The thing to remember about this question is that in January 1944, Eckert wrote a short document titled "Disclosure of Magnetic Calculating Machine" that discussed using disks or drums for primary storage reflecting initial discussions related to the EDVAC. In it, he includes the statement: "If multiple shaft systems are used a great increase in the available facilities and for allowing automatic programming of the facilities and processes involved may be made, since longer time scales are provided. This greatly extends the usefulness and attractiveness of such a machine. This programming may be of the temporary type set up on alloy discs or of the permanent type on etched discs." How much earlier than January 1944 this idea of programming was floating around the Moore School is not known, but it's clear that at least by early 1944, there was a recognition that storing programming on the same media as data had substantial advantages. There is every indication that von Neumann first learned of the work at the Moore School in July 1944 at a chance encounter with Herman Goldstine, when Goldstine told him of the ENIAC. At least by August 1944, von Neumann visited the Moore School to see the ENIAC, at which time they were testing a two-accumulator configuration. The bottom line is that there's not much question that by the time von Neumann was involved with the EDVAC project, about which his "first draft" was written, the idea of the stored program was already established at the Moore School. However, the situation is not as simple as this. The idea of the stored program would not have been in any way alien to von Neumann, and it would not surprise me at all if he did make significant contributions to the conversation about what form it would take. This is because von Neumann was well aware of Turing's "On Computable Numbers." Remember that Turing did his PhD at Princeton after the publication of that paper. Although the IAS isn't really a part of the university, there can be no doubt that there was a lot of contact among people like Church, Turing, Godel, and von Neumann. Further, in von Neumann's examinations of cellular automata, he specifically mentions the Turing Machine. We also have no record of when Turing first started thinking about embodying the universal machine of his paper in physical form. He didn't really write much of anything about that during the war. However, because he moved so quickly into working with NPL after the war, it seems very likely that he was already thinking about it at Bletchley. Ultimately, trying to assign an individual and a name to the concept is unproductive. The ideas themselves are much more interesting and important than credit. BLS

On 2015-Feb-26, at 7:51 AM, Evan Koblentz wrote: Thanks for the link, the articles were very interesting. However: The authors of the articles carefully avoid declarations of ENIAC as a stored-program machine. They separate out 3 aspects from the "First Draft of a Report on the EDVAC" (the "modern code paradigm" as defined by them, the "von Neumann architecture paradigm", and the "EDVAC hardware paradigm") and show how the 1948-modified ENIAC (to provide it with an instruction-set-based programming capability) implemented the "modern code paradigm". While that was a couple of months before the Baby, it is not the same (by the authors own definitions and intent) as making it a stored-program machine. In modern terminology, I would suggest the 1948 ENIAC would be classed (as far as it can be) as Harvard-Architecture or a form of modified-Harvard-Architecture machine. Von Neumann was around the project years before the 46-48 modifications. He produced the Draft Report on the EDVAC - from which those modifications were conceptually derived - in 1945. As presented in the articles, the modifications were von Neumann's idea and implemented in consultation with him.

On 2015-Feb-27, at 1:15 PM, Evan Koblentz wrote: Which is why they're avoiding making a "stored-program" declaration in regards to the ENIAC. Note also, as they present, "stored-program" as a phrase is itself post-ENIAC. So M&E, or their proponents, claim. You should have said what you meant then. Earlier, I said: and you replied: The articles do not in any way get to the bottom of this, they do not address the origins of the stored-program concept, nor do they contradict or diminish von Neumann's place or influence in the stored-program matter. (And to be clear, I'm not disagreeing with the articles, they were very informative.) I know you like to support the ENIAC camp, Evan, but from what I see, you're not helping it.

On 2015-Feb-27, at 3:02 PM, Evan Koblentz wrote: Then you're not reading very closely or understanding the article. The concept Evan, not an implementation, or partial implementation, or aspect, derived from the concept. 1945, not 1948.

Tom stated (in an email last fall): "we do conclude that an entirely practical version of the programming method described in the seminal 1945 First Draft of a Report on the EDVAC was implemented on ENIAC in March-April 1948, some months before the Manchester ?Baby? ran what has usually been considered the first modern program."