Visualization & Science forum: topic
This is a public forum
Mapping Nature
Hilary Spencer
Wednesday, 30 May 2007 18:05 UTC
An aesthetically pleasing version of a citation network, created by “sorting roughly 800,000 scientific papers (shown as white dots) into 776 different scientific paradigms (red circular nodes). The author describes the picture as "hairy” and “organic”.
Frankly, I’m not sure how useful a visualization like this is. The image is not interactive, however it seems that a project like this would benefit significantly from features like mouse rollovers. Finally, I’m curious about how the paradigm clustering was done and, from a philosophical standpoint, exactly what a “paradigm” is in this context.
Updated 12 December 2007 21:51 UTC
-
Replies
-
The concept is interesting…
-
The visualization does look visually pleasing. But I doubt it is useful.
My academic colleague, Nobel Laureate Barry Marshall, published his key findings about Helicobacter pylori in 1983.
Warren, J.R.; Barry Marshall (1983). “Unidentified curved bacilli on gastric epithelium in active chronic gastritis”. Lancet 1 (8336): 1273-5. PMID 6134060It had to be published as a letter to the Lancet, yet Barry’s papers have gone on to become some of the most cited papers of all time.
I feel that when trying to depict paradigms, it is more instructive to look at the gap between the initial date of publication, and when the frenzy of “me-too” citation started in earnest.
For, at least based on Barry’s experience, it took a decade for this paradigm to gather much inertia at all.
I suspect that if one could depict such delays in important paradigm shifts, it would tell us a lot about the state of science in the 21st century.
Frankly, I have been publishing since the 70’s and I have never seen the level of resistance to new ideas that I encounter today, especially in circles that like to think of themselves as ‘scientific.’ IMO, true paradigm shifts are not depicted solely by the number of citations, but by the temporal pattern with which they occur.
-