Friday, August 21, 2015

Internet-aided serendipity in science (was: How the Internet can help chemists with serendipity)

The ACS Central Science RSS feed in Feedly.
Finding new or useful knowledge to solve your scientific problem, question, etc, is key to research. It also is what struck me as a university student as so badly organized (mid-nineties). In fact, technologically there was no issue, so why are scientists not using these technologies then?? This question is still relevant, and readers of this blog know this is a toy research area to me, and I have previously experimented with a lot of technologies to see how they can support research, and, well, basically, serendipity. Hence, internet-aided serendipity.

This happened to be the topic of an article by Prof. Bertozzi (@CarolynBertozzi), editor-in-chief of the gold Open Access ACS Central ScienceHow the Internet can help chemists with serendipity, part of the website. I left a comment, which is currently awaiting moderation, but to keep the discussion on twitter going, here is what I left (the comment on the article may turn out to have lost the formatting still present here):
    Dear Prof Bertozzi,

    the browsing of TOCs is not a lost art, and neither has the Internet solved everything. Where I fully agree that Twitter and other social media have filled a niche in finding interesting literature, it is basically kind of a majority vote and does not really find you the papers interesting to your research. This has to extend, of course, to #altmetrics, which capture the attention on social media and allows creating TOCs on the fly, as do (good) paper bookmarking services like CiteULike (see Similarly, people developed tools to find science in blog posts, like the no longer existing, continued/forked as Chemical blogspace (see, but consider this code has not been updated in the past 2-3 years). So, creating cross-journal TOCs is a daily habit for many of us still. (BTW, will ACS Central Science fully adopt #altmetrics, as data provider as well as showing #altmetrics on the website?)

    Returning to the single journal TOCs. Here, RSS feeds have shown to be critical, happy to find a RSS feed for ACS Central Science ( It is good to see that the journal's RSS feed for the ASAP papers contains for each paper the title, authors, the TOC image, and the DOI (possibly, it could also include the abstract and ORCIDs of the authors). Better, it should adopt CMLRSS and include InChIs, MDL molfiles, or SMILES of the chemical compounds discussed in that paper (see this ACS JCIM paper: With proper adoption of CMLRSS, chemists could define substructures and be alerted when papers would be published containing chemicals with that substructure (and it does not have to stop there, as cheminformatically it is trivial to extend this to chemical reactions, or any other chemistry). After all, we don't want to miss the chemistry that sparks our inspiration!

    I personally keep track of a number of journals via RSS feeds which I aggregate in Feedly, which filled the gap after GoogleReader was closed down. Feedly does not support CMLRSS (unfortunately, but I have other tools for that) and there are a few alternatives.

    So, I hope the ACS Central Central journal will pick up your challenge and continue to support modern (well, CMLRSS was published in 2004) technologies to support your past workflows! For example, make the link to the ACS Central Science RSS feed more prominent, and write an editorial about how to use it with, for example, Feedly.

    Maastricht University
    The Netherlands
Of course, there is a lot more. It should not surprise you that adoption of PDF and ReadCube as killing internet-aided serendipity, where HTML+RDF, microformats,, etc would in fact enable serendipity. Chemistry publishers do not particularly have a track record in enabling the kind of serendipity Prof. Bertozzi is looking for. Good thing is that as editor-in-chief of an ACS journal, she can restore this serendipity and I kindly invite her to the Blue Obelisk community to discuss how all the technologies that have been developed in the past 15 years can help chemists. Because we have plenty of ideas. (And where is that website again aggregating chemistry journal RSS feeds...?)

Or, just browse this posts in blog, where I have frequently written about the innovation with publishers (in general; some do better than others).

Update: Other perspectives