The recent ResBaz (Research Bazar) conference at Melbourne Uni gave us the chance to demonstrate how figshare was being used by a number of publishers. We showed how figshare was being used for the publishing of supporting material and data for inline display (widgets) on a few papers, and how the papers were essentially being complemented by a figshare “collection”. This is a significant advancement from “e-prints” which were essentially online print versions, either PDFs or HTML text. The content can be dynamically displayed (for some file types), shared and even downloaded in some cases. Some interesting examples…
- http://doi.org/10.1021/acs.jmedchem.6b01594 (see supporting info)
- http://doi.org/10.1371/journal.pone.0171833 (this is the paper, next item is the figshare collection)
- https://figshare.com/collections/Resilience_through_adaptation/3691519
However, this supplementary content is still essentially a static publication, a fixed point in time. Now, you can use figshare to publish your own supplementary material and data. I thought I’d demonstrate how this might work for a researcher wanting to publish an article, but wanting to retain some control over presenting and linking materials and data.
I’ve created a figshare collection of material related to an older publication of mine. I obtained a DOI for the collection https://doi.org/10.4225/49/58b01d05a9b6f so it can be cited, then I published it. The collection reflects the publication, references it and links to it. (Ideally, I should have reserved the collection DOI before submitting the paper so I could reference it within the paper.) At first the collection included figures and diagrams, things that I might want to share or publish elsewhere in order to cross promote my work and the paper.
I then thought I would share some additional material after the paper was published. I added slides from the paper presentation as well as a digram I created to explain how the simulation experiment was constructured. Figshare required that I declare this a new version of the collection before these changes were visible. This is great for integrity, as everyone with the DOI gets the updated content, but you can see the version change and go back in time to previous versions. The ability to update published collections would be great for supporting future data formats, for better visualisations or presentations of the work, or even linking to relevant further work. But for the moment, I can include my power point presentation inline in this blog post! (Forgive the outdated presentation style… but it is what it was.)
Now, this took a while to setup, as metadata was needed for each data item, and each data item needed to be published first, then finally you create the collection and publish it and get a DOI. But using collections appears to give the most flexibility and control, and the setup time is definitely not worse than submitting a paper for publication. The figshare collection setup emulates how some publishers are using figshare. Linking from each data item to the collection needs to be setup by hand (if desired, and I used links in the description) and whole collections do not have certain functions (eg. download, widgets) but in a way this gives you options to restrict how your collections are accessed and viewed. Overall, I still think it’s a powerful tool for promoting and sharing your work.
Now, pretending that I then wrote a blog article to cross promote my work (which is sort of this article), then I’d probably include some references for people to use.
Paper reference:
Lyle J. Winton. 2005. “A simple virtual organisation model and practical implementation.” In Proceedings of the 2005 Australasian workshop on Grid computing and e-research – Volume 44 (ACSW Frontiers ’05), Vol. 44, p57-65. Australian Computer Society, Inc., Darlinghurst, Australia.
Cite the collection:
Winton, Lyle (2017): A simple virtual organisation model and practical implementation. University of Melbourne. https://doi.org/10.4225/49/58b01d05a9b6f