that the repository is a wone stop shop. Also highlighting improvement in Googlability has been crucial – one researcher said that when talking in the US you get Googled so having material visible there is a great driver.

• Richard Jones (CottageLabs) – SWORD2

Ignore the slides – they are just for Ian!

I will talk about SWORD2 the process (swordapp.org documents the protocol). This has been an awesome community project. I ran about asking people for feedback and comments. We took that away and turned our white paper into a JISCpress page – a great way to get commentary on your content paragraph by paragraph. We had people commenting from all over the world and that was very useful and fantastic and helped us see where we were going with SWORD2. At that point JISC funded us for the rest of the year to turn that proposal into a real thing. The advisory group is open – do join us! That group included many of the people that had commented on the white paper. A great blend of senior and technical people and we had a savage discussion that led to a technical paper and a business case. And we tore that apart too. And then we shared an alpha version, made available on the website with version control.

We have had developers from all the main repositories, we have programmers, we have lots of people developing different implementations of the same standard. So we have a massive amount of technology to let SWORD be used in contexts we haven’t even thought about yet. The current version is still only a beta – it was launched just before OR11 and led to whole new requests and use cases for SWORD. JISC have now given further funding to investigate data deposit aspects. We also have some money for client developments we have a call out to develop our for SWORD in your system.

We’ve worked hard to make this come out of the grassroots needs of this community. All who wanted to be involved can be involved. We’d like you to continue contributing to that and giving us feedback. I haven’t told you what SWORD is, that’s not so important.

That’s the last of our proper Pecha Kucha’s – for voting on – but this is a wee presentation of a similar overall time.

• Extra talk: Charles Duncan (Intrallect) – Deposit from a mobile phone (PK with a difference!) *

This is about using SWORD as a standard to deposit via your mobile device – whether tablet or phone. So first an example with a video!

A pupil doing a school project on Mary Queen of Scots at the National Museum of Scotland. She’s taking pictures on her mobile and she can deposit her images on her phone. That could be images, sound, text etc. whatever you want to upload in this way. The whole thing is built into the system – so we take a picture outside the Intrallect office, share the image, and one of the ways on the list is IntraLibrary. You can see that you are asked for key information – Title, group, tags, etc. This is inormal metadata. Can then hit the magic Upload button. Now when you login to the repository – works on all IntraLibrary repositories – it has already grabbed the files name, size, etc. and it appears in your resource page – you can choose to automatically publish when you post stuff depending on your settings.

To get set up you need to have your username, password, and IntraLibrary URL (for SWORD deposit).

This is currently in beta. Only works on Android at present and supports IMS packaging – intended for learning object repositories. We are very interested in partnersing and projects and of course if anyone wants to test it.

Q&A

Q1) Was wondering whether Intrallect is specific just to one type of repository or whether you could use it for any others?

A1) Currently it supports IMS packaging but other than that it could work with any IMS and SWORD compatible repository

Comment) Would be a great app to go forward for the call for SWORD 2 App development

Q2) For the culture presentation – you mentioned that sometimes repositories are quite hard to find, they have weird names – can’t you tell them to call their repository something obvious?

A2) Maybe we need a rebranding exercise but really comes down to advocacy

Q3 – Peter Burnhill, EDINA) Shameless plug: have you considered “put it in the Depot?”!

A3) We did have a workshop on terminology and we did try various phrases on them.

Q4 – Peter Murray Rust) If SWORD2 is awesome! why is there SWORD 3 or will there be?!

A4) Actually the SWORD2  is the fourth SWORD project. But we think SWORD2 covers all we think we’ll need so shouldn’t do a SWORD3 as we shouldn’t need to.

Q5 – Balviar Notay, JISC) HIghlight for Kultur project – there is a huge engaged community around the Kultivate project here. There is more work being done in the research project in terms of research papers. The project has done an amazing job of pulling in the community and engaging them in preservation, in terms of advocacy.

And with that we are done with the main bit of Day 1! Thanks to all who have followed the live blogging today. We will be back tomorrow and there’ll be tweets coming out of the drinks reception that we’re all heading to up on the lovely informatics roof terrace just now!