Multiples and mashups


I've had the good fortune of meeting Matt McAlister of Yahoo, great thinker, really nice guy.


For some reason, an old post of his cropped up in my reader today. Not only did the piece point out some good mashup / Web 2.0 material (that's my excuse for posting on this blog), but it also highlighted a phenomenon I find fascinating. Some call it scientific serendipity, others (Merton, see below) have framed it as 'singletones' vs. 'multiples', others have termed it as 'scientific zeitgeist'


First, Matt provides some background:



"The other day I posted about giving a talk on mashups to people who were new to this concept.  I find out since then that there were lots of other people doing the same thing at nearly the same time.  Then I saw that several editors at high profile publications were publishing stories explaining mashups to nontechnical audiences at about the same time, as well."

Rex Hammock, Richard MacManus and Joshua Porter gave mashup presentatons and even posted the powerpoint files on their blogs.  Verne Kopytoff of The SF Chronicle and Elinor Mills of CNET posted news analysis pieces on mashups."


Since the advent of the Wonderful, Wonderful Web, this kind of realization must be happening more often - the realization that others are thinking *a lot* like you.  Not surprising really - the network and its search engines have made it increasingly easy to discover these common trains of thought.  In one sense, the more of the web I discover, the more I realize that an 'original thought' is a very rare happening indeed. In another sense, the more I discover of the web, the more original thoughts I seem to have - original to me that is.


Matt goes on:



"I find it bizarre yet not necessarily surprising that people process the same ideas at nearly the same time.  What surprises me is that there isn't more scientific understanding of why this occurs."


There is a fair bit of research in this area. I point out a summary of one theory- Robert Merton's 'multiples' theory. The following is an except from this Stanford paper (pdf).   The context is the 'state of the art' of developments in speech recognition, natural language processing, and AIs:



"Are these multiples to be considered astonishing coincidences? A well-known hypothesis by sociologist of science Robert K. Merton (1961) argues, quite the contrary, that



all scientific discoveries are in principle multiples, including those that on the surface appear to be singletons.


Of course there are many well-known cases of multiple discovery or invention; just a few examples from an extensive list in Ogburn and Thomas (1922) include the multiple invention of the calculus by Leibnitz and by Newton, the multiple development of the theory of natural selection by Wallace and by Darwin, and the multiple invention of the telephone by Gray and Bell. (3)


But Merton gives a further array of evidence for the hypothesis that multiple discovery is the rule rather than the exception, including many cases of putative singletons that turn out be a rediscovery of previously un-published or perhaps inaccessible work. An even stronger piece of evidence is his ethnomethodological point that scientists themselves act under the assumption that multiple invention is the norm. Thus many aspects of scientific life are designed to help scientists avoid being “scooped”; submission dates on journal articles; careful dates in research records; circulation of preliminary or technical reports.


(3) Ogburn and Thomas are generally credited with noticing that the prevalence of multiple inventions suggests that the cultural milieu and not individual genius is the deciding causal factor in scientific discovery. In an amusing bit of recursion, however, Merton notes that even this idea has been multiply discovered, citing sources from the 19th century and earlier!"


I love the self-referential point in the footnote. Btw, in case you're wondering, 'ethnomethodology' simply means the study of the ways in which people make sense of their social world.


Back to Matt:



"Maybe the next leap in search after social and semantic connections is the study of spontaneous order.  If so, then somewhere in the Silicon Valley there are super secret research projects that aim to identify memes before they actually get expressed."


Wouldn't that be cool....I've played with this 'memes before they boom' idea. The approach I've taken has been to identify existing memes that are pre-tipping point. But alas (and entirely in keeping with the topic at hand), meme tracking, predictive theory of vectors of memes and memetic propagation are flogged horses.


Now, Matt's idea of identifying before they actually get expressed, I'd invest in that one...


--


Further reading on the topic multiple discoveries:



--


Tags: , ,

Comments (1)

  1. Google the phrase "Steam Engine Time".

Skip to main content