Friday, July 21, 2006

A time-saving step

This weekend I'll catch up w/ the cond-mat archive. In the meantime, I wanted to point out one amusing piece of Lubos Motl's latest blog posting:
The previous paragraph also clarifies my style of reading these papers. The abstract has so far been always enough to see that these fundamental gerbes papers make no quantitative comparison with the known physics - i.e. physics of string theory - and for me, it is enough to be 99.99% certain (I apologize for this Bayesian number whose precise value has no physical meaning) that the paper won't contain new interesting physics insights.
This attitude is surprisingly common among physicists. In a graduate seminar course at Stanford, someone else in the class showed our (then pre-)Nobel Laureate theorist professor a paper on high temperature superconductivity. After glancing at the title, author list, and abstract, he tossed the paper face-down on the table, and said, "I don't even have to read this to know that this is crap." Sometimes this approach (or its converse) really does work. I certainly have a list of condensed matter and nano experimentalists whose work I presume to be extremely good, because everything I've ever seen from their research groups has been elegant and solid. However, pre-judging results based on who did the work and what the abstract says is exactly the kind of non-scientific, unobjective attitude that emboldens social science types to argue that science and its findings are largely a social construct, etc., a conclusion that I think is way off base (when I drop my pencil from above my desk, it will fall toward the ground at 9.8 m/s^2, regardless of my sociology, preconceptions, or personal beliefs).

5 comments:

Dave Cobden said...

Hi Doug,

Of course the problem is that we are all caught between a rock and a hard spot nowadays. There is simply no time to study all the papers we know we should. Lubos is unusually frank about his strategy for dealing with this, but we all have to make painful compromises and play to our instincts to some extent or we can't survive. Like you I have an 'A-list' of experimentalists (and theorists) whose work I know must be read. The difficult thing is finding the time to give a fair try to all the others. By the way, there is not much common or typical about Bob Laughlin, I hope you realize!

Congratulations for keeping this blog, a lone condensed matter voice in the pervasive noisy HEP/cosmo background. A Martian trying to work out what 'physics' meant from blogs might never realize that here on earth the majority of us are actually interested in things that can easily be measured in a lab. I promise to try harder to find time to read your posts (sadly at the inevitable expense of some no-doubt worthy papers listed in the Virtual Journal.) IMO there are lots of interesting and controversial questions in CM that could be entertainingly discussed in a blog thread.

Dave

Incoherent Ponderer said...

I read the title, abstract, the author list, maybe the conclusions, before I decide whether the paper is something of interest. Deciding something is "crap" based on abstract and author list alone is a dangerous practice. However, I suspect in your story the nobel prize winner has heard similar arguments from the same group of people before and therefore didn't have to read it to know what was their point.

In general I don't think it's wise to decide that certain authors only produce "crap" (a lot of times work done by postdocs and grad students who have short life-time crossing through the group). Even when reading papers that strike me as questionable, I often give the authors benefit of the doubt. I know others who don't - my colleague describes approximately 90% of talks or papers as being "complete BS". His paranoia is really entertaining to watch, especially since he has spoken out against some pretty established facts as well (such as band theory calculations for some elemental metals). He is a funny guy! :)

BhpG said...

Doug--

Well, I barely dip my toe in the stream of current research, much less drink from the firehose of the preprint whirlwind. But I don't think I agree that judging a paper on its title and abstract is wildly off-base. If the author can't "hook" you with a well-written abstract, then -- while it's possible this is a diamond in the rough -- it's more likely that things have gone awry.

I also understand the reluctance to categorize based on the author's name. In general that's a noble sentiment. On the other hand, I've never gone wrong picking up a book with, say, Neal Stephenson on the cover -- while I don't have to delve too deeply to know that I'm not going to enjoy the latest Danielle Steele opus. It's possible I'm missing some really good stuff that way -- had I known Colleen McCoullough wrote The Thornbirds, I might never have picked up First Man in Rome, and I think that would have been a shame. But you gotta play the odds, and generally, decisions made on past experience have merit.

How to keep this from degenerating into a popularity contest, of the sort that scientists are accused of by, say, Flat Earthers? Well, firstly, utmostly, and always, one has to keep an open mind and be willing to take some risks -- and certainly, to listen to colleagues who take them and report pleasant surprises. Most likely, with papers as with many things in life, the best bet is a triage: Pile "A" for the authors whose previous work has been always stellar; pile "C" for authors generally on the edge if not the fringe; and the huge pile "B" for everyone else. And then exert the self-discipline to pick up a "C" pile paper every once in while. Oh, and reconcile yourself to the fact that, more often than not, your "C" pile paper is going to be a crock of BS and a waste of your time. The thrill of discovering the few exceptions, though, will probably more than pay you back.

Anonymous said...

This has been going on for a long time, sadly. We haven't learned anything from it by trying to correct it. This behavior is what tempts renouned scientists to steal ideas too (unfortunately).

It's particularly dangerous when a well-known scientist gets to a review a paper in which something revolutionary is proposed by an unknown person. Everyone knows why the term "Bose-Einstein" has Einstein in it !

Alfonso Fanjul said...

I read the title, abstract, the author list, maybe the conclusions, before I decide whether the paper is something of interest.