Snowflake Science

Last Friday, I gave a talk on the snowflake effectfor science in my own department – funny how I give presentations all the time but almost never “at home”… This was a follow-up to an earlier blog post on how I need help…

The slides are below, and so is the audio recording. This is a bit of an experiment, as I’ve tried to synch the sound and with my slides on slideshare. Do let me know if that works for you. It does seem to work fine for me, though I am not sure whether it is possible to download the slides and sound on an ipod – something I depend on for keeping me up to date. (In a much more convenient and easy way than I have access to my science reading – a theme I elaborate a bit on in the talk…)

| View | Upload your own

(By the way, I didn’t edit out the fire alarm around slide 5… In case you’re wondering, this wasn’t The Real Thing, though it did show that the alarm is kind of useless as everybody stayed in the room 😉 … The annoying background sound of the alarm goes away after a few minutes.)

At slide 42, there is some discussion on the first part of the talk. One of the questions challenged that people really want to know who cites them. The basic argument is that you should know if an important person cites you anyway, or you are not a Serious Researcher. I wonder about that… It may be true that good researchers are better at dealing with citations, but that doesn’t seem to be a reason to make it harder to find out about citations than it has to be? This theme comes up in other contexts too: a serious music lover will know how to discover new music? A serious learner or teacher will know where the good quality resources are? Again, that doesn’t seem very convincing to me: maybe only those that are good at discovering or finding stuff can be good at their jobs now, but if we can make the problem of finding stuff go away, then maybe that skill is no longer all that relevant?

Another remark from the audience was that it is very difficult to do really good extraction of bibliographical data. I would agree that it is really difficult to do this in a perfect way. But this seems to be another case of “the perfect is the enemy of the good” to me: if you would be informed of 10 new citations of your work, and, on close inspection, it would turn out that 2 or 3 of those were not really relevant, I think that many of us would still find such a message quite useful and valuable.

The second part of the talk is more about how different the coverage of google scholar, web of science and citeseer is. As slide 51 illustrates, web of science and citeseer cover 35% and 38% respectively of scholar, for the faculty in my department! This means that, if my university switches the way it measures academic output, we could magically just triple our productivity 🙂

Moreover, for some people citeseer covers most of what they do, for some web of science does, and for some, me for instance, neither citeseer nor web of science covers even 5% of what they do! Actually, I think that this says something about the state of our domain and the fragmented nature of research on Technology Enhanced Learning! This is something I want to explore further…

As when I talked about this subject before, it seems to touch a bit of a “raw nerve”… Apparently, the talked encouraged Phil to take up blogging again :-), and I actually did get some help! In follow-up messages, I received suggestions including sciencedirect, scitation, eventseer and rexa. Seems like there may be more innovation in this space than I suspected… Please do keep sending me your help. I’ll report back!