Talk:John Mauchly

From Wikipedia, the free encyclopedia

	This article is within the scope of WikiProject Biography. For more information, visit the project page.	Biography Portal
???	This article has not yet received a rating on the Project's quality scale. Please rate the article and then leave a short summary here to explain the ratings and/or to identify the strengths and weaknesses of the article. [FAQ]

	This article is within the scope of WikiProject Physics, which collaborates on articles related to physics.	Physics Portal
???	This article has not yet received a rating on the assessment scale. [FAQ]
???	This article has not yet received an importance rating within physics.
Help with this template Please rate this article, and then leave comments here to explain the ratings and/or to identify the strengths and weaknesses of the article.

Contents 1 Pronunciation 2 Flip-flop circuit claim 3 ENIAC patent case 4 ABC computing speed 5 Programmability

[edit] Pronunciation

Mauchly is pronounced MAWKly: there is no "ch" sound. Jfgrcar 04:41, 15 December 2005 (UTC)

[edit] Flip-flop circuit claim

According to articles in Wikipedia on Flip-flop (electronics) and William Eccles, the flip-flop circuit was invented and patented by Eccles and F.W. Jordan in 1919, not by Mauchly. If there is evidence that Mauchly invented it at Ursinus College, it should be provided here. (But Mauchly was at Ursinus from 1933 to 1941)

Mauchly never claimed he invented the flip-flop. He was well aware of the Eccles-Jordan circuit. He always claimed that it was the use of the flip-flop as a high-speed counter to count cosmic rays at Swarthmore that first inspired him to compute electronically. --Zebbie 20:00, 10 January 2006 (UTC)

[edit] ENIAC patent case

In his October 30, 1968 deposition in the Honeywell Inc. v. Sperry Rand Corp. case, Mauchly testified that he arrived at the concept of an electronic, general purpose computer while at Ursinus, and had discussed it with others, but could not recall the names of anyone to whom he had revealed this. He was able to produce no evidence of an interest in digital circuits prior to June of 1941, when he visited Atanasoff and examined the ABC Computer.

These quotes from the findings of Judge Larson (Oct. 19, 1973) in the ENIAC patent case constitute settled law:

• 3.1.17 – Prior to his visit to Ames, Iowa, Mauchly had been broadly interested in electrical analog calculating devices, but had not conceived an automatic electronic digital computer.
• 3.1.18 – As a result of this visit, the discussions of Mauchly with Atanasoff and Berry, the demonstrations, and the review of the manuscript, Mauchly derived from the ABC "the invention of the automatic electronic digital computer" claimed in the ENIAC patent.

This information can be found in Clark Mollenhoff's biography of Atanasoff. Mollenhoff was both a Pulitzer prize winning journalist and a lawyer.

This is not to argue that Mauchly made no advances on the work of Atanasoff, only that his claims of originality and priority in attempting to patent ENIAC were false. --Blainster 21:28, 24 August 2005 (UTC)

There is, however, a difference between something settled as far as a court is concerned and something settled as far as history is concerned. (Especially as the question now has no legal meaning, all patents having expired). The answer really depends on "what do you mean by invent?" and "what do you mean by computer?" There is no single inventor, but ENIAC is a much more important machine historically than ABC and was bigger contribution.--67.53.1.193 05:46, 11 September 2006 (UTC)

[edit] ABC computing speed

The calculations of the ABC were done with vacuum tube logic circuits, and thus were order of magnitude faster than electro-mechanical computers such as the much later Harvard Mark I. It did have a rotating drum memory system, which limited it to 30 operations per second. The one second per operation that anonymous 141.158.42.148 refers to, was the printout speed per page, not the calculation speed. --Blainster 00:14, 11 September 2005 (UTC)

I don't think it's valid to try to guess how fast one component of the machine would be if it were somehow used in another design. Arthur Burks, certainly an authority on it's workings, puts the ABC's speed much lower that 30 per second; I believe he thinks you must consider the time it takes to solve a problem, including the human handling of intermediate results on punch cards. 30 OPS is the "peak" rate: one second to do 30 simultaneous additions. --70.90.6.230 18:17, 20 January 2006 (UTC)

[edit] Programmability

Today's definition of computer programmability did not apply to either ABC or ENIAC. ABC was single purpose, and programming the ENIAC meant a few days of rewiring. Use of stored programs was not part of the ENIAC patent claim, even though it was programmable in the sense of being manually rewirable for general purpose use. However the prior design disclosures in Von Neumann's document apparently contributed to the weakness of Mauchly and Eckert's claim through no fault of their own. Ironically, Eckert's advances in memory design for the EDVAC (which served as a basis for Von Neumann's design) employed Atanasoff's ideas of binary operation and regenerative memory. --Blainster 22:58, 19 January 2006 (UTC)

I disagree with (Blainster 19-Jan) comments above and the edits that removed "programmability."

Defined by the Wikipedia: "A computer is a machine capable of undergoing complex calculations. The calculations proceed according to a program — a list of instructions."

The ENIAC was programmable and therefore a computer. "Rewiring" is probably a misleading term; the machine was designed to follow a sequence of instructions, and the sequence was defined by the positions of patchcords and switches. It was in every sense a program; just not a stored program. The Z3 and the Harvard Mark I are also considered programmable machines, although perhaps not "Turing-complete."

The ABC was just not programmable. It did one thing at a time, under operator control. Besides read and write punch cards and convert them from decimal to binary, the only operation it performed automatically was to add or subtract the same number repeatedly.

The point is that there is a specific definition of computer, but also a more general one. The general class of things that manipulate numbers will include the ABC, as well as including a desk calculator or a punch card tabulator. The specific computer must have calculations that "proceed according to a program." Within that class of computer we can then differentiate whether the program was hard-wired, set by switches, stored on paper tape, stored in the same memory as the data, etc.

But the ABC wasn't in that class. The complex steps needed to solve a series of simultaneous equations were not part of the ABC. Like a calculator, it required a human to choose and activate the next step. (In addition, the operations available were not general purpose, so one could argue that even as a calculator it was a special-purpose calculator.)

Programmability is absolutely fundamental to the definition of a computer - and that is why Judge Larson's bold statement has been questioned. I think it belongs in the text.

--70.90.6.230 18:03, 20 January 2006 (UTC)

1) I agree that one common understanding of the word computer today is that it includes reprogrammability. This definition is given in the Wikipedia article about computers. However, Wikipedia does not qualify as a recursive source for its own material, and I see no references for that information in the article. According to the Microsoft Computer Dictionary, third edition, 2000, a computer is any machine that does three things: accepts structured input, processes it according to prescribed rules, and produces the results as output. This published definition does not happen to include a requirement for reprogrammability, and it is a relatively recent definition. I expect you can find one that supports your position. The point is, it is not as clear-cut as you suggest. --Blainster 05:35, 14 August 2006 (UTC)

2) The definition of a computer has evolved since the early 1940s, when it typically referred to a person who computes. You are applying your 21st century definition to a mid-20th century machine. You are certainly free to do so, but you should not assume that the word means the same thing today as it did in 1946, or even in 1973 when Judge Larson made his decision. --Blainster 05:35, 14 August 2006 (UTC)

3) It happens that the ABC shares two characteristics with all modern computers, that ENIAC lacked. The ABC used binary logic and regenerative capacitor memory, so it is a more advanced design in that respect. However, ENIAC was a great leap in speed and general purpose design. No surprise, considering it was a vastly larger and more expensive project. Eckert & Mauchly deserve the credit for building a successful machine, and creating the computer industry. But that does not diminish the accomplishments of Atanasoff, without whom Mauchly might never have built a computer. --Blainster 05:35, 14 August 2006 (UTC)

4) Regarding my recent edit, of course the ABC did follow a sequence of computational steps, (or program). It's "program" was hardwired, so could not be changed. --Blainster 05:35, 14 August 2006 (UTC)