Feb 12, 2008
Search and Cite for Science Bloggers
A couple of weeks ago I was reading a post by Will Griffiths on the ChemSpider Open Chemistry Web blog about how the DOI citation system of journal article lookups might be improved. The DOI system basically assigns each research paper a unique number depending, with an embedded publisher tag. Enter a DOI into a look up box (e.g. the DOI lookup on Sciencebase, foot of that page) and it almost instantaneously takes you to the paper in question. I use the DOI system for references in Sciencebase posts all the time.
There are a few cons that counteract its various pros, for instance, not all publishers use it and among those that do there are some who do not implement the DOI for their papers until they are in print, as opposed to online. Despite that it is very useful and commonly used. Having read Griffiths’ post about OpenURL a non-proprietary DOI alternative, I thought maybe it would be even more powerful if the concept were taken back another step the author level and I came up with the concept of a PaperID, which I blogged about on Chemspy.com. PaperID, I reasoned could be a unique identification tag for a reasearch paper, created by the author using a central open system (akin to the InChI code for labelling individual compounds). I’m still working out the ins and outs of this concept and while a few correspondents have spotted potentially fatal flaws others see it as a possible way forward.
Meanwhile, CrossRef, the association behind the publisher linking network, has just announced a beta version of a plugin for bloggers that can look up and insert DOI-enabled citations in a blog post. I’ve not investigated the plugin in detail yet, but you can download from a Crossref page at Sourceforge.net. the Crossref plugin apparently allows bloggers to add a widget to search CrossRef metadata using citations or partial citations. The results of the search, with multiple hits, are displayed and you then either click on a hit to go to the DOI, or click on an icon next to the hit to insert the citation into their blog entry. I presume they’re using plugin and widget in the accepted WordPress glossary sense of those words as the plugin is available only for WordPress users at the moment with a MoveableType port coming soon.
According to Geoffrey Bilder, CrossRef’s Director of Strategic Initiatives, “CrossRef is helping jumpstart the process of citing the formal literature from blogs. While there is a growing trend in scientific and academic blogging toward referring to formally published literature, until now there were few guidelines and very few tools to make that process easy.” Well that reference to a jumpstart sounds like marketing-speak to me, as Sciencebase and
dozens
of other
science blogs have
been using
DOI for yearsSciencebase and dozens of other science blogs have been using DOI for years.
Whether or not I will get around to installing what amounts to yet another WordPress plugin I haven’t decided. I may give it a go, but if it works as well as is promised you will hopefully not see the join. Meanwhile, let me have your thoughts on the usefulness of DOI and the potential of OpenURL and PaperID in the usual place below.
Unfortunately from experience, users don’t like having to recreate a paper database from scratch.
I’m speaking really early into the game on this one, but I’m recoding the chemrank site I made last year to only use Connotea’s database. Since Connotea allows users to submit hard URLs or DOIs, piggybacking on Connotea’s own indexing system may be viable.
Mitch
@JCB I think there is a lot of potential in this little idea of PaperID, although as I’ve said elsewhere I’ve not quite worked out the ins and outs of how to get it started as a viable and realistic supplant to DOI/OpenURL that would be controlled at the author, as opposed to publisher, end of the publication process
db
baoilleach, I just installed and ran the plugin, so I had a better idea of what it can do. I tried a Science paper I wrote about a few weeks ago, and it found two ScienceDirect entries, so it doesn’t bode well…nor could it find the PNAS paper I’m writing about for spectroscopynow.com given author name…
Another annoyance is where it slots into the WordPress GUI sidebar, I’d prefer it above the fold while I’m writing a post.
db
Unfortunately, it’s only as good as the data the journal publishers have deposited with CrossRef. It seems that even CrossRef don’t have the author list in some cases. See comments by Geoffrey and Chuck of CrossRef on my blog post.
David,
We’re always willing to experiment new ways of packaging and indexing experimental info. In terms of DOI and author generation, that is pretty much what I see Nature Precedings doing and we’ve used it a few times to do that. I assume a DOI refers to a specific version of a document (although Precedings does allow uploading of multiple versions I am not sure if you get a new DOI every time). For experiment pages, we still don’t have a “final version” for most of them – I am working with students in the next few weeks to get to that point. Of course if we discover errors down the road we will correct them and the wiki version tracking system will keep a record of that. We would like to get the associated raw data off our server as well though to get third-party time stamps. Maybe ChemSpider can help with some of that down the road (reaction monitoring spectra for example).