User talk:Joseph Grcar

From Wikipedia, the free encyclopedia

[edit] Matthew Effect

Hi Joe, thanks for your message.

First off, welcome to Wikipedia, and don't forget to sign your comments to talk pages with four tildes (~~~~).

I will try to reword the offending text so that it is attributed and properly cited, as it should be on Wikipedia, and hopefully this will alleviate your objections. Unfortunately, this may take some time, as I am out of the country and away from my library of notes and resources on this subject.

The only remark you made that seems certainly wrong to me is the bit about "all of these people could have written something, but none of them did", which seems wrong on both counts. It neglects that Mauchly and Eckert were working on a secret project and could not have published until the ENIAC's unveiling. It also neglects that they did write internal disclosures, including Eckert's on a magnetic calculator which predates von Neumann's involvement and which most will admit intimates the stored-program concept made explicit in the First Draft. This disclosure was entered into the exhibit evidence in HvSR. To my mind, an implementation description implies a logical design but the reverse is not the case. "All of these people could have written something" becomes "not all of these people could/should have written something, not even von Neumann" and "but none of them did" becomes "but anything they wrote was overshadowed by sheer readership numbers of the First Draft."

You also say that "he specifically took pains to avoid mentioning anyone's hardware", but a read of the text clearly shows that he was convinced by the success of the ENIAC's engineering that electromechanical implementations were passée ("It is clear that a very high speed computing device should ideally have vacuum tube elements"--he further goes on to talk about delay times for electronic components) and that cathode ray tube memories were going to be improvements over Eckert's ideas for high-speed serial access memories using acoustic delay lines ("The solution to which we allude must be sought along the lines of the iconoscope. This device in its developed form remembers the state of 400 x 500 separate points..."). Indeed, von Neumann held out for the development of cathode tube memories for his IAS machine, whereas the EMCC developers gave up their experiments with them and went ahead with mercury delay lines.

As to Norbert Weiner, speaking practically, he had nothing to do with the development of computing machines, and Mauchly and Eckert were, IIRC, in contact with Aiken and Stibitz and so von Neumann had no particular leg-up on them as far as connections to other computing developments went. Indeed it was von Neumann who was the lucky one to have been introduced to the Moore School group, as Goldstine was always fond of recounting. While von Neumann probably would have found out about it sooner or later, Goldstine's admiration for him probably had probably no small part in von Neumann being invited to consult for the EDVAC.

Cheers, thanks, and good work. Robert K S 15:56, 7 March 2007 (UTC)

Hi again Joe,

You say, "Correct me if I am wrong, but I believe the only publicly distributed description of a modern computer though about 1950 is the First Draft Report. As you say, it came out in May/June of 1945." I cannot confirm that you are correct on the first point, but might offer as one potential counterexample among many the Moore School lecture volumes, which were surely much more exhaustive in depicting the state of the art in computing than the First Draft was. I must correct you on your second point, as I didn't say it came out in May/June of 1945, and I have no reason to doubt Goldstine's assertion that no copies of it were circulated to anyone prior to June 25, 1945.

"The excuse about security is a red herring." The facts are these. An incomplete document meant for internal distribution ended up being read by people outside of the project for whose members it was written and from whose members some ideas were drawn unattributed. It was the security officer on the project who was responsible for the distribution. Some months after leaving the Moore School, Eckert, Mauchly, Goldstine, von Neumann, Pender, and others met at the Moore School to resolve issues related to intellectual property, and all parties came away with the understanding that the First Draft's publication precluded patentability of any of the inventions contained therein. The genie was out of the proverbial bottle. Von Neumann would have liked to patent the EDVAC if for no other reason than to preempt a commercial monopoly on the electronic digital computer, and indeed sought to do so.

"After March 1946 when Mauchly and Eckert technically resigned (actually were fired) from the University Pennsylvania they were free to write anything they pleased, but they wrote nothing." All of which is irrelevant as per the above, except your insistence that Eckert and Mauchly "actually were fired", which is not exactly true and somewhat bias-revealing. Mauchly and Eckert were posed the choice of signing away the intellectual property they had been promised or else submitting letters of resignation, and so they chose the latter, but by this time they had already spent considerable energy sending out feelers for investment in the commercialization of the technology. I am fairly certain you can read their letters of resignation, as copies of them still exist.

"...But Mauchly and Eckert did nothing." I could try to provide you with lists of their disclosures and publications, but you've caught me at a disadvantage since I don't have access to my research. "Did nothing" seems a little strong when you consider that they taught the first computer course, began the commercial computer industry in the United States, and Mauchly helped found the Association for Computing Machinery. Though once out of academia they had little incentive to publish their research, they were cheerful to disseminate their knowledge. Should they have been more worried about their legacies, or their names being attached to certain concepts? I'm not sure either of them ever gave a thought about it, except retrospectively in their retirements.

Finally: "Norbert Wiener deserves some credit for [von Neumann's abstract description of a machine] because in the 1940s he was working on cybernetics... Wiener organized a series of gatherings on the topic... and he invited von Neumann to attend. It is from these meetings that von Neumann got the idea to use biological terminology in the First Draft Report: memory, etc." I'm interested by this. Can you source the assertion that that von Neumann used biological analogies as a result of his attendance of the Wiener meetings? Do you assert that the use of the word "memory" in particular was original to von Neumann?

It is a pleasure. Robert K S 18:36, 7 March 2007 (UTC)