Congratulations to George! 

Citations are an interesting sociopolitical phenomenon, particularly 
when it comes to methods.  Most people cite computer programs that they 
used, but certainly not all of them, and almost never cite "laboratory 
tricks".  Case in point: one would think that T.-Y Teng's J. Appl. 
Cryst. (1990) paper describing a (then "new") method for mounting a 
crystal into a "loop" and plunge-cooling it in liquid nitrogen would be 
the most heavily cited work in all of protein crystallography today.  
Yet Google Scholar tells me it has only been cited 311 times (in 20 
years).  This is in stark contrast to "the most heavily-cited work in 
all of science": O. H. Lowry et al. JBC (1951), which describes the 
"Lowry assay" of measuring protein concentration and has been cited 
about 200,000 times.   However, I think if you count up all the 
citations to misspellings of Laemmli's name, the U. K. Laemmli, Nature 
(1971) paper might have more citations than Lowry.

Even more interesting (I think) is that if you read Laemmli's paper you 
will find that the description of the method we now refer to as a 
"Laemmli gel" is entirely crammed into a single paragraph at the bottom 
of one figure caption.  The rest of the paper is about phage head 
proteins.  The funny thing is, a lot of young biochemists seem to think 
that Laemmli invented electrophoresis (actually, that was A. Tiselius, 
Trans. Faraday Soc. (1937), who was awarded the 1948 Nobel for that, and 
a few other things like ion exchange, reverse phase and affinity 
chromatography).  Tiselius is one of my favorite scientists because he 
transformed his field so much that noone can remember his name.  
Admittedly, this does not have to be the case: a series of papers in 
1905 by a fellow named "A Einstein" (mostly in Ann. Phys.) transformed 
physics and biophysics alike.  People remember his name, but ironically 
almost never cite his papers.  The paper explaining Brownian motion and 
diffusion coefficients has only been cited ~1500 times in more than a 
century.  Still, this is more citations than his paper on something 
called "Special Relativity", published the same year.

So, what do you have to do to get people to cite your methods paper?  
Near as I can tell, the method you describe must be highly useful (but 
not too useful) and also very difficult to comprehend.  I don't mean 
that the paper should be poorly written, but the sad truth is that 
people don't generally cite "methods" that they think are "obvious" 
(inasmuch as they understand how it works, and think everyone else does 
too).  People also don't cite methods that they think they could have 
come up with themselves, and especially not those they see as a "common" 
commercial product (like mini-prep kits).  Generally, something from 
"outside the field" must be part of the method for it to be "citation 
worthy", so for biologists this can be chemistry (copper binds to ALL 
proteins?  Really?), physics, or especially mathematics.  Computer 
programs are particularly well-suited for this, but it can't be a 
computer program that does something "transparent" like sequence 
alignment or collecting diffraction data (ahem...).  In these cases, the 
user knows (or thinks they know) exactly what the program is doing, and 
assumes that their audience will too, so why cite it? 

On the other hand, one must also be very careful not to produce an 
algorithm that is "too useful" and rapidly becomes incorporated into 
everyday life.  An excellent example of this is L. Ten Eyck's work on 
something called "FFT" (L. F. Ten Eyck, Acta A 1973).  I think he wins 
the prize for the largest "unfairness ratio", which I define as: (papers 
that used the method)/citations.  Perhaps it is important to give your 
program a memorable name.  However, as long as you have some fancy math 
in there (like "direct methods" or "likelihood"), or at least pretty 
graphics AND the program can do something that no other program can 
(such as solve a structure that was "hard" enough to end up in a 
high-impact Journal ... like Acta A), then you've definitely got a 
"citation classic" in the making.

BTW.  I hope everyone understands that in no way do I mean to belittle 
the efforts of those who write heavily-cited computer programs.  Quite 
the contrary.  I think they are simply fortunate to have an "unfairness 
ratio" close to 1.

-James Holton
MAD Scientist




Paul Emsley wrote:
>> Well, good luck to all the methods-folk who are up for tenure, here 
>> is your chance guys and girls ... it will not last long!!!
>
>
> Indeed.
>
> http://community.thomsonreuters.com/t5/Citation-Impact-Center/What-does-it-mean-to-be-2-in-Impact/ba-p/11386 
>
>
>
> p.s. "methods-folk who are up for tenure"? - haha...