3 ms·

Still a semi-positive outcome no? Computers becoming cheaper?

masswerk 6y ago
J. Presper Eckert (the Eckert in "Eckert & Mauchly") in his oral history interview [1], p.46f: > "Originally we called ENIAC the "MANIAC" when it didn't work right. And later they borrowed that name for some other actual machine. But if you worked for von Neumann on the MANIAC, then if you invented something it belonged to you. Well, on some relatively short notice, like it might have been a week, or a month, or something, a short time before the deadlines hit, von Neumann went down and published all that stuff. All the reports of the engineers went to the Library of Congress which put a bar on any patents being obtained by any of his employees. And when they complained about it to him, he just said, "Well, that's tough; that's the way I think; that stuff should be in the public domain." Now there is a perfectly obvious reason for this. He was consulting with people like IBM. If the things weren't patented that would be a problem for IBM. The idea was he was selling ideas to other people...if it wasn't covered by patents, he would have been selling something they couldn't use. They would have come back and said, now, "what kind of a consultant are you, coming up with new ideas which are already patented by others and we can't use them." But if these ideas could come under the public domain, then he could go around and sell them to people. That was his game." [1] Oral history interview with J. Presper Eckert (1977): https://conservancy.umn.edu/handle/11299/107275 https://conservancy.umn.edu/handle/11299/107275
- adrian_b 6y ago
  For the humanity, the publication of the von Neumann report was certainly a very positive thing because it launched a large number of research projects with the purpose of building electronic computers, both in USA and in many other countries. There is absolutely no doubt that without this publication and with a patent claiming the design of a computer with stored program, the evolution of the computers would have been much slower. Whoever pushed the report to be published did a huge service to everybody else but Eckert and Mauchly. Also, Eckert and his coworkers were deluded if they thought they could have gained anything by having a patent on the stored-program computer. Nobody prevented them to build good computers even without a patent, but they did not have the intellectual and financial resources to do much until they were bought by Remington Rand, which lead to the introduction of UNIVAC in 1951. If they had a patent, the result would have been that Remington Rand would have had a monopoly on the computer market, with inferior products, and all the great advances in computers that have been done in other places, like NIST, MIT or IBM, would have happened only many years later. I have read both the von Neumann report, which is a model of clarity, and all the early documents that remain from Eckert, Mauchly et al. It is likely that they already had some ideas about stored-program computers, but there is no doubt that their ideas were very confused. Von Neumann distinguished what was right from what was wrong in their ideas, then clarified, organized and completed all those concepts and synthesized them in a coherent theory that could be easily understood and used by anyone else. The contribution of von Neumann is much more important than Eckert tries to imply. Moreover, while Eckert and al. were upset that their own contributions were not acknowledged as well as they should have been, they also failed to acknowledge the important ideas about electronic computers that they themselves have taken from Atanasoff, without ever mentioning him. Unlike Eckert and al., but like von Neumann, Atanasoff was also someone able to express very clearly his concepts, so you could easily learn from him. For example a report about the classification of computer memories, written by Atanasoff at the beginning of WWII, was extremely advanced for those days (Atanasoff invented what is now called DRAM).
  - adrian_b 6y ago
    I want to add that the fast evolution of computers caused by the publication of the von Neumann report and the lack of "IP protection" is paralleled by the fast evolution of semiconductor devices after the discovery of the transistor, which was also caused by the lack of "IP protection" in the modern sense. Even if the transistors have been patented, the patents have been licensed to anyone for negligible fees or even for free (e.g. for hearing aids). If the essential transistor patents had been used like worthless patents are used today in preventing competition, the history of the computer and electronics industries would have been very different, with progress delayed by decades. So both the publication of the von Neumann report and the generous handling of the transistor patents have been very influential in shaping the word of today, while many inventions where the inventors clinged to the patents have been either doomed or they began to be used only after the patents have expired, more than 20 years later.
  - masswerk 6y ago
    Regarding "very confused ideas": To be fair, Mauchly proposed combined program and data storage in a mercury delay line, much like it was done on the UNIVAC I, in 1945, before von Neumann arrived. (Allegedly, there had been ideas for stored program in the ENIAC group even earlier than this.) Mauchly also invented Short Code (1949, originally named Brief Code), the first programming language used in production, and the Critical Path Method for automatic scheduling. The group around Maurice Wilkes attended his 1946 lectures on the design of digital computers and went from there to build the EDSAC, the very first modern computer.

J. Presper Eckert (the Eckert in "Eckert & Mauchly") in his oral history interview [1], p.46f: > "Originally we called ENIAC the "MANIAC" when it didn't work right. And later they borrowed that name for some other actual machine. But if you worked for von Neumann on the MANIAC, then if you invented something it belonged to you. Well, on some relatively short notice, like it might have been a week, or a month, or something, a short time before the deadlines hit, von Neumann went down and published all that stuff. All the reports of the engineers went to the Library of Congress which put a bar on any patents being obtained by any of his employees. And when they complained about it to him, he just said, "Well, that's tough; that's the way I think; that stuff should be in the public domain." Now there is a perfectly obvious reason for this. He was consulting with people like IBM. If the things weren't patented that would be a problem for IBM. The idea was he was selling ideas to other people...if it wasn't covered by patents, he would have been selling something they couldn't use. They would have come back and said, now, "what kind of a consultant are you, coming up with new ideas which are already patented by others and we can't use them." But if these ideas could come under the public domain, then he could go around and sell them to people. That was his game." [1] Oral history interview with J. Presper Eckert (1977): https://conservancy.umn.edu/handle/11299/107275 https://conservancy.umn.edu/handle/11299/107275

adrian_b 6y ago
For the humanity, the publication of the von Neumann report was certainly a very positive thing because it launched a large number of research projects with the purpose of building electronic computers, both in USA and in many other countries. There is absolutely no doubt that without this publication and with a patent claiming the design of a computer with stored program, the evolution of the computers would have been much slower. Whoever pushed the report to be published did a huge service to everybody else but Eckert and Mauchly. Also, Eckert and his coworkers were deluded if they thought they could have gained anything by having a patent on the stored-program computer. Nobody prevented them to build good computers even without a patent, but they did not have the intellectual and financial resources to do much until they were bought by Remington Rand, which lead to the introduction of UNIVAC in 1951. If they had a patent, the result would have been that Remington Rand would have had a monopoly on the computer market, with inferior products, and all the great advances in computers that have been done in other places, like NIST, MIT or IBM, would have happened only many years later. I have read both the von Neumann report, which is a model of clarity, and all the early documents that remain from Eckert, Mauchly et al. It is likely that they already had some ideas about stored-program computers, but there is no doubt that their ideas were very confused. Von Neumann distinguished what was right from what was wrong in their ideas, then clarified, organized and completed all those concepts and synthesized them in a coherent theory that could be easily understood and used by anyone else. The contribution of von Neumann is much more important than Eckert tries to imply. Moreover, while Eckert and al. were upset that their own contributions were not acknowledged as well as they should have been, they also failed to acknowledge the important ideas about electronic computers that they themselves have taken from Atanasoff, without ever mentioning him. Unlike Eckert and al., but like von Neumann, Atanasoff was also someone able to express very clearly his concepts, so you could easily learn from him. For example a report about the classification of computer memories, written by Atanasoff at the beginning of WWII, was extremely advanced for those days (Atanasoff invented what is now called DRAM).
- adrian_b 6y ago
  I want to add that the fast evolution of computers caused by the publication of the von Neumann report and the lack of "IP protection" is paralleled by the fast evolution of semiconductor devices after the discovery of the transistor, which was also caused by the lack of "IP protection" in the modern sense. Even if the transistors have been patented, the patents have been licensed to anyone for negligible fees or even for free (e.g. for hearing aids). If the essential transistor patents had been used like worthless patents are used today in preventing competition, the history of the computer and electronics industries would have been very different, with progress delayed by decades. So both the publication of the von Neumann report and the generous handling of the transistor patents have been very influential in shaping the word of today, while many inventions where the inventors clinged to the patents have been either doomed or they began to be used only after the patents have expired, more than 20 years later.
- masswerk 6y ago
  Regarding "very confused ideas": To be fair, Mauchly proposed combined program and data storage in a mercury delay line, much like it was done on the UNIVAC I, in 1945, before von Neumann arrived. (Allegedly, there had been ideas for stored program in the ENIAC group even earlier than this.) Mauchly also invented Short Code (1949, originally named Brief Code), the first programming language used in production, and the Critical Path Method for automatic scheduling. The group around Maurice Wilkes attended his 1946 lectures on the design of digital computers and went from there to build the EDSAC, the very first modern computer.