Repository records now in City’s new Library Search

Over the last few months I’ve been involved in implementing Summon here at City, which we have called Library Search. One of the things we wanted to make sure we included in Library Search’s content set were repository records, and they have now been uploaded to the service’s index. You can view all repository records in Library Search, and the first hit for this search is an example of a City Research Online full text research paper coming up in a real search.

It will be very interesting to see if our download statistics improve as a result of this, and I expect that they will- presumably City users of the service will be interested in City academics’ research! You can read more about the development of this service over at our Summon at City University London blog.

Advertisement

Filed under: City Research Online, discovery, Library Search, Summon, system architecture

Using City Research Online to serve papers to RePEc

One of the promises of the creation of a network of institutional repositories was that this would truly be a network, in the sense that there would be facility for appropriate transfer of material between services (I wrote about this for UKCoRR’s blog a while ago if you want more context). For example, an academic should be able to post a paper in the home repository, and also see this transferred automatically to e.g. the ArXiv.

We saw an opportunity to do this here at City when we began archiving our Department of Economics Discussion Papers Series. It soon emerged that the main point of discovery for economists looking for papers was the Repository of Papers in Economics (RePEc). The person in charge of the series had set up a page on the Economics website that pushed the papers in the series to RePEc, but this required an awful lot of maintenance, in particular ensuring that data could be transferred to RePEc in an appropriate format as RDF files.

So, we offered to take care of ensuring the series was automatically transferred from City Research Online (CRO) to RePEc. This involved some work with Eprints services and the people at RePEc to set up an area at CRO which indexed the papers as RDF files using the eprints2redif script. This is then used to push these files to City’s Department of Economics page at RePEc. The CRO RDF file-set updates overnight, meaning that additions, deletions and changes to the files therein will quickly be reflected on our RePEc page.

This will hopefully be a convenient and useful service for our economists- add your discussion paper to CRO, and it will automatically appear in RePEc! For us it’s a real win as well- we can take the administrative and technical burden off the economists’ hands, and also demonstrate that we are able to offer this kind of service to other departments. Also, it means that we should see a significant improvement in our download statistics, since the papers’ records in RePEc actually point back to full text papers in CRO when people hit the download button (see the URL to download this paper, for example). So it really is a win-win situation!

I would encourage other repository managers to have a think about this. I found the Department of Economics to be very receptive when we approached them, particularly when it became clear that we take on work they were spending time upon. There is some technical work that has to be done, but nothing that should flummox an experienced Eprints administrator. The next thing I’m going to think about is whether we can arrange something similar for our Centre for Mathematical Science, who are keen users of CRO and the aforementioned ArXiv.

Filed under: City Research Online, Systems, discovery, RePEc, repositories, scholarly communication, system architecture, web integration

City Research Online search functionality

We’ve just launched search functionality for publications data held in City Research Online. We’ve created a dedicated search page on the Research area of City’s website, as well as a supporting page with information about the service in general. This is the first time we’ve surfaced data from Symplectic (our Current Research Information System) to the web, and it took us a while to sort it out as well as some dedicated web development time, but we’re pleased with the results.

The search has been created by using Symplectic’s Application Programming Interface (API). The API pushes out “approved” (i.e. items validated by their author(s)) publications (in the form of citations plus abstracts) to a cache. The cache is then indexed and ranked by Funnelback, City’s corporate website’s indexing tool. The indexed data is then exposed to a keyword search via the form at the page linked to above.

There are a few features of the search, and the results it creates, worth flagging:

• We have top-ranked search results where there are full text open access papers associated with those results. See, for example, a search for Jason Dykes’ publications– you’ll note that the first 40 or so hits allow you to click through to an openly accessible paper. This was done on the rationale that people are more likely to be interested in results with papers associated (and it doesn’t hurt or download statistics!)
• As mentioned above, the search’s index includes abstracts, where present in the publication’s metadata. This means that search terms can sometimes appear a little fuzzy, particularly when you get towards the bottom of a list of hits- see for example this page, which is the fourth page of four when searching for the term “concrete”. We’re not too worried about this, given the propensity of searchers to only look at the first couple of pages of hits for any given search.
• The advanced search is not particularly advanced. Our web developer is going to include a date range for results, but generally we weren’t looking to re-create a City version of e.g. Scopus, so we felt that relatively few advanced search options would be adequate.
• We still need to do a bit of re-jigging of the formatting of the main search page, for example to include some text fields after the search form, to make the layout look a bit nicer. We’ve also included the service’s Twitter stream and an RSS feed of new items on this page, to give an idea of full text content being made live.

As ever, any feedback on any aspect of this new functionality much appreciated- you can email the team at publications@city.ac.uk.

Filed under: City Research Online, Systems, discovery, search, system architecture, web integration

Business as usual

I became conscious we’ve not updated this blog in a while. This is because it has been business as usual, getting hold of and adding content to City Research Online. We’re now up to over 300 papers in the open access repository- not bad, considering in August we had nothing!

A few pieces of work we have in hand, or will be looking at in the New Year are worth mentioning:

• Further integrating data from Symplectic and Eprints into City’s web presence, specifically the Research area of City’s website. We’re currently working on a publications search, allowing users to query our holdings in both Symplectic and Eprints.
• In the New Year, we’ll start looking at using City Research Online to store and serve City’s PhD theses. This will be an interesting piece of work- the infrastructure side should be relatively straightforward, but the policy side of things (including intellectual property issues) will be more challenging.
• We’re going to be examining whether Gold OA is something that we need to be more actively supporting at City, particularly in light of recent government pronouncements on this issue.
• I would also like to do some scoping of the extent to which City is producing working paper series (I know of at least a couple), and if and how we might archive these series, or even support their publication.

Seasons greetings to our readers, see you in the New Year!

Filed under: City Research Online, Gold open access, Open access, search, Symplectic, system architecture

OpenURL and City Research Online

We’ve been grappling with OpenURL here at City, partly as a result of work being done at LSE (see comments in addition to the main post which give useful context), where I used to manage the repository. A little healthy rivalry never hurt anyone!

We wanted to make sure that repository users are able to reach the published versions of research as quickly as is possible, even though we’re currently associating full text with (nearly- see below) every record. To this end, we include DOIs in every record that has one, meaning that people can click through to the published article.

Where an OpenURL resolver comes in handy is by extending the concept of discovery-by-DOI above to query whether your home institution holds the article in question, then if it does to send you directly to the published version of that article. We’ve now configured this so it works for any UK institution. In other words, you can run an OpenURL query via an ePrint record, and it will check against any OpenURL-enabled holdings at UKHE institutions.

By way of example, if you go to this record, hit the tools button at the top of the record, and hit the “Find a copy” button, you will be able to find out if your institution has access to the journal Language Learning and Development for the year in question (in this case 2010). If it does, you will be able to click through and (after logging in) see the published version of this paper.

Why, you might ask, does this matter if all of City Research Online’s records have full text associated with them? There are a few reasons why it might be a useful feature, I think. First, you might wish to compare the open access version of a paper with the published version. Second, you may come across a record which is still under a publisher’s embargo period and is hence not available from the repository, but you still wish to access it. Third, it may be the case in future that we open up the repository to non-full text citations, in which case discovery of items away from the repository (via DOI or OpenURL) becomes much more important.

Finally, a note on getting repository material into our OpenURL holdings. If we were to do this, it would mean that we would enable click-throughs from citations databases (e.g. Google Scholar, Scopus, Web of Science etc.) using “Find this article” (or similar) buttons, making material in the repository more discoverable. To do this, we need to track repository holdings, something that our OpenURL tracker can’t currently do. However we have one or two leads on this, so watch this space…

Filed under: Systems, discovery, OpenURL, system architecture

Progress!

A few excellent pieces of progress on the City Research Online project to report.

First, we now have academic colleagues interacting with the Current Research Information System to upload their papers, and have already received some material via these channels. This is in addition to material sent to us via email to the shared service inbox. We’re now working to make research live, as can be seen in the latest additions area of City Research Online.

Second, we have set up a Twitter account which streams new additions to the service. Tweets from this account have already been noticed and retweeted by City colleagues and by City’s main Twitter account and (of course!) the Library’s account. On top of this, repository colleagues have added the City Research Online Twitter stream to this aggregated list of UK repository tweets.

Finally, we’re making the first steps toward integrating repository data with City’s web presence elsewhere, by adding an RSS feed of new research from the Department of Journalism to the Library’s subject guide for that department. We’re going to roll out more of these feeds to other library subject guides and other areas of the library’s web presence soon.

Though incrementally small, these feel like real progress toward telling people about the service, and repurposing data from it in useful ways.

Filed under: City Research Online, RSS, system architecture, Twitter, web integration

Making systems talk

We’re currently in the process of asking our Symplectic Elements current research information system (CRIS) talk to ePrints, our repository software. The idea is that academic colleagues should be able to upload full text versions of research directly to the CRIS, at which point the item gets cross-walked into ePrints automatically. This gives a one-stop-shop for academics wanting to manage their research profile and open access material, and makes administering the workflow seamless for those of us behind the scenes. Fantastic: a seamless service for academics and for administrators, and a sound basis on which to set up a  new repository.

The problem seems to be actually getting this to happen. It has proved difficult to get the necessary code from Symplectic to pass to ePrints to make the cross-walk happen; then there were issues with opening up the server ports, as City’s network security people were (quite justifiably) interested in what systems were being given network access and why; and now we think it’s possible that when Symplectic uploads records to ePrints, it may overwrite material already there (fine if these are just test records as currently, not so good if genuine records we create in ePrints down the line get over-written).

None of this is to point the finger of blame, and I’ve actually not had to manage any of this directly (my colleague Andrew has had that pleasure). The point seems to me to be that systems architecture is difficult, and particularly so when you’re trying to manage a  bi- or tripartite relationship (in this case Symplectic, City, and ePrints). If I were to draw a lesson from all this it would be, in a project setting, to always factor in plenty of time to allow for delays in making these relationships work, because things never run as smoothly as one would like.

Filed under: Systems, ePrints, projects, Symplectic, system architecture