Open Research Online

This morning I gave a talk on ORO to the Health & Social Care Group of the Health & Social Care Department of the Health & Social Care Faculty (I think I’ve got that right!). Thanks ever so much to Sarah Earle for inviting me, and to Caroline Holdaway for organising the laptop, data projector etc.

Anyway, perhaps inevitably, even though I managed to talk and answer questions for almost an hour, there was quite an important aspect of ORO that I forgot to mention, and that is how you (either as individuals or groups) can link to, or get feeds from, ORO.

Thinking first about linking to ORO, I neglected to show you how nice your publications lists should now look. This is thanks mainly to our technical person (Miriam) who has painstakingly edited the code which underpins the reference style used for ORO. (Miriam has been very patient with me requesting italics, full stops, commas, en rules instead of dashes etc. – thank you Miriam!).

So, if you look at Sarah’s publications, for example, you will see that not only are they all nicely ordered chronologically, but also that each item conforms pretty well to Harvard-style referencing. The thinking behind this is that you can then consider adding a link from your staff homepage, comfortable in the thought your publications are all formatted in a standard way. Plus, each article will have a link to the full text (if supplied) and the definitive published version.

‘Feeds’ are ways of getting information out of ORO. For example, it would be possible to set up an RSS feed of the latest articles from the Health & Social Care Group added to ORO and have this embedded into an appropriate area of your department website. This kind of thing would need involvement from your IT department, but it is entirely do-able and certainly worth thinking about. If anyone is interested in this kind of thing, send me an email and we’ll try to set up a meeting between the appropriate techies!

This posting is aimed at OU researchers who deposit in ORO, but may well also be of interest to any repository managers or administrators from other universities that read this blog.

For those of you who regularly publish in Elsevier journals, you may well be interested to learn that they (Elsevier) produce an online version of your paper that can be downloaded and deposited in ORO without any worries about copyright. As many of you will know, Elsevier have an ‘Articles in Press’ service for most of their journals. This involves the early online publication of your paper at three stages of production prior to it being assigned to a volume and issue. These stages are: ‘Accepted Manuscript’, ‘Uncorrected Proof’, and ‘Corrected Proof’. It is the first of these three (the ‘Accepted Manuscript’) that you can deposit in ORO.

Here’s how to do it…

Upon acceptance of your paper, you should receive an email from the journal’s editorial office. In that email there should be a hyperlinked DOI number (Digital Object Identifier). By clicking on that hyperlink you should be taken through to the current online version of your paper on Science Direct (Elsevier’s online journals platform); so, that could be the ‘Accepted Manuscript’, the ‘Uncorrected Proof’, the ‘Corrected Proof’, or the final published version, depending on how soon after you received your email of acceptance you are doing this.

Assuming you click on the DOI pretty promptly after getting the email, you should be taken through to the ‘Accepted Manuscript’ version of your paper. You’ll be able to tell because it will be labelled as such, it will have an Elsevier-branded cover sheet detailing what version it is, and it won’t be formatted in to journal style. All that’s left to do is to save the PDF and include it in your ORO submission when you’re ready.

So why is this a good way to do things? Well, it saves you the bother of having to dig through all of your Word files (or LaTex if you’re a mathematician or a physicist!) looking for the accepted manuscript version of your paper to deposit in ORO. You also don’t have to worry about uploading any elements of your paper that might be in separate files (e.g. figures and tables). But most importantly of all, it means you are building ORO into your publishing routine; as soon as you get that email of acceptance, think ORO!

Papers that you deposit in ORO in this way will of course not yet have a volume, issue, or page numbers assigned to them. But don’t worry – we will initially deposit the item as ‘In Press’ and then we (not you) will update the bibliographic details when the paper is published.

Here’s an example of a paper already on ORO that is ‘In Press’ and has an Elsevier ‘Accepted Manuscript’ attached (of course, if you’re reading this in a few months time it may no longer be ‘In Press’):

http://oro.open.ac.uk/12755/

Finally, a reminder. Elsevier (and most other publishers for that matter) only allow the ‘Accepted Manuscript’ to be deposited in open access repositories like ORO. If it is an uncorrected or corrected proof (i.e. it has been copyedited and typeset in to journal style), or it is the final published version (the PDF the publisher sends to you upon publication) we cannot take it.

During our visits around the university departments it became clear that removing the ‘PDF-only’ restriction and allowing a number of different file formats to be uploaded to ORO (especially Word files) would be extremely useful and a time-saver for those who do not have access to PDF convertors. As a result we have recently improved the upload process so that researchers can now upload a wider range of file types.

The aim behind this recent move is to encourage our researchers to always upload the final accepted manuscript (whatever format it is in) and leave it to library staff to check the copyright of the full-text submitted, enter any embargo, and convert the paper to PDF.

There is a growing amount of evidence that making the full-text of a paper available on an open access repository like ORO can improve your chances of being cited. Therefore it is important that researchers continually add their full-text to ORO, especially since the forthcoming Research Excellence Framework (REF) will be based partly on bibliometrics. The message is simple – always send us your full-text and leave it to us to decide if it can be used.

We are very happy to add documents to existing records on ORO. Please direct these or any questions about the deposit process or copyright to the ORO Team.

There are lots more developments happening in the world of ORO – watch this space for more information!

Today I attended an interesting event on repository services, organised by the RSP (Repositories Support Project) and hosted by the University of Bath. The services featured were not only of interest to repository administrators and managers, but also of relevance to researchers. Specifically, I’m talking about content searching services; so, the ability to cross-search the content of multiple open access repositories.

We were given demonstrations of these services by the speakers; specifically, OpenDOAR, ROAR, BASE, DRIVER, and Intute. All of these are very good tools for researchers, enabling you to search across multiple platforms for open access content. I couldn’t help but wonder though, while I and many other repository staff in the audience are aware of these services and have played around with them to see what they can do, how many academics actually think of using repository search engines like these? If you’re a researcher and you can’t find what you’re looking for within the bounds of your university’s subscriptions and other information resources, or, dare I say it, Google, would you think of visiting a service that searches across open access repositories? Did you even know that such services exist?

According to the Registry of Open Access Repositories (ROAR), based on our number of records this is where we (ORO) sit in the ‘league table’ of UK institutional repositories:

1. Cambridge University
2. University of Southampton
3. LSE
4. UCL
5. Strathclyde
6. University of Bath
7. Lancaster
8. Bournemouth
9. Open University

Until quite recently we were fourth in this table… then fifth for a while… and now all of a sudden we’ve plumeted to ninth. On the face of it this looks like bad news and poor performance; but, as you’ve probably guessed by the title of this post, I’m about to explain that it isn’t, why it has happened, and why we may well drop even further.

The main problem with comparing repositories simply by the number of records they house is that we are not comparing like-for-like. For example, Cambridge – who sit at the top of the pile – not only produce a larger amount of research than the OU (and thus have more available to deposit in their repository in the first place), but they also accept a much broader range of material (e.g. datasets, multimedia files, images, learning objects, administrative material, and so on). Our current policy for ORO is to only accept peer-reviewed published research. This is clearly a limiting factor when thinking about and looking at repository size.

The point about policy extends beyond the type of material a repository houses to the way in which that material is collected. A decision was taken quite early on with ORO that a self-archiving approach would be used; that is, academics themselves deposit their work, or at the very least make a decison to instruct a member of administrative staff within their faculty or department to do it for them. Either way, they have knowledge of what’s being deposited and, more importantly, engage actively with ORO.

The alternative, non-self-archiving approach, is for library staff to mass-deposit items on behalf of staff in the background, often without their knowledge. For example, automatic imports from commercial databases such as Web of Science or Scopus could be used, or perhaps library staff could surf their university’s staff pages and manually add items from existing publications lists, CVs etc. Not relying on self-archiving by the individual user in this way inevitably boosts the size of your repository in a relatively short space of time, and this is another reason why ORO has been, and will probably continue to be, overtaken in the above league table.

In short, there is absolutely no point in comparing ourselves to institutions that have different policies and different models of deposit.

Why then, you may ask, do we stand by self-archiving? If we really want ORO to take off and be a true reflection of the OU’s research output why don’t we just get on with it and do it all from within the library and not bother our very busy academics in the first place? The broad answer is one of sustainability; slow and steady growth of ORO and the gradual establishment among OU staff of what it means in terms of access to their research, as well as what it can provide in return.  How can this happen without their active involvement?

In my opinion, however, the biggest disadvantage to not adapting a self-archiving appraoch relates to open access. Third-party depositing without the involvement of the author will almost always result in a metadata-only record; that is, there will be no full text attached to the item, and thus no open access to the research it reports. The third-party depositor could perhaps contact the author and ask for the full text, but this is time consuming and generally unsuccessful, especially if the author has had little to do with the repository in the past and therefore, frankly, doesn’t really care about it.

Of course all this depends on what you want from your repository. If all you want is a static publications database that can occassionally be used to output data for administrative purposes, then a third-party approach might be ok. However, if you – as we do with ORO here at the OU – really want to provide a service to academics by opening up their research to a wider audience through open access, then the depositor has to be actively involved.

Self -archiving is the best way to obtain full text, the best way to provide open access, and therefore the best way to showcase your university’s research. Once you have the involvement of researchers at the front end of the system, you can then begin to demonstrate what they can get in return, and hopefully what you end up with is a sustainable open access repository embeded into the culture of your institution.