[SIGCIS-Members] Semi-rare John Mauchly document

[PeterEckstein at comcast.net]

Evan, 



Below I am repeating my comment but using the Reply All option to include others. I have since read the Rosenberg summary cited by McMillan, and I think that the part that attempts to counter the myth is a fair--but far too short--summary.  I did interview Goldstine over the phone one time, but I do not offhand remember if we covered this issue.  I hope I transcribed my notes on that interview; if so, I would be happy to send those, too, to anyone working on this issue seriously. I would also add my agreement with JM's point that the btrief discussion in the Metropolis "Trilogy of Errors" article was far closer to the mark than most other histories.  As I recall, Scott McCartney's more recent book dealt with the issue pretty fairly, too. 



Peter Eckstein 



Evan, 



I am a little rusty on this story, but the oral histories that Nany Stern created--even better than her book itself--make it clear that the specific details of the stored program concept emerged from (roughly) weekly seminars that included JPE & JM, a couple of Moore School professors including Carl Chambers and Arthur Burks, and John von Neuman. Don't remember about Goldstine. JVN. wrote up the results of their joint work, to which he, like others, had made substantive contributions, and he did it in a consistent notation system of his own devising, and the typescript became the "First Draft" that was immortalized. Pres Eckert to the end resented that he and JM were working under top secret wartime conditions, but that JVN felt free to make speeches about the coming of the computer age--which created the impression (without saying) that he was its chief inventor--and that Goldstine's distribution of the First Draft with only JVN's name on it was a great injustice.   Some of the popular histories build on this to make the fantastic sugestion that the engineers who built ENIAC were too stupid to realize that it would need a stored program and that it took JVN to figure this out.  But the need to freeze the design and build a machine to help win the war was what prevented the stored program from being developed for ENIAC. This is a little more complicated story than the summary that Bill Mauchly offers but seems consistent with the John Mauchly letter. 



I'm not convinced that this story has ever been done full justice. Most (but unfortunately not all) of the Stern inerviews are in the Babbage collection. My own interviews with JPE and, more briefly, with AB include discussions of this.  I could make those portions available to anyone seeking to tell the accurate story. 



Peter Eckstein 
----- Original Message ----- 
From: evan at snarc.net 
To: members at sigcis.org 
Sent: Thursday, July 30, 2009 6:00:32 PM GMT -05:00 US/Canada Eastern 
Subject: [SIGCIS-Members] Semi-rare John Mauchly document 

Hi all. I received this document from Bill Mauchly today. I asked him if it's okay to share freely; he said yes. File is attached. Bill added: "In 1979, just a few months before he died, John Mauchly had a letter published in DATAMATION.  Examples of his writing are rare, but here he clearly wanted to have his say.  In it he laid out very clearly how he and Eckert, in the wee hours of 1944, worked out the stored-program architecture of EDVAC, the successor to ENIAC.  Later they told von Neumann, and he published it as his own work.  He also talks about BINAC.  This was at the time that Burks and Goldstine were trying to drain as much credit away from Eckert and Mauchly and towards von Neumann as they could.I've attached a copy of the letter; it a good read if you like this sort of thing.  This short essay does not seem to be available on the web.  I typed it in from a copy I had that was among my mother's things, apparently a draft from his TRS-80 word processor, on the prompting of Jean Bartik.  Enjoy.Bill Mauchly" 
_______________________________________________ This email is relayed from members at sigcis.org, the email discussion list of SHOT SIGCIS. The list archives are at http://sigcis.org/pipermail/members/ and you can change your subscription options at http://sigcis.org/mailman/listinfo/members
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.sigcis.org/pipermail/members-sigcis.org/attachments/20090801/92ed39b3/attachment-0001.htm>