User talk:Softtest123

From Wikipedia, the free encyclopedia

Contents

[edit] Church-Turing thesis, history etc.

Yes, go for any additions and changes you want to make. The way I look at it is: nothing is really gone (as it resides in the history). There seem to be a few others with a lot of interest in this and other topics in theoretical computer science, in particular you may see edits (of yours and my stuff) by CBM (another PhD, I believe he's a recursion theorist) who has risen in the ranks of wikipedia to become a kind of super-editor. He is not shy about editing out stuff but he is always very polite and explains why he did it on the article's talk page (see the Recursion theory talk page for a recent example). wvbaileyWvbailey 16:49, 3 September 2007 (UTC)

Hi, wvbailey here. Martin Davis 1965 (a student of Post and Kleene -- he's still quite alive and kicking -- we've actually corresponded about the origin of phrase ("the Halting problem" (he is the originator)) is the editor of "The Undecidable: Basic Papers on Undecidable Propositions, Unsolvable Problems and Computable Functions". He assembled the papers from the various journals, had translated for him when necessary (e.g. Godel 1931), and then added commentary to the various articles of Godel, Church, Turing J.B. Rosser, Stephen C. Kleene and Emil Post. In the process he had contacts with those who where alive at the time (in particular: Godel). Some of his correspondence is almost as famous as the original articles themselves. So sometimes the article is the source of the quotation, and sometimes Davis's comentary is the source of the quotation, and sometimes Godel's (in particular) responses to Davis's queries is the source of the quations. Something similar happened to van Heijenoort when he was doing a similar task. Hope this helps: wvbaileyWvbailey 23:17, 3 September 2007 (UTC)
Hi, I've had a hard time with "The Undecidable" citations too. One way I've done them is something like this (which seems to me the best way, but the most laborious):
(Gödel in Davis 1965:24)
And perhaps when Davis is commenting, something like this:
"Dr. Gödel has stated in a letter that that he was, at the time of these lectures, not at all convinced that his concept of recursion comprised all possible recursions . . .."(Davis in Davis 1965:40)
I believe I've also done it this way -- put a note at the top of the page that "U" in the citations indicates reference to The Undecidable, so you could then do either of these (this method is maybe better when you have a bunch of co-authors or long names like (Whitehead and Russell (1910) in van Heijenoort 1967:216):
(Gödel in U:24), or more vaguely (U:24)
I suppose any of these are okay so long as the reader can figure out where to look up the quote. wvbaileyWvbailey 13:51, 4 September 2007 (UTC)

[edit] DTSS

Hi. You asked about, "the first timesharing system [] developed by a joint effort between General Electric and Dartmouth. I presume that would be the Thayer School of Engineering."

Dartmouth, but not specifically Thayer. Kemeny and Kurtz were professors in the math department:
"The Dartmouth Time-Sharing System, or DTSS for short, was the first large-scale time-sharing system to be implemented successfully. Its implementation began at Dartmouth College in 1963 by a student team under the direction of John Kemeny and Thomas Kurtz with the aim of providing easy access to computing facilities for all members of the college. By 1964 the system was in use where it remained so until the end of 1999. DTSS was originally implemented to run on a GE-200 series computer with a GE Datanet 30 as a terminal processor that also managed the 235."
—wwoods 16:23, 4 September 2007 (UTC)
Interesting. What the person wrote above is correct. And the history goes way back. With regards to George Stibitz (also see the Valley News article that I typed in on that article's talk page) I took a photo a couple weeks ago of the brass plaque in McNutt Hall at Dartmouth College. The plaque reads as follows:
"In this building on September 9, 1940, George Robert Stibitz then a mathematician with Bell Telephone Laboratories first demonstrated the remote operation of an electrical digital computer. Stibitz, who conceived the electrical digital computer in 1937 at Bell Laboratories, described his invention of the "Complex Number Calculator" at a meeting of the Mathematical Association of America held here. Members of the audience transmitted problems to the computer at Bell Labs in New York City, and in seconds received solutions transmitted from the computer to a teletypewriter in this hall."
What Stibitz invented (1936-7) was the binary (relay) adder. Turing must have heard about it when he was at Princeton and he built himself a binary multiplier (Davis 2000. Stibitz' use of a teletype was apparently the first instance of a "time sharing" use of a computer. By the time I arrived on the campus (1966) the folks who were funding this was ARPA or DARPA as it was known then -- the Defense Advanced Research Projects Agency. They had an on-staff retired general or colonel or something like that who was ensconced in the Kiewit Computation Center (now demolished). As I remember, the time sharing computer was in the basement of either in McNutt or a hall next door. About midway through fall of my freshman year (1966-67) the Kiewit Computation Center was finished and everything moved over there (I knew Kemeney but didn't have him for a prof). Two research/design projects I did when I was at Dartmouth and then Thayer School were funded by DARPA. Some of the Seven Sisters' colleges (I know for certain this included Mount Holyoke, and most likely Smith College) were also linked via teletype, as were all the local high schools. The historical connection between this and the Internet as we know it is unclear to me, but they are connected. wvbaileyWvbailey 18:19, 4 September 2007 (UTC)
It is interesting that the main article makes no reference to General Electric's contributions to that effort. Later GE commercialized that effort making it available through its Information Systems Division. in the late 70's, I designed the "CRD-8" microprogrammable synchronous communications controller that was key to operation of what was later known as the "backbone" of GE's timesharing network. It was capable of handling 4 voice grade lines up to 57 KB per line, but of course the voice grade lines in reality could only handle about 50 KB. GE, having sold its process control division to Honeywell, commissioned Honeywell to build data concentrators, MRC, and RC (Mini-remote concentor, remote concentrator.)
Too bad Wikipedia does not allow "original reseach" so that it could include what those of us that were there saw with our own eyes.
Softtest123 15:29, 5 September 2007 (UTC)
Adding facts that you happen to know from personal experience isn't a problem, as long as you don't add personal theories about them. Ideally, having some documentation to corroborate your statements would be good, but this stuff doesn't sound likely to be controversial. (Just don't dwell on the underappreciated importance of the CRD-8, or the brilliance of its designer.) If in doubt, tell your version on the talk page, and see if someone else can pick up the ball.
—wwoods 20:22, 5 September 2007 (UTC)
Thanks. I'll keep that in mind.
I'm not particularly interesting in promoting the CRD-8 or any of the professional work that I have done. I have, however, had occassion to observe quite a bit of computing history in the last 49 years. <grin>
21:31, 5 September 2007 (UTC)


I'm not sure the above is true. Example: I contacted Martin Davis directly to ask him he was the source of the phrase "the halting problem". I had traced the phrase to him via a proof in a book of his, a proof without citation. In an e-mail to me he confessed his crime. I pasted this into the talk page. But it was inadmissable in the court of wikipedia -- when I entered, in the halting problem article, a weasel-worded sentence to the effect that "Martin Davis claims to have coined the phrase" it was challenged and expunged (in my opinion: rightfully so; I was curious to see what I could get away with). Eventually someone found a footnote in a book that answered exactly the same question exactly the same way. I journeyed to the library, verified the quote in the book and entered it with a proper citation, and thus it remains evermore as a footnote in the article. I believe that the rule is this: if you didn't read it in a book or article that can be found by someone and fact-checked, it is not admissible in the article article. But it can sit on the talk page and maybe a reader will make a kindly/helpful suggestion (as happened in my example). Similar cases have happened more than once to me (e.g. busy beaver). The talk page proves to be very useful this way. wvbaileyWvbailey 21:03, 5 September 2007 (UTC)
Huh! We collided again! I'm learning how to merge these things in, though.
I'll bet the rule has to do with hearsay, or some such. And the controversial nature of it.Softtest123 21:31, 5 September 2007 (UTC)
Here's their official policy statement:
"The threshold for inclusion in Wikipedia is verifiability, not truth. "Verifiable" in this context means that any reader should be able to check that material added to Wikipedia has already been published by a reliable source. Editors should provide a reliable source for quotations and for any material that is challenged or is likely to be challenged, or it may be removed."
"It would seem to me that, according to this wiki-rule, the information must be "publically accessible", i.e. not in the inaccessible/private/hidden-away files of some company or some personal collection; rather, the info should be locatable in a public library or the equivalent. Don't get me wrong: I don't agree with everything that wikipedia promulgates. But I try to find "peer-reviewed" stuff, or real physical evidence (e.g. photograph of), and I assume that anything that is in a book or a magazine in a library is more-or-less (kind of, hopefully sort of) "peer-reviewed". A strict adherence to "Bill's rule" leads to misgivings about web-available papers of such folks as Gurevich at Microsoft; the papers are published internally at Microsoft but available to the public and thus some are not peer-reviewed. But sometimes I break the rules, esp. with regards to Gurevich. There are also problems when someone is pushing their very Non-Neutral Point Of View and citing weird sources/information as "truths" (e.g. Nazi propaganda -- 2 years ago I ran into this on the Hilbert article. The guy was finally blocked. He had been using "sock puppets" from public computer terminals and got caught. He started spouting vile stuff in German...'twas a horrible episode)). wvbaileyWvbailey 00:18, 6 September 2007 (UTC)
I think the operative word there is "may". First person knowledge "may" remain in an article if it adds to the article and is not "controversial". Just about anything can be controversial and if it is and cannot be verified, then it probably shouldn't be left in.
I just started reading a book about the origins of phrases and had to put it down because it was so badly written. Lots of references listed at the end of the book, but none cited. <tsk. tsk.>
Anyway, there is enough work on Wikipedia to keep me as busy as I want to be without writing anything new. If I write something new, I'll get it published, then cite it in Wikipedia. <ho ho ho>
Softtest123 04:47, 6 September 2007 (UTC)

[edit] An awesome job

I quickly re-read the Church-Turing thesis history. The amount of work you did was really awesome. One of these days I'll go through and check the quotes and citations. Because someone stole the last two volumes of Godel's complete works from the Dartmouth library I have not been able to verify some other important quotes that I'm aware of, so they are not in the article. Wow, what a nice job!

Another topic, I was reading your bio and observed that you are code-checking theorist. Here's another, related problem. Actually: pretty scary problem, there's even an international standard related to this (I can't remember the number, IEC 950?). The dangerous stuff we built (plasma cutters) had one little scary problem that we could not seem to work around -- eventually they had to start (!) Someone had to push a button. Scary thought: I saw once and only once (and so did someone else), 20 years ago, a plasma machine start up on its own. Some glitch somewhere, set it off all on its own. My first eye-witness of the aftermath of an industrial accident -- in a coal mine, a miner was sucked under a coal-digger machine when it started up on its own (some fool didn't turn off the main power -- the guy lost his legs, we were there nearby doing something else and helped in the rescue).

In the last couple years of my work I got involved in various litigations (had to testify in a Canadian lawsuit, for instance) that happened because of human errors. But the potential for machine-made errors always really bothered me. AT the very end, I was researching a scheme for a Motorola MC68HC05 micro to self-check by running, periodically, a computation (I had it runnning a tiny random number generator)... the point being that if something was wrong with the guts of the micro, it would produce bad numbers, and (somehow, there had to be a internal scheme to detect this, force a loop, or whatever, and ) an external watchdog equipped with a flip-flop would then catch the "hang" and disable the machine (we always used external watchdogs together with the internal one, at least we had that). But all this had to be done cleverly, so that the "hanging" would not be fooled .... Did you ever run into anything like that? I got to the point where I was simulating the entire micro on a spreadsheet -- I mocked up the entire micro and ran tests on it, forcing bits high or low and seeing what would happen. The only work in that direction that I know of, I saw on the web that at Stanford where E. J. McCluskey (we called him McFlip-Flop) had a team in the early 2000's working on this sort of thing. wvbaileyWvbailey 18:03, 1 October 2007 (UTC)

Not an uncommon problem when the power-on feature of a system is computer driven, i.e., the internal computer that detects the "on" switch being depressed and issues the "system on" command. There is a basic problem in the software development side of programmable systems development rooted in the culture of software development: "A hardware problem is not my problem." I will be writing more on this in the next few weeks on a different wiki and I'll let you know. This is not autonomic. I have designed and implemented an operating system for malfunctioning systems.
In the process control business it was common for us to lock up a computer when a system failure was detected and a "dead-man" timer would trigger and reactivate the processes on a standby machine. Somehow we seemed to have lost that technology.
Your technique works well when the system being monitored is designed to work with a monitor, but not otherwise. There is a fault injection technique that works quite well for determining how well a program works and responds to system failures. See Appendix G of my Ph.D. Dissertation.
Thanks for the kudos, but my work was just cleanup for your excellent research work. The article is quite long and probably could be decomposed some more. The desire is that articles be limited to 32k so this one is a little long. The rule is no longer hard and fast and having many sections may avert any size problems.
My work on this article does need checking and I could not resolve some of the "reprinted in" issues.
Perhaps you can get the books you need on interlibrary loan. If there is something you cannot get, let me me know what the issue is and I'll see if I can get the book and resolve the issue.
Softtest123 03:14, 2 October 2007 (UTC)

[edit] My Help Desk Questions

Getting rid of unwanted watchlist item.

Getting rid of unwanted watchlist item.

[edit] Watchlist

Did you ever fix that spurious item on your watchlist? I suggested on the help desk that you try at the village pump, but I noticed it seems to have been archived off of there by the robots without getting any comment. If the WP:SOURCE item is still there I'd suggest asking User:Brion Vibber. He's one of the main software devs and can check the watchlist table in the database to find out why it's there (and easily remove it). It may have been only a one-time database glitch that caused it, but the fact that you cannot remove it makes it sound like possibly a reproducible bug, so you could also submit it at Bugzilla, where the devs will see it immediately. • Anakin (talk) 02:02, 12 February 2008 (UTC)

Nope. It is still there and I still cannot delete it. I will report this to User:Brion Vibber and see what happens. --Softtest123 (talk) 15:14, 15 February 2008 (UTC)