April 2012 - Members - lists.sigcis.org

New CHM video: Bell Labs
by Dag Spicer 10 Apr '12

10 Apr '12

Friends, CHM has posted a lecture video from our recent program: The Idea Factory: Bell Labs and Great Age of American Innovation with New York Times Magazine writer John Gertner. Watch the program here: http://www.youtube.com/watch?v=e90sw6-qxmE&feature=digest_sat Info on the lecture: [Recorded: March 28, 2012] Bell Laboratories, which thrived from the 1920s to the 1980s, was the most innovative and productive institution of the twentieth century. Long before America's brightest scientific minds began migrating west to Silicon Valley, they flocked to this sylvan campus in the New Jersey suburbs built and funded by AT&T. At its peak, Bell Labs employed nearly fifteen thousand people, twelve hundred of whom had PhDs. Thirteen would go on to win Nobel prizes. It was a citadel of science and scholarship as well as a hotbed of creative thinking. It was, in effect, a factory of ideas whose workings have remained largely hidden until now. New York Times Magazine writer Jon Gertner unveils the unique magic of Bell Labs through the eyes and actions of its scientists. These ingenious, often eccentric men would become revolutionaries, and sometimes legends, whether for inventing radio astronomy in their spare time (and on the company's dime), riding unicycles through the corridors, or pioneering the principles that propel today's technology. In these pages, we learn how radar came to be, and lasers, transistors, satellites, mobile phones, and much more. Even more important, Gertner reveals the forces that set off this explosion of creativity. Bell Labs combined the best aspects of the academic and corporate worlds, hiring the brightest and usually the youngest minds, creating a culture and even an architecture that forced employees in different fields to work together, in virtually complete intellectual freedom, with little pressure to create moneymaking innovations. In Gertner's portrait, we come to understand why both researchers and business leaders look to Bell Labs as a model and long to incorporate its magic into their own work. Join author Jon Gertner for a fascinating conversation with KQED's Dave Iverson about the people and history of Bell Labs, and the ways it fostered a culture of innovation and ideas. Best wishes, Dag

1 0

Message from Atsushi Akera re Stored Program
by Thomas Haigh 06 Apr '12

06 Apr '12

[Looks like the list is having a bad week - there's some kind of issue with our Go Daddy domain registration that sometimes interacts with certain ISPs to bounce mail with a message from "mailstore1.secureserver.net" that the address is rejected. If anyone has expertise in this area and can help us solve the problem that would be appreciated. In that case the messages never make it to our listserv manager. Anyway, Atsushi asked me to forward this - see below for my reply]. From: Atsushi Akera [mailto:akeraa@rpi.edu] Sent: Wednesday, April 04, 2012 5:47 AM To: 'David Alan Grier'; 'Thomas Haigh' Cc: members(a)sigcis.org Subject: RE: [SIGCIS-Members] "Stored program" -- anyone know origins of the PHRASE Hi David, Hi Tom, Just in this context, the term "program" is used extensively in the context of ENIAC. It's been a while since I've gone through the records, and I didn't go back to check it. But as I recall, Mauchly was among those who brought the terminology of 'program' based on his amateur radio experience. I assume this was referring in general terms to a sequence of activities, similar to a conference program or a program for an organized luncheon. You would know this part better, but I think the other conventional term from the period would have been "plan of calculation," based on the work of human computers. The reference to "instruction tables" (in considering Paul's post) strikes me as something that could come out of the fact that there was contemplation of using the ENIAC function tables to store instructions. (This would require tracking the dates & influence, however. It's just a guess for now) Meanwhile the term "storage" would have of course gone back to Babbage (mill and store), but gained significant currency once everyone developed an interest in mercury delay lines as a possible storage device. What's perhaps interesting in all this is that technological and material artifacts, as manifestations of both human and machine computation practices, were heavily influencing the terminology people used to understand the early process of computation. New language of course has to be assembled in order to sort through and understand new technological phenomena (electronic computing), and we see during this period various efforts to combine past terminology into a workable description of the new technology-followed by some process whereby consensus emerges around those terms. I think it's useful to think of "stored program" in that context-and when that particular articulation became important for the practitioners. As I recall, stored programs weren't all that important in the early years. Von Neumann's single bit used to designate instruction from data was more from the point of view of efficient storage of the two types of information, not from any notions about stored programs related to computabilitiy. Even when the term came into existence, it may not have had this kind of reference either. perhaps worth keeping in mind. Best wishes! - Atsushi From: Thomas Haigh [mailto:thomas.haigh@gmail.com] On Behalf Of Thomas Haigh Sent: Wednesday, April 04, 2012 6:39 PM To: 'Atsushi Akera' Subject: RE: [SIGCIS-Members] "Stored program" -- anyone know origins of the PHRASE Thanks Atsushi, I think von Neumann might have felt the need to flag data vs. code as he planned delay line storage at this point and was concerned with interleaving them for efficient access. Possibly he recognized the danger of accidentally overwriting code, and saw this as a kind of memory protection feature as later implemented on multitasking systems to protect programs from each other. Although the general organization of the 1945 EDVAC instruction set does not betray a concern with the convenience of programming. The point about Babbage and the "store" is something I'd been wondering about myself, and it's interesting to see that you have the same idea. As I look more at the 1945-48 documents, it does seem that the idea of storing programs and data in the same memory was quickly recognized as one of the key advantages of EDVAC-type machines but mostly on the grounds of efficiency, simplicity and flexibility rather than the potential for self modification. E.g. Eckert in the Moore School lectures volume says the big advantage is that storage can be allocated as needed for a particular application. After all, ENIAC has already shown that you could do things like conditional branching and indexing through an array without needing to modify stored program code. I agree with the bigger point about the importance of categories and terminology. It's interesting that with all the endless discussion of "stored program" nobody ever seems to have tried to find out where the term comes from. Tom From: Atsushi Akera [mailto:akeraa@rpi.edu] Sent: Thursday, April 05, 2012 12:35 PM To: 'Thomas Haigh' Subject: RE: [SIGCIS-Members] "Stored program" -- anyone know origins of the PHRASE Agreed. By the way, I tried to send the message to the group, but it bounced. Not sure why. Please feel free to forward if you can. Also keep in mind that with circulating storage, there's no way to slow down what's coming out of the memory. So the instruction/data bit can be used as an efficient way to implement (in engineering) where information goes in a machine. - Atsushi (p.s. if you encounter problems with RPI's spam filter, please send the message to my alternate email account: <mailto:atsushi_akera@hotmail.com> atsushi_akera(a)hotmail.com, preferably with a quick note to this account indicating that you sent a message there since otherwise I don't check that account regularly. - Thanks!) _________________________________________________________ Atsushi Akera Associate Professor, Department of Science and Technology Studies Director, First Year Studies Program--Sage 5206 Rensselaer Polytechnic institute 110 8th Street Troy, NY 12180 USA ph: 518.279.9708/fx:518.276-2659/e:akeraa@rpi.edu /w: http://www.rpi.edu/~akeraa

2 1

Message from Len Shustek re Stored Program
by Thomas Haigh 04 Apr '12

04 Apr '12

[Len is also having posting problems with the list. While I'm here, to address his final point, self modification of code goes much earlier than Wilkes, Wheeler & Gill. The instruction set described original 1945 EDVAC Report would have relied on code modification during execution for looping, conditional branching, or indexing through a matrix. However this capability was confined to address modification rather than unrestricted instruction modification. This restriction was relaxed in subsequent work by von Neumann's group and, as mentioned earlier, in Turing's 1946 description of the ACE design. Designers of early stored program computers added conditional branch instructions and indirect addressing capabilities such as Manchester's famous index register to sharply reduce the need for self modifying code. There's a clear description of this in Priestley's Science of Operations, particularly pages 164- 172.] Tom, I'm not sure why the first appearance of the precise bigram "stored program" is so significant. There were certainly very early near-misses, such as this from the 1947 Burks, Goldstine, and von Neumann "Preliminary Discussion" paper: "It is evident that the machine must be capable of storing in some manner not only the digital information needed... but also the instructions which govern the actual routine to be performed. ... Hence there must be some organ capable of storing these program orders." Even earlier is this from the 1945 "Description of the ENIAC and Comments on Electronic Digital Computing Machines". It is farther from the precise linguistic target, but exactly on point conceptually: "This requirement as to instruction speed can be met in a direct and simple fashion by storing instructions (in numerical code, so to speak) in memory elements or registers of the same type as those used to store the numerical data." If your focus is on the ability of programs to modify themselves, I don't know whether there is a reference earlier than the one in Wilkes, Wheeler, and Gill that Allan Olley pointed out. The significance of that ability is interesting to ponder. Doron Swade has done some research in this area, and in his article in the December 2009 "EDSAC at 60" issue of The Computer Journal he says of the expert opinions he received, "whether the primary benefit was one of principle or practice was frustratingly blurred." -- Len Shustek

1 0

Turing at Princeton
by Drucker, Thomas L. 03 Apr '12

03 Apr '12

The current (4 April 2012) issue of the Princeton Alumni Weekly includes two brief articles of interest. W. Barksdale Maynard has an article 'Daybreak of the Digital Age' on Turing at Princeton and Jon R. Edwards (an undergraduate classmate of mine) has a note entitled 'Before Turing, There Was Veblen'. In addition, there's an advertisement for three events: (a) an exhibition at the University Library featuring Turing's dissertation and graduate file, (b) Andrew Hodges' upcoming lecture on St. George's Day, and (c) the conference on Turing's contributions to take place May 10-12. Further information is promised at http://www.princeton.edu/turing/. Thomas Drucker University of Wisconsin-Whitewater druckert(a)uww.edu

1 0

Reply from Paul McJones re Turing's ACE proposal
by Thomas Haigh 03 Apr '12

03 Apr '12

(Forwarded, as the list bounced his message.) From: Paul McJones [mailto:paul@mcjones.org] Sent: Tuesday, April 03, 2012 11:21 AM To: Brian Randell; Thomas Haigh Cc: members(a)sigcis.org Subject: Re: [SIGCIS-Members] "Stored program" -- anyone know origins of the PHRASE This is to confirm that Turing's 1945/6 ACE proposal describes the ideas of stored programs (starting on the very first page) but does not use that terminology. Turing does use the word "storage" extensively, but he refers to "instruction tables" rather than "programs". He demonstrates the use of stored programs with an instruction table called INDEXIN that simulates what we would call an indexed load. He also mentions that various aspects of program preparation (e.g., relocating subroutines) could be done within the machine itself, but for pragmatic reasons suggest it be done externally using punched card machines. For those interested, a scan of the 1972 NPL reprint is available here: http://www.npl.co.uk/turing/pilot-ace-proposal and a scan of the original 1946 report is here: http://www.turingarchive.org/browse.php/C/32 For anyone interested in studying Turing's report I recommend this paper: The other Turing machine B. E. Carpenter and R. W. Doran Computer Journal, Volume 20, Issue 3, Pp. 269-279. http://comjnl.oxfordjournals.org/content/20/3/269.abstract (I learned of the paper by Carpenter and Doran in Brian Randell's The Origins of Digital Computers: Selected Papers.) Paul On Apr 2, 2012, at 1:44 PM, Brian Randell wrote: Hi Tom: Not directly responding to your query, and commenting from memory, rather than having rechecked my sources, my recollection is that the 1945 First Draft of a Report on the EDVAC did not allow operations on stored instructions, but Turing's 1945 proposal for ACE did - to my mind a necessary final step towards the generality that I associate with the stored program computer concept. Cheers Brian On 2 Apr 2012, at 21:34, Thomas Haigh wrote: Hello everyone, I have a query related to a project I am working on concerning the conversion of ENIAC to stored program control in 1948, initially to run the first computerized Monte Carlo calculations. All this took place prior to the first operation of the Manchester Baby. That makes the question of what one means by "stored program" a very interesting one. This question was much discussed in the early days of the history of computing (1970s, early 1980s). I am starting to dig back into primary sources for early use of the phrases "stored program" and "stored program concept" to get a better idea of how these terms were used in the 1940s/early 1950s and what people thought they meant at the time. To clarify, almost everyone who has written about this cites the 1945 "First Draft of a Report on the EDVAC" as the initial dissemination of the stored program concept although there has been considerable debate as to the source of the ideas contained therein. However that document does not contain the phase "stored program." Or indeed use the word "program" in the body of the text. Or, remarkably, "EDVAC." "Stored" shows up a few times, though less frequently than "remembered." So, ignoring for the moment the relationship of the report to later definitions of the concept, we can agree that it was not the source of the phrase. The most obvious summary of the idea using the report's own terminology would be "remembered instruction device" rather than "stored program computer." I had thought about the 1946 Moore School lectures as a possible vector for the phase "stored program" as well as the concept. The phrase shows up many times in the Moore School lectures book but so far I have spotted it only in the 1980s editorial material rather than in the original lecture summaries. By 1954 "stored program computer" is showing up without explanation or citation required in the description of the IBM 650 published in the inaugural issue of Journal of the ACM. It is not particularly common in the ACM DL material for the rest of the decade ("automatic computer" and "digital computer" are more prevalent) but continues to pop up occasionally. The best the OED can do is 1957, which is even later. So, any thoughts on who came up with this phrase and when? I'm planning to dig deeper in search of early usage, for example into the 1950 "High-Speed Computing Devices" ERA book and some of the other CBI reprints from the 1940s, but it occurred to me that someone on the list might already know the answer to the question. Tom _______________________________________________ This email is relayed from members(a)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 -- School of Computing Science, Newcastle University, Newcastle upon Tyne, NE1 7RU, UK EMAIL = Brian.Randell(a)ncl.ac.uk PHONE = +44 191 222 7923 FAX = +44 191 222 8232 URL = http://www.cs.ncl.ac.uk/people/brian.randell _______________________________________________ This email is relayed from members(a)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

1 0

Re: [SIGCIS-Members] "Stored program" -- anyone know origins of the PHRASE
by Evan Koblentz 03 Apr '12

03 Apr '12

>From Pres Eckert's Tokyo speech (thanks to Thomas Miller for observing this): "My big idea was the idea of the stored instruction sequence or program, using a single fast memory for both data and instruction, with no distinction between registers used for many purposes. At the time I first though of this idea, (in January 1944)..."

3 2

"Stored program" -- anyone know origins of the PHRASE
by Thomas Haigh 03 Apr '12

03 Apr '12

Hello everyone, I have a query related to a project I am working on concerning the conversion of ENIAC to stored program control in 1948, initially to run the first computerized Monte Carlo calculations. All this took place prior to the first operation of the Manchester Baby. That makes the question of what one means by "stored program" a very interesting one. This question was much discussed in the early days of the history of computing (1970s, early 1980s). I am starting to dig back into primary sources for early use of the phrases "stored program" and "stored program concept" to get a better idea of how these terms were used in the 1940s/early 1950s and what people thought they meant at the time. To clarify, almost everyone who has written about this cites the 1945 "First Draft of a Report on the EDVAC" as the initial dissemination of the stored program concept although there has been considerable debate as to the source of the ideas contained therein. However that document does not contain the phase "stored program." Or indeed use the word "program" in the body of the text. Or, remarkably, "EDVAC." "Stored" shows up a few times, though less frequently than "remembered." So, ignoring for the moment the relationship of the report to later definitions of the concept, we can agree that it was not the source of the phrase. The most obvious summary of the idea using the report's own terminology would be "remembered instruction device" rather than "stored program computer." I had thought about the 1946 Moore School lectures as a possible vector for the phase "stored program" as well as the concept. The phrase shows up many times in the Moore School lectures book but so far I have spotted it only in the 1980s editorial material rather than in the original lecture summaries. By 1954 "stored program computer" is showing up without explanation or citation required in the description of the IBM 650 published in the inaugural issue of Journal of the ACM. It is not particularly common in the ACM DL material for the rest of the decade ("automatic computer" and "digital computer" are more prevalent) but continues to pop up occasionally. The best the OED can do is 1957, which is even later. So, any thoughts on who came up with this phrase and when? I'm planning to dig deeper in search of early usage, for example into the 1950 "High-Speed Computing Devices" ERA book and some of the other CBI reprints from the 1940s, but it occurred to me that someone on the list might already know the answer to the question. Tom

4 4