« women | Main | playing dumb »
November 29, 2003
history, credit and identity
Like many students of computing, i was inspired by Vannevar Bush from my earliest days. "As We May Think" and follow-up writings on the Memex helped define a century of thought and computational effort.
Yet, as Michael Buckland is uncovering, much of Bush's fame is misplaced. Bush's seminal ideas around the Memex were actually developed and patented over a decade before Bush by a Jewish chemist named Emanuel Goldberg.
Goldberg's role in history was eclipsed in part because of his identity as a Jewish German man. (The German's did not respect the thought of the Jews and post-WWII, the world did not respect the thought of the Germans.) He is not alone. Only recently has Ada Lovelace been given credit for her contributions to scientific computing; she was eclipsed by Charles Babbage because of her identity as a woman.
I wonder how many other inventions in technology are not properly credited to their creators because of their identity.
Posted by zephoria at 05:50 AM in History | Permalink
Comments
Goldberg invented a document retrieval system. You gave a specific filename to the machine, you got back a specific document, and that's as much as the machine ever did. Bush extended the idea of such a machine to the concept of links, paths through links, and paths as legitimate works of authorship.
Goldberg may be the father of document retrieval, but Bush is the father of hypertext.
Posted by: Julius at Nov 29, 2003 6:36:10 AM
There was the Nobel Prize for discovery of DNA, which was given to the two male researchers involved, but not to Rosalind Franklin, their female collaborator. Googling I discover though that Franklin had died, four years before the Nobel prize was awarded (at just 37!), so perhaps they just don't award the prize posthumously.
Here's one version of Franklin's story; here's a list of women who've won the Nobel Prize.
While I'm sure identity and group membership makes your visibility far simpler, I think it's quite common for a great scientific discovery to go unnoticed until someone else rediscovers it a few years later. Quite probably many of those cases are due to gender, ethnicity, social connections and so on.
Posted by: Jill at Nov 29, 2003 6:40:40 AM
Vannevar Bush's fame is not singularly tied to 'as we may think'. he was very much a polymath of sorts, and was a significant figure in several fields, this allowed him more popular press access of course, but if you check out his accomplishments on the wikipedia page here, I think you'll see that his fame is a bit broader based and that some of his other works are foundational in other fields. he was even on the cover of Time magazine for his work in physics
a short biblio includes such works as:
modern arms and free men: a discussion of the role of science in preserving democracy
principles of electrical engineering
science is not enough
pieces of action
endless horizons
two codicils:
1. bush cetainly falls into the 'great man' problem of history, he is usually individuated and put forth out of his myriad of contexts, so we have to be careful about what his role really was in regards to certain concepts that he put forth, which could in fact be hinting at another problem in science studies, which is that we traditonally put undo emphasis on the people that do something first, usually singling them out in opposition to others, when they very well may have been aware of the others work and thought they were working in a larger framework. so saying 'goldberg or bush' is problematic, when it could have been that bush was seeking to popularize goldberg or something else.
2. memex has nothing to do with hypertext and everyone knows hypertext is dead (said snarkily) ;)
Posted by: jeremy hunsinger at Nov 29, 2003 9:43:01 AM
>"Goldberg invented a document retrieval system. You gave a specific
>filename to the machine, you got back a specific document, and that's as
>much as the machine ever did.
>
This statement is not correct. As is clear from Goldberg's papers and patents, the originality of Goldberg's "Statistical Machine" is that it was a genuine search engine. You entered a query by specifying a code for the characteristic you were interested in (i.e. the code for a sales region; a product code; price; or whatever the application was concerned with). The machine would then find each and every document coded in that way. Previous document retrieval machines required one to know the address (location) of the desired document. The filename could be used only if the documents were in filename order.
http://www.sims.berkeley.edu/~buckland/goldberg.html
>Bush extended the idea of such a machine to
>the concept of links, paths through links, and paths as legitimate works
>of authorship.
>
This is difficult to comment on: Bush's notion of "links" was not different, in principle, from any other form of indexing that brings together documents with a shared characteristic, such that starting with any one (or the link) leads to all others. The creation of indexical links is and always was a creative descriptive activity and can be seen as a form of authorship, but usually the indexers are not so recognized.
>Goldberg may be the father of document retrieval, but Bush is the father
>of hypertext."
>
There are several roots to the concept of "hypertext."
Of particular interest is the largely forgotten work on hypertext by Wilhelm Ostwald. 1853-1932, who did a kind of technical transfer from "monos," cards distributed in Germany as a marketing gimmick. Ostwald's name for hypertext was "the Monographic Principle." In 1910 he formed an institute in Munich to work on it. He got quite lyrical in 1911 about how one could use hypertext to build a "World Brain," a theme developed for a while by H. G. Wells. Ostwald spent his Nobel Prize money on the topic.
How far Bush was aware of what others had already done is unclear.
I recommend:
Serres, Alexandre. 1995. Hypertexte: Une histoire à revisiter. Documentaliste - Sciences de l'information Vol. 32, no 2:71-83. Summary: The commonly held view that hypertext was conceived by Vannevar Bush in 1945, developed by Douglas Engelbart and Ted Nelson, implemented through Hypercard and html simply ignores most of the history of hypertext. Five origins of hypertext are reviewed: Recent: 1. Documentalists 1930s onwards. 2. Computing, esp. 1960s onwards. Older: 3. Techniques of printed book design; 4. Utopian plans of universal knowledge; 5. Memorization techniques.
Posted by: Michael Buckland at Nov 29, 2003 3:05:47 PM
I don't think there's much percentage in the "who invented hypertext?" debate as formulated by some above. It's like saying that people had concepts closely related to the talking cure before Freud (via Breuer, via Anna O). The Freudian formulation is important because it was influential and it is in this work that we find the definition of the talking cure, whatever predecessors and later reformulations may exist. We return to Bush's work because of its profound influence on Nelson and Engelbart. And it is in the work of Nelson, who coined the term, that we find the definition of hypertext, whatever predecessors and later reformulations may exist.
Posted by: noah at Nov 29, 2003 6:01:47 PM
Madame Wu should have shared the Nobel Prize for her amazing experiment in Parity violation.
Four women scientists who should have won Nobel prizes:
http://www.voting.ukscientists.com/womensci.htm
(and I know enough physics to realize that these claims are just the obvious ones and, if anything, understated)
Posted by: ricardo at Nov 29, 2003 8:21:20 PM
But - big gaping hole, noah. You almost certainly know that whose ideas get credited (are "influential") has a lot to do with politics and personality; political loyalty to Newton/Leibniz notation, or Marie Curie remarking that she had to move to France to work because the social pressure on an immigrant to be 'ladylike' was weaker than that on a native in either Poland or France. Ramanujan, Hardy. Gauss voluntarily not publishing on non-Euclidean geometry, to avoid fuss!
Posted by: clew at Nov 29, 2003 10:02:29 PM
Clew, I'm not sure I understand what you're getting at. It's clear that many people talked about things that Bush also talked about, and did so before he talked about them (we can probably add Buckminster Fuller to those already listed). The same is true of Freud. But the fact that Bush and Freud authored enormously influential formulations of certain ideas is part of our intellectual history. I don't think these influential texts came from lone genius within these authors. I think these texts came from the world, through these authors. And it's no surprise that other people were serving the author function for similar formulations, and doing it earlier. But, as it happens, the texts they functioned to author didn't turn out to be the influential ones.
Certainly, the process by which it gets decided which authors' texts will have influence is a social one Bush's position of power got "As We May Think" prominent placement in two very influential US publications. His position also made it possible for him to get his research on analog computing funded, etc.
But just like the talking cure, or natural selection, hypertext is a term that comes from a particular author in this case, Ted Nelson and his work serves to define it. We've adopted the term, and the concept, and with it we've adopted some of Nelson's reported intellectual heritage (e.g., Bush). Nelson has had his influence in turn. If, instead, it had been Ostwald's "Monographic Principle" that was adopted, or that had influenced Nelson and Engelbart, Bush might mostly come up in reference to the Manhattan Project.
I don't think there's anything wrong with unearthing these alternate histories. They show the different ways these ideas were taking form in different places, and as expressed through different people. But these discoveries are only shocking if you believe in lone genius, achieving insight where no one else could have, followed by these insights fanning out meritocratically to enlighten the world.
Which is to say, Clew, I don't see any contradiction between your comment and mine.
Posted by: noah at Nov 30, 2003 1:40:00 AM
Lovelace was likely a fraud as explained by Jim Holt in a New Yorker article a while back. Sorry no reference.
And Wu was an experimentalist. Although ingenious, and demonstrating the predictions of Lee and Yang was indeed amazing, theoreticians usually almost always get the credit. So I'm not sure how true claims are she didn't receive a Nobel Prize "because she was a woman" are exactly. Perhaps you could design an experiment around it though :D Anyway, this isn't to denigrate the achivements of women in science and engineering! Instead, it's to preemptively counter some highly visible, but not so very good, examples of discrimination in order to present a stronger case for recognition of their work, rather than a straw man (or woman, if you will :) no matter how well conceived and with the best of intentions!
Posted by: Bradley S. Felton at Nov 30, 2003 10:43:58 PM
Bradley, that New Yorker article's conclusions are, let's say, far from generally accepted. See, for example, the Tech TV piece on it.
Excerpt: But according to Toole, this characterization is "pure hogwash." Toole says Babbage's correspondence with Ada reveals that he gave her very little help. In fact, Toole argues, it was Ada who suggested programming the Bernoulli numbers -- a claim Holt and Woolley both support -- and using indexes, much like those used in modern computers.
Posted by: noah at Nov 30, 2003 11:21:54 PM
Hey, found it :D Hmm, noah very intriguing indeed... I don't want to turn this into a "he said, she said," so you be the judge!
Posted by: Bradley S. Felton at Dec 1, 2003 12:22:08 AM
