[ExI] Paper and patent citations

[David Lubkin]

On another list, a poster reported a citation search to cast 
aspersions on a scientist:

>Breakdown of search results by citation
>
>All papers Published only
>Renowned papers(500+ cites): 0 0
>Famous papers (250-499 cites) : 0 0
>Very well-known papers (100-249) : 0 0
>Well-known papers (50-99) : 0 0
>Known papers (10-49) : 0 0
>Less known papers (1-9) : 4 1
>Unknown papers (0) : 2 0

I was aware of looking at the numbers of cites as a metric of the 
influence of a paper, but I hadn't seen these bins and labels.

(1) Would they be generally accepted as reasonable?
(2) What would an equivalent set of bins be for patents?

Have you seen a paper or web site that gave the distribution of 
cites? I'm looking for both papers and patents.

The labels for the bins seem overblown, even in a niche.

I don't track cites of my papers, but I do track patent cites. I have 
a patent for inventing heterogeneous configuration management. I'm 
proud of the work. I think I cleverly solved a tough technical 
problem. The patent is cited in 150 subsequent patents.

But I would guess that most programmers have not heard of HCM, let 
alone the patent. Out of over 10^6 programmers, I'd guess that there 
are 5 who have seen the patent, and haven't read it. Maybe 2 who read 
it,  because they are working with a patent attorney.

On the other hand, I'd guess that 50-100,000 or so were familiar with 
the ideas embedded in the patents, aggregating the applicable 
fraction of our user base with people who read one of the papers I'd 
written on the subject.

And, on the other hand, some of the best inventions in software were 
not patented.

Is number of cites a good metric for the influence of a paper? What 
would be a good measure for inventions (patented or not)?


-- David.