A look at Pubmed’s new commenting platform

Screen Shot 2013-11-04 at 10.00.50 AMPubmed is implementing a new function that enables researchers to share their thoughts about scientific publications. By allowing readers to comment and debate about specific papers publicly,  PubMed Commons is trying to extend the peer-review of manuscripts after their publication. If successful, PubMed Commons will become a platform for scientific discussions that could foster constructive criticism and eventually improve published papers and science. The service is currently in its pilot version which is restricted to recipients of NIH or Wellcome Trust funding. But I’ve put together a couple of screeenshots to give you a quick look at this interesting new feature.

Nothing much has changed from the familiar Pubmed abstract page. But notice that at the bottom of the page, a new “Reader Comments” section has appeared. Comments can be directly typed in the text box.

Screen Shot 2013-10-28 at 5.48.47 PM

Although comments of any length can be posted, it is recommended to keep it short and clear. Simple formatting options are offered, but need to be written as code in the text box.

Screen Shot 2013-10-28 at 5.51.20 PM

For example, links to other articles can be added by including the PMID number.  A direct link with author and year appears when published.

Screen Shot 2013-10-28 at 5.51.26 PM

All contributions are published under the Creative Commons Attribution 3.0 License and can be edited or deleted after publications. The comments are assigned a unique and permanent link so that they can be easily referred to. It is also not allowed to use anonymous pseudonyms to maximize accountability for comments. As of today about 200 articles have comments link to them, which is negligible compared to the 23 million referenced papers in Pubmed. But that will surely increase greatly with the service going live in the near future.

Similar initiatives such as Pubpeer and publishers such as Faculty of 1000, Frontiers and Plos already offer similar functionalities. However, the relatively small size of their user-communities and the lack of incentive for researchers to comment publicly on papers have limited their success up to now.  Pubmed Commons is starting off with a very large number of regular users, so perhaps this will be a different story. I believe that post-publication review will start being a serious alternative to the traditional post-publication peer review when at least two factors are combined:

  1. Commenting needs to happen on a centralized system, which is already familiar to researchers and that counts a certain critical mass of active users. Pubmed common has the potential to make that happen and Pubpeer’s talks about joining forces with Pubmed Commons could only help.
  2. Researchers need to accept that publications are not absolute truths, set in stone for eternity. Everyone should be allowed to make (honest) mistakes, if they are ready to accept constructive criticism. Publication should be dynamic documents, living with the evolution of knowledge. Papers are constantly being commented, criticized and applauded, but why keep these behind closed doors?

Comment on published manuscripts with PubPeer

Logo_PubPeerThe other day, I ran into PubPeer, which allows readers to comment on publications. Here’s a description directly taken from the “about” section:

PubPeer seeks to create an online community that uses the publication of scientific results as an opening for fruitful discussion. 

  • All comments are consolidated into a centralized and searchable online database. 
  •  Authors, as well as a small group of peers working on similar topics, are automatically notified when their article is commented on.
  • Pubpeer strives to maintain a high standard of commentary by inviting first and last authors of published articles to post comments.
  • The chief goal of this project is to provide the means for scientists to work together to improve research quality, as well as to create improved transparency that will enable the community to identify and bring attention to important scientific advancements. 

PubPeer is democratizing the peer review process. This is driven by the idea that publishing research results should be open to all since publishing costs are driven down by massive digitization. However open discussions and reviews should be retained to assure good science and generate new ideas.

Shifting the peer review process from before to after publication is an ongoing effort shared by others. The idea is usually to first build a community around a collection of papers then get discussion started.  I love to concept, but feel like the system is taking its time to get adopted by the masses. Why is that? Could it be because the communities are too small? Because they are too diverse maybe? Or perhaps because such comments are not taken into account to measure research impact?