My paper for Chercher le texte, the Electronic Literature Organization’s 2013 conference in Paris next month, is about ready and I’ve uploaded it to the ELO conference website and to Figshare.

Figshare was new to me. It’s an open access non-peer-reviewed research repository built to contain more than just papers: the idea is that you share your datasets, images, posters or whatever – and if you like a traditional paper as well – and Figshare sets you up with a nice website for your research materials all citation ready.

My paper is a network analysis of links to creative works of electronic literature from dissertations on electronic literature, so Figshare makes it easy to share the paper along with that data, allowing anyone else who is interested to import the same data into Gephi or some other software and do their own exploration of it. Excellent! It’s like GitHub for academics! Of course, some academics already use GitHub in fascinating ways, but I do like the idea of a site specifically for academic datasets, and I especially appreciate the citation information they provide. It also provides versioning, which is useful for this kind of data – I hope to get more PhD dissertations analysed before the conference, and when I do, I’ll update the dataset and the paper too – and Figshare will keep track of that for me.

I would like Figshare even better if there was some layer of peer-review here. You could leave uploads open to everyone, but with some kind of peer-review on top of that to give the best work a stamp of approval from your peers, you could make this sort of publication the only kind we would need. On the other hand, you do to some extent get this from the shares and views – although my own two shares are the only ones here, and at least 20-30 of my views, maybe more, are because I used this URL when I had participants in our workshop last week download the dataset as a test set to learn to use Gephi.

Another point is that Figshare was very obviously built for the natural sciences, and there are very few humanities papers in there. I’m not sure how much that matters. Less community for us humanists, certainly, but we can still use it for sharing.

I’ve never done research before the last year or so where I’ve thought of myself as using datasets – traditionally that’s not how we think about our work in the humanities. Obviously doing visualisations with data from the ELMCIP Electronic Literature Knowledge Base does generate datasets that others could reuse. What I do have from previous research is piles of screenshots of works and blogs that I’ve written about but not used in my publications. Most of that is sort of half-lost on forgotten hard drives. Perhaps if sites like Figshare become part of our workflow, even we humanists will start thinking about our notes and screenshots and other materials as datasets that could be shared?

Figshare also promises to help write your data management plan for grant applications, because they provide the services required by funding agencies – a system for keeping data safe. Not private or confidential, but archived, permanently. I’m not sure I’d trust a private company with that, to be honest. Figshare was started by a PhD student and is owned by Digital Science, a subsidiary of Macmillan Publishing that “creates software and tools rather than content”. Which sounds good, but how long will they be around? How independent are they? What is their plan for monetizing this? This is exactly the sort of system universities should be working together to provide, for example through GÉANT, the European network that among other things provides Eduroam, secure roaming wifi for students and employees at hundreds of universities worldwide. Actually, most universities’ open access research archives, such as my own university’s BORA, probably support uploading datasets along with papers. But I don’t think they suggest it – you can upload multiple files, but it’s certainly not the default expectation, and poking around BORA I couldn’t find any examples of research publications archived with their datasets.

There’s also a shift here to a culture of sharing research before it’s quite finished, and preprints before they’re in a journal. My paper here is less finished than it will be when it goes to publication, so it’s not really a preprint, just a draft of a conference presentation that I plan to do further work on. I’d certainly love any feedback on it at this point, though!