November 11, 2014
nobabies.net
Nature
correspondence@nature.com
Sir:
I have read with interest Code Share, NATURE vol. 514 no. 7524 October 30, 2014 page 536.
In 1973 J. Calhoun published a paper “Death Squared,” reporting his experience with a population of captive mice that had been given all the food, water, nesting material and space they ever needed. As expected, the population increased exponentially. Then unexpectedly and abruptly on about day 600 all live births ceased. The importance of this is impossible to exaggerate.
Over a period of several years I developed a computer program that modeled the process. Actually I had not at that time been able to find the Calhoun paper, but there was ample other evidence that indicated that this would inevitably happen.
I was reluctant simply to release the source code. At least one person, had had no real interest in the subject, seemed to undertake to steal it. But were I to post the code either it would be ignored or it would take wings and I would lose any control at all, almost certainly the former. One other person who I trusted enough to show the code only went pale and stammered, “My God. It’s ten ply.” I have simplified it in some ways since then and extended it in another way.
I went to the trouble and expense of placing the program on a disc which could not be copied and from which the source code was not available. This I have given away free to anybody I thought might have the slightest interest. So far as I am aware interest has been zero.
I found the program valuable in a paper a colleague and I published Herbert, M. L.; Lewis, M. G., Fluctuation of Fertility with Number in a Real Insect Population and a Virtual Population, African Entomology 2013 21(1), 119–125, and am working on a paper where the same program applied to further questions again is valuable.
But having read your article I am still at a loss as how to proceed in the best interest of everybody. A commercial outfit might be able to make some money off it, but hardly if I had thrown it away and never unless there is a proven market. Of course I can’t use you as a resource; I’ve never published with you.
I understand it is frustrating for a researcher not to have access to source code of interest that the researcher is unwilling to develop. But that seems to me to be the way to go. Develop your own code, and if results to not tally make sure the two programs are giving the same results to the same input. If not, yes, then there is a problem I don’t know how to handle. But if they do give the same results, than one can bask in the satisfaction of knowing that replication has been successful.
Would you like a copy of the CD?
Sincerely,
M. Linton Herbert MD
There have been 121 visitors over the past month