Charles Basenga Kiyanda

It is time

If I’m going to be serious about this new way of disseminating science, it has to start now! I just saw on slashdot that a new science site, which they titled as “YouTube for Science” had started. It’s called SciVee. This is not exactly what I was thinking of (so I haven’t been scooped yet, pfiou!), but it has elements of what I’d like to see (so it’s dangerously close, bummer!). If I understand well, then SciVee allows you to upload a video (or just audio) of yourself talking about your paper and you can synchronize it with plots from your paper. This isn’t what I really had in mind, but it’s getting close. Their system has some limitations.

For starters, it’s looking at papers that have gone through the classical peer-review process. I’m aiming at something that, ultimately, can bypass the classical peer-review process. Conceptually, it would be easy to extend the SciVee service to include papers which haven’t gone through the peer-review process. You’re already uploading a video, it’s not much different to upload a paper. What’s harder then, is how do you determine what’s a good paper from a bad paper then? Their system is probably not really set-up for that. Plus they don’t have a review process and the standard “click here to make this paper (or SciVee as they call it) a favorite” is not good enough for scientific publications.

In the same vein as the previous comment, their uploading and reviewing process is linear. I believe there is a way to rate and comment on the videos of other users (from the explanation, it’s unclear whether you’re rating the video or the science, so would you rate low a bad video of bad science?), but it’s what I’d call a direct user input. You look at a video and you rate how much you think it’s worth. Let’s say, for the sake of the example, that it’s a 5 point scale. You would rate something 3/5 or 4/5. Any such system has several drawbacks. One is that you’re not always sure that users understand or at least share the same rating scale. 4/5 for me might be an outstanding paper with 5/5 being nobel prize material. 4/5 for you may be an average paper with 5/5 being anything between a really good paper and Nobel prize material. This makes such a scale less relevant. What I’d propose is to have a system whereby users can communicate with each other about the different papers available (or the SciVees as they call them). Papers are then rated based on how people interact with each other about those papers. If a paper gets lots of positive recommendations between users it gets rated higher. There also seems to be a system to comment on a particular SciVee. I don’t like comments as a main way of determining the value of a piece of work. I really don’t want to have to sift through 12 billion comments before I can decide whether the work was well done.

So far their system isn’t a community per say. Like I said, it’s linear. You upload, people download. Sometimes they comment and review. In their defense, they have a community aspect on the way. You can sign-up as a beta tester.

I should point out that I’m saying all of this as someone who hasn’t used the site as an author. One of the big drawbacks of the site right now is that you can only use it with papers in open access journals. The journals of PLoS are probably much much easier to use with this system. I don’t have papers in open access journals and more so not in PLoS journals. They do say that they’ll, at some point, accept getting abstracts from articles published in traditional journals (not open-access). So the expansion of their system relies on people publishing in open-access journals and with licensing options that permit re-using the material.

So like I said, it’s a good idea. It’s not completely MY good idea, but it’s a good idea. So whatever it is I want to do, I have to do it NOW! This definitely has the potential, with some added functions and maybe a complimentary service from someone else, to fill the niche I’m looking at. I still think by-passing the traditional peer-review system is the way to go. Not to abolish peer-review, but to change it drastically. The greatest hurdle of such a new service will be community acceptation. The traditional peer-review system has been used for a long time as a way to validate the worth of individual scientists. The community is probably not going to let go very quickly, so giving the right incentives to use the new system is going to be important.

1 comment to It is time

  • Tommy

    Nice post. I like your idea of a rating system whereby users interact with each other, thus producing a more accurate rating, but I wonder how you would be able to achieve this, as it may be way harder to code than a simple rating system (a la youtube). Then you would have to specify the criterias on which the ratings are achieved so the process is transparent enough. But hey, I’m sure it can be done, with a lot of determination 🙂

    Regarding the peer review system, I totally share your views. I read so many horror stories abusing of this system that it’s not even funny. For starters…

    More info regarding the weakness of the peer review system here :

    Keep up the good work !! You have gotten quite an interesting blog here !

Leave a Reply




You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>