Semantic Molecular Future: Article accesses during first 30 days

I believe that we seriously need a set of new metrics for scholarly publication. It should be multidimensional and one of these should be accesses. [Yes, I know it’s possible to game the system. Everything can be gamed. The only thing that can’t is scientific reality, but that’s increasingly irrelevant to academia]. So here is my contribution to access-metrics.

Our special issue of “Visions of a Semantic (molecular) Future” in J. Cheminformatics (BiomedCentral) has now been out for a month. Since BMC publish the article accesses for each article (a) confidentially to the author and (b) publicly for the highest accessed articles each month I can start to do some simple analysis. (I’m only missing one author, Cameron Neylon).

Here are the stats, after exactly one month (October 14th -> Nov 14th). The actual stats are for a window of 30 days.

2639 Accesses Openness as infrastructure
John Wilbanks Journal of Cheminformatics 2011, 3:36 (14 October 2011)

1185 Accesses Open Bibliography for Science, Technology, and Medicine Richard Jones, Mark MacGillivray, Peter Murray-Rust, Jim Pitman, Peter Sefton, Ben O’Steen, William Waites Journal of Cheminformatics 2011, 3:47 (14 October 2011)

1018 Accesses Semantic science and its communication – a personal view Peter Murray-Rust Journal of Cheminformatics 2011, 3:48 (14 October 2011)

936 Accesses Open Data, Open Source and Open Standards in chemistry: The Blue Obelisk five years on Noel M O’Boyle, Rajarshi Guha, Egon L Willighagen, Samuel E Adams, Jonathan Alvarsson, Jean-Claude Bradley, Igor V Filippov, Robert M Hanson, Marcus D Hanwell, Geoffrey R Hutchison, Craig A James, Nina Jeliazkova, Andrew SID Lang, Karol M Langner, David C Lonie, Daniel M Lowe, Jérôme Pansanel, Dmitry Pavlov, Ola Spjuth, Christoph Steinbeck, Adam L Tenderholt, Kevin J Theisen, Peter Murray-Rust Journal of Cheminformatics 2011, 3:37 (14 October 2011)

822 Accesses Ami – The chemist’s amanuensis Brian J Brooks, Adam L Thorn, Matthew Smith, Peter Matthews, Shaoming Chen, Ben O’Steen, Sam E Adams, Joe A Townsend, Peter Murray-Rust Journal of Cheminformatics 2011, 3:45 (14 October 2011)

681 Accesses The past, present and future of Scientific discourse Henry S Rzepa Journal of Cheminformatics 2011, 3:46 (14 October 2011)

531 Accesses OSCAR4: a flexible architecture for chemical text-mining David M Jessop, Sam E Adams, Egon L Willighagen, Lezan Hawizy, Peter Murray-Rust Journal of Cheminformatics 2011, 3:41 (14 October 2011)

429 Accesses CML: Evolution and design Peter Murray-Rust, Henry S Rzepa Journal of Cheminformatics 2011, 3:44 (14 October 2011)

420 Accesses Mining chemical information from open patents David M Jessop, Sam E Adams, Peter Murray-Rust Journal of Cheminformatics 2011, 3:40 (14 October 2011)

313 Accesses The semantics of Chemical Markup Language (CML): dictionaries and conventions Peter Murray-Rust, Joe A Townsend, Sam E Adams, Weerapong Phadungsukanan, Jens Thomas Journal of Cheminformatics 2011, 3:43 (14 October 2011)

280 Accesses The Quixote project: Collaborative and Open Quantum Chemistry data management in the Internet age Sam Adams, Pablo de Castro, Pablo Echenique, Jorge Estrada, Marcus D Hanwell, Peter Murray-Rust, Paul Sherwood, Jens Thomas, Joe Townsend Journal of Cheminformatics 2011, 3:38 (14 October 2011)

265 Accesses Adventures in public data Dan W Zaharevitz Journal of Cheminformatics 2011, 3:34 (14 October 2011)

^^^ cut-off of top 25 papers this month ^^^

264 Accesses CMLLite: a design philosophy for CML Joe A Townsend, Peter Murray-Rust Journal of Cheminformatics 2011, 3:39 (14 October 2011)

263 Accesses The semantic architecture of the World-Wide Molecular Matrix (WWMM)
Peter Murray-Rust, Sam E Adams, Jim Downing, Joe A Townsend, Yong Zhang Journal of Cheminformatics 2011, 3:42 (14 October 2011)

??? Accesses Three stories about the conduct of science: Past, future, and present Cameron Neylon Journal of Cheminformatics 2011, 3:35 (14 October 2011)

=== and a previously-published article ===

235 Accesses
ChemicalTagger: A tool for semantic text-mining in chemistry
Lezan Hawizy*, David M Jessop*, Nico Adams and Peter Murray-Rust Journal of Cheminformatics 2011, 3:17 doi:10.1186/1758-2946-3-17 (This has been out for several months and is noted as “Highly Accessed”)

Total 10063 for 15 papers in 1 initial month == ca 20 accesses per day per paper

So what conclusions?

  • Clearly the figures aren’t random, or created solely by bots (I joked about the “wilbots”, but I am sure these are real people reading John’s article. There were about 50 tweets mentioning this article so people wanted other people to know about it.
  • The top articles are not significantly about chemistry. So it doesn’t matter where material is published. This is a really important message. If you have something to say, then people will find it.
  • A LOT of people read Open Access material. Note that after 14 days the articles were also available on Pubmed so we probably double the figures.
  • I can’t believe that these papers would have been nearly so widely read in the – effective Closed competitor – J. Chem. Inf. And Modeling. We have also had high readership there – OPSIN paper was “highly accessed” but I have no figures. But I suspect that only a small fraction of the current readership would have access. So closed access publication constrains innovation and constrains multidisciplinarity.


Egon writes in a comment:

Peter, why not give Total Impact a try… click the ‘Manually edit this collection’ link in the left menu under ‘Collect IDs from:’ and add the DOIs for those papers, and hit the red ‘Get Metrics’ button…

PMR: I don’t get anything back that makes sense… Nor do I know how to add PMIDs instead of DOIs


This entry was posted in Uncategorized. Bookmark the permalink.

10 Responses to Semantic Molecular Future: Article accesses during first 30 days

  1. Peter, why not give Total Impact a try… click the ‘Manually edit this collection’ link in the left menu under ‘Collect IDs from:’ and add the DOIs for those papers, and hit the red ‘Get Metrics’ button…

  2. I also recommend Total Impact. An alternative is my ScienceCard ( project, but here you put in people and not papers. ScienceCard currently only knows about PLoS PDF downloads and HTML views, but I hope to include more journals in the future.

  3. FWIW Just checked the statistics on my article: 260 total downloads and 196 in the last 30 days. Not entirely sure what that means but another piece of data to add.

    • pm286 says:

      It means that a lot of people thought it was worth looking at your article. More than if it had been in the closed equivalent. Since we’ve only had 2 days not in the 30-day window it suggests about 60 read it in the first two days and you are now running at ca 200/month. Not bad

  4. Just checked up with the Total-Impact site as well. There is some bug with adding of DOIs. I’ve started up a collected but so far all of the metrics are zero. Not suprising for citations as there won’t be too many citations yet (actually Google Scholar picks some up but they don’t have an API so aren’t included in TI) and BMC don’t provide a public download statistics API. Less sure why it didn’t pick up tweets or bookmarks. Tweets possibly because its hard to track unless people tweet the DOI, which they don’t mostly. Euan’s is probably capturing that aspect of things better.

    • Actually, no something is definitely wrong. Not picking up any biblio metadata from Crossref either so either the dois aren’t yet live which would stop things working or something else has gone wrong…

    • pm286 says:

      Egon got one tweet for DanZ but I got zero.
      I think the accesses are very convincing – does TI index BMC? because BMC does not give accesses publicly apart from the top 25. Actually I trhink BMC is missing out there as early access metrics, done daily would be a great encouragement to read. And TI-like tools could spray those out to people.

      • Ok. Doing slightly better now. Although now some of them seem to have been picked up as datasets which is rather odd. Nonetheless I think this gives a reasonable picture of the overall state of things. Yes, its a shame that BMC don’t make the download data public. Anyway, this picks up the 50 odd tweets for John’s paper and a few for mine as well as blog mentions for a couple of the others.

Leave a Reply

Your email address will not be published. Required fields are marked *