The following is an archived discussion of a featured article nomination. Please do not modify it. Subsequent comments should be made on the article's talk page or in Wikipedia talk:Featured article candidates. No further edits should be made to this page.

The article was promoted by Karanacs 20:36, 28 September 2010 [1].

---

Manchester Mark 1[edit]

Manchester Mark 1 (edit | talk | history | protect | delete | links | watch | logs | views)

• Featured article candidates/Manchester Mark 1/archive1

Toolbox
Analysis Citation bot External links

Nominator(s): Malleus Fatuorum 19:31, 17 September 2010 (UTC)[reply]

This may be the only time you'll ever see a Cultural impact section in an article about an early valve-computer, but this one and its predecessor, both built at the end of the 1940s at Manchester University, caused quite a stir when the university's professor of neurosurgery delivered an address debunking the idea that computers could ever display intelligent behaviour. Oh, and it was also the prototype for the world's first commercially available stored-program computer, the Ferranti Mark 1. Malleus Fatuorum 19:31, 17 September 2010 (UTC)[reply]

• Driveby question (I'll read it properly later): I appreciate that it was scrapped, but does any part of it still exist? I suspect that will be many readers' second question (after "but what about ENIAC/Colossus/the Difference Engine?"—and yes, I know the answer to that one). These things have holy-relic status among a certain subset; I can easily imagine people making a special trip to see a couple of surviving valve-tubes if they exist. – iridescent 19:42, 17 September 2010 (UTC)[reply]
  • Not so far as I'm aware. It was made largely out scrap anyway, old GPO steel racks and war surplus valves. I seem to remember reading somewhere that the racks were returned to the GPO, but I may be imagining that. I suspect that the valves would have been recycled into the next development machine, Meg. Malleus Fatuorum 20:14, 17 September 2010 (UTC)[reply]

Comment. One dablink needs fixing: National Physical Laboratory. No problems with deadlinks. PL290 (talk) 20:12, 17 September 2010 (UTC)[reply]

Fixed. Malleus Fatuorum 20:17, 17 September 2010 (UTC)[reply]

• Comment: While I agree with the use of a non-free image in this case, the rationale needs to be expanded, author/publishing information needs to be added if known and the image itself should probably be reduced. J Milburn (talk) 00:37, 18 September 2010 (UTC)[reply]
  • I've scaled down the image size by 50% and rewritten the fair-use rationale. There's no information on the author that I've been able to find, but the copyright is claimed by the University of Manchester. Malleus Fatuorum 17:32, 18 September 2010 (UTC)[reply]
    • Much better, thanks. All the other images are fine, obviously. J Milburn (talk) 22:31, 18 September 2010 (UTC)[reply]

Support Comments - it's not a hangman! Still interesting, though - some nitpicks below. Nikkimaria (talk) 00:41, 18 September 2010 (UTC)[reply]

• "the British press, which used..." in lead vs "who used" in article body
• Is there a difference between a "stored-program computer" and an "electronic stored-program computer"?
• Manchester University or University of Manchester?
• "£35,000 per year (£950 thousand today)" - why use "thousand" for one and not the other?
• Some wikilinks (like millisecond and Ferranti Mark 1) are unnecessarily repeated
• "Main store consisted of" -> "The main store consisted of"?
• What does the term "seconded" mean? Perhaps explain/link?
• "Thirty-four patents" in lead vs "34 patents" in article body
• Is Leavitt 2006 or 2007?
• Lavington 1980 is not in Notes, nor is Williams 1997
• Be consistent in how volume and issue numbers are formatted in notes. For example, note 3 uses "2 42", and note 15 uses "21 (1)"
• Is Turing 1936 or 1937?
• Be consistent in including or not including publisher locations
• "Resurrection (The Computer Conservation Society)" or "Resurrection, The Bulletin of the Computer Conservation Society"

• Replies

Thanks for your comments.

• Standardised on "the British press, which".
• I suppose that in principle there could be stored-program computers that aren't electronic, but I'm not aware of any, or at least I can't immediately think of one, so I've removed "electronic".
• At the time, the university's official name was "Victoria University of Manchester", but it's commonly called Manchester University, and it seems a little stilted to refer to "the Victoria University of Manchester's Department of Neurosurgery" rather than "Manchester University's Department of Neurosurgery", for instance. I don't there's any ambiguity in the terms is there?
• Changed to £950,000.
• Duplicated wikilinks have been removed.
• I've stuck a "the" in front of "main store".
• "Seconded" is a common or garden everyday term isn't it? It means a temporary transfer of employer.
• It had to be "Thirty-four" in the lead as that's at the beginning of the sentence.
• It's Leavitt 2007, now fixed.
• I've removed Lavington 1980 and Williams 1997 from the Notes.
• The paper was submitted and published in 1936, in the 1936–37 issue of the journal. Hopefully the citation now makes that clear.
• Removed all publisher locations.
• I've standardised on Resurrection (The Bulletin of the Computer Conservation Society).

Malleus Fatuorum 01:03, 18 September 2010 (UTC)[reply]

• Only that the article uses both "University of Manchester" and "Manchester University", and it's not clear that those are one and the same

• I don't feel strongly about it, so I've changed all the "Manchester University"s to "University of Manchester". Malleus Fatuorum 12:20, 18 September 2010 (UTC)[reply]

• Perhaps it's a British term? I've only heard it used as "supported", as in "He seconded the motion". Nikkimaria (talk) 01:25, 18 September 2010 (UTC)[reply]

• Perhaps, I'm not sure. Anyway, if it's unfamiliar to you in this context then I guess it may be unfamiliar to others from the colonies as well, so I've changed it to "temporarily transferred". Malleus Fatuorum 12:09, 18 September 2010 (UTC)[reply]

Leaning to support on prose and comprehensiveness. Some comments, many of them minor:

• We learn in the lead about the world's first stored-program computer. That term then first appears in the first section's second paragraph; meanwhile, its first paragraph has discussed Turing's model and von Neumann's architecture definition. That first paragraph should explicate the extent to which the term is applicable to Turing's model and von Neumann's architecture (i.e., even if the term was not then in use, the reader needs to be told explicitly whether those embody the same concept and hence are what developed into the stored-program computer).
• The wrong expectation is produced (in me at least) by three section titles. Looking at the TOC, we have Programming, First programs and Later developments; to me (in the context of a Mark 1 article, as opposed to a History of computing article) this implies a continuing focus on work done with the Mark 1. The article structure would be made clearer with something like, for the sake of illustrating the point, Software architecture, Programs written and After the Mark 1
• "In July 1949 IBM invited Williams to the United States on all expenses paid trip" - missing word (on an); also, the compound adjective needs hyphenating (an all-expenses-paid trip).
• "The most significant design legacy of the Manchester Mark 1 was perhaps its incorporation of index registers" - needs attributing to Lavington, assuming he indeed speculates thus.
• In a couple of places there is what is, to me, a rogue word "for": *"which allowed for 1,024 (2¹⁰) different instructions" ("allowed 1,024"), and "eight 5-bit teleprinter characters were required to encode for each word" ("to encode each word")
• "Because the Mark 1 had a 40-bit word size" - for consistency, word length, wikified earlier, would be a better term.
• "it was able to carry out the necessary decimal to binary and binary to decimal conversions" - needs hypnenation (the necessary decimal-to-binary and binary-to-decimal conversions)
• "The only system software on the Mark 1 was a few basic routines for input and output; it had no operating system" - recast/rewikify to avoid leaving the reader wondering what "system software" is at the start of the sentence.
• "rather than the more conventional "00001"" - "now conventional" would be clearer
• The Cultural impact section has a "main article" hatnote to History of artificial intelligence; linked article does not discuss the cultural impact of the Mark 1, and in any case the cultural impact of an early computer is wider than that, so this should be a "see also" hatnote.
• On a related note, given that the Ferranti Mark 1 was the world's first commercially available general-purpose computer, the significance of that impact in terms of popular computing today needs to be stated. In addition, there are surely applicable See also links of that kind that should be added.
• "Not until a machine can [...] be angry or miserable when it cannot get what it wants" - as an aside, one is forced to wonder whether Jefferson foresaw that the machine would, however, have the ability to intensify said propensity in its users
• "The Times newspaper reported" - we know it's a newspaper (though the first occurrence needs wikifying)

I look forward to supporting this nomination once these points are addressed. PL290 (talk) 09:31, 18 September 2010 (UTC)[reply]

• Replies

Thanks for taking a look.

• I've made a small change to hopefully better explain the relationship of the stored-program computer architecture to the Turing machine and von Neumann architecture.[2]
• The article's structure works for me. The term Software architecture has a very specific meaning that's nothing to do with how the Mark 1 (or any other computer is programmed), which is the purpose of the Programming section. First program describes, well, the first programs written for the machine, which was in use for some time; who knows what other programs were written for it? Later developments is I think better than After the Mark 1, because to a large extent the Ferranti Mark 1's development was in parallel with the Manchester Mark 1's, it didn't happen after it. Similarly, IBM's licensing of the Williams tube and the decision to build Meg were directly influenced by the Manchester team's experience with building the Mark 1.
• "all expenses paid" changed to "an all-expenses-paid".[3]
• "The most significant design legacy of the Manchester Mark 1 was perhaps its incorporation of index registers" - needs attributing to Lavington, assuming he indeed speculates thus." The statement is implicitly attributed to Lavington as indicated by the citation at the end of the sentence, but I'd certainly prefer not to explicitly attribute it to him as it's a widely held and uncontroversial view. It could easily be attributed to dozens of other computer scientists, the sense of which would be lost if only one of them was chosen almost at random. I think it's just about as close to general knowledge as you could get in this field.
• I agree with you about "encode for", which I've changed to "encode", but I think there's a subtle distinction between "allowed for" and "allowed". The unadorned "allowed" carries the implication of permission having been granted, whereas "allowed for" implies that the possibility was taken into account. Not certain I've explained that very well, but I'm pretty sure that "allowed for" is correct.
• I've changed "word size" to "word length".
• The purpose of hyphens is to resolve ambiguity, but I don't think there's any plausible ambiguity in "decimal to binary conversion", for instance. The literature itself is inconsistent in whether or not hyphens are needed,[4] so on balance I'm against them here.
• I've recast "The only system software on the Mark 1 was a few basic routines for input and output; it had no operating system" to "The Mark 1 had no operating system; its only system software was a few basic routines for input and output".
• "rather than the more conventional '00001='". It's always been the mathematical convention that increasing powers progress from right to left. In fact Turing confused many of the listeners to his presentations by neglecting to explain that the conventional binary representation was reversed in the Mark 1. Remember that the modern binary system was developed by Leibniz in the 17th century, predating the Mark 1 by 300 years.
• I've changed the "main" hatnote to a "see also".
• This is an article about the Manchester Mark 1, not the Ferranti Mark 1. Although the Manchester machine was a prototype for the Ferranti the two were very different machines, and nobody imagined then that computers would become as pervasive as they are today. A contemporary account published in Popular Mechanics said: "While a calculator on the ENIAC is equipped with 10000 vacuum tubes and weighs 30 tons, computers of the future may have only 1000 vacuum tubes and weigh only 1.5 tons", As recently as 1977 Ken Olsen of DEC was quoted as saying that "There is no reason why anyone would want a computer in the home", so I think we can safely say that the Manchester Mark 1 had zero effect on popular computing.
• I've added a link to the first occurrence of The Times and removed the word "newspaper", although I fully expect that someone else will demand that it's put back.

Malleus Fatuorum 11:49, 18 September 2010 (UTC)[reply]

Responses to those responses: most of my points have been addressed to my satisfaction.

• I remain unconvinced about those section titles; the alternatives I suggested were quickly picked out of the air merely to try and illustrate the sort of contrast I think would help, but perhaps they failed to do that. If you still think what's there can't be bettered, it's not a showstopper for me.
• On "The most significant design legacy of the Manchester Mark 1 was perhaps ..." being just about as close to general knowledge as you could get: in that case, my issue is with "was perhaps". It would be preferable to say "is generally considered to be ...".
• I would say "allow for" has connotations of "take into account" rather than "make possible". But it's a very minor point and I will leave it with you.
• On ""rather than the more conventional '00001", in that case my issue is with the immediately preceding "in contrast to the modern convention". From what you say, it should be "in contrast to the established convention".
• Point taken about lack of direct effect on popular computing, meaning it would be inappropriate to devote prose to much later developments; would not some See also links on subsequent developments in computing nevertheless be appropriate? Currently there are none.

PL290 (talk) 20:02, 18 September 2010 (UTC)[reply]

Responses to those responses:

• I won't be changing the section titles.
• See also links are generally a sign that something's been inadequately covered in the article itself by not being linked. If you can point to any such subject then no doubt a See also section could be included.

Malleus Fatuorum 21:01, 18 September 2010 (UTC)[reply]

Response to those responses: my unresolved niggles are not actionable per FA criteria and I am happy to now Support. I leave you with an unaddressed couple of those unresolved, unactionable niggles:

• The second "was perhaps" needs the same treatment
• "Modern" remains misleading: reading about an old computer, we first learn that "there was no separate program stored in memory, as in a modern computer" (so far so good); we then learn that the machine's storage was arranged with the least significant digits to the left "in contrast to the modern convention". The latter unnecessarily leads the lay reader to presume that, like the former, the "modern" state of affairs only developed later. But, as you reply above, "the modern binary system was developed by Leibniz in the 17th century, predating the Mark 1 by 300 years." Hence it would be preferable to state, "in contrast to the established convention" or perhaps even "in contrast to the convention that had been established for 300 years".

PL290 (talk)

• You've persuaded me about the binary representation issue, so I've changed that to "in contrast to the established mathematical convention". Malleus Fatuorum 11:59, 19 September 2010 (UTC)[reply]

Sources issues: None. All sources OK Brianboulton (talk) 23:14, 19 September 2010 (UTC)[reply]

• Image copyright review: File:Von Neumann architecture.svg is a derivative work of an image without a valid source. Other images OK, but oppose pending resolution or removal of this one. Stifle (talk) 20:36, 20 September 2010 (UTC)[reply]
  • Would you mind translating that into English? Malleus Fatuorum 20:40, 20 September 2010 (UTC)[reply]
    • File:Von Neumann architecture.svg's description page indicates that it is adapted from File:Von Neumann architecture.png. That file has been tagged on Commons as lacking source information: "This media file is missing essential source information. The author and source of the file must be given, so that others can verify the copyright status. Unless the source is given, the file can be speedily deleted seven days after this template was added and the uploader was notified: (20 September 2010)." Not to put words in Stifle's typing fingers, but I think he wants you to deal with that tag. Nikkimaria (talk) 20:52, 20 September 2010 (UTC)[reply]
      • Jeez. And while I'm doing that perhaps he'd like me to do his ironing as well. Malleus Fatuorum 21:07, 20 September 2010 (UTC)[reply]
        • I'm sure he'd appreciate it! And while you're doing housekeeping, refs 17 and 19 are identical. Cheers, Nikkimaria (talk) 21:12, 20 September 2010 (UTC)[reply]
          • Well then clearly I made a big mistake in bringing this under-prepared article to FAC. Malleus Fatuorum 21:18, 20 September 2010 (UTC)[reply]
            • Obviously, seeing as it doesn't have universal support ;) Oh, and about your last edit summary: don't worry, it's a bit early in the season for raking yet. Nikkimaria (talk) 23:14, 20 September 2010 (UTC)[reply]
              • I can't be arsed to argue the toss and I can't be arsed to make a new version of it, so the image is now deleted. Is the article better or worse as a result? Who can tell. Malleus Fatuorum 23:40, 20 September 2010 (UTC)[reply]

• Support. Wonderfully written article about (what I find to be) an extremely interesting topic. Quick comment, however: is "The resulting machine, known as Meg, ran its first program in May 1954." really important enough to merit a mention in the lead? "In 1951 they started on the development of a successor to the Mark 1, which would include a floating point unit." suffices, I think. ɳOCTURNEɳOIR^♯♭ 15:44, 27 September 2010 (UTC)[reply]
  • Thanks. It seems to me that dropping that last sentence of the lead leaves the question of what was the Mark 1's successor dangling, but I don't feel particularly strongly about it. Malleus Fatuorum 18:41, 27 September 2010 (UTC)[reply]
    • I think my beef is mostly with the mention of the date, which I view as largely irrelevant to the subject. I don't feel particularly strongly about the matter either, so it's up to you at the end of the day. ɳOCTURNEɳOIR^♯♭ 19:11, 27 September 2010 (UTC)[reply]
      • That's a fair point. I've dropped the date and combined the last two sentences of the lead. Malleus Fatuorum 19:32, 27 September 2010 (UTC)[reply]
        • Looks good. Full support as previously noted. ɳOCTURNEɳOIR^♯♭ 20:53, 27 September 2010 (UTC)[reply]

• Support; I can't see any obvious omissions or anything I'd change. One minor point (and I appreciate there may be no answer); did they have any particular task in mind for it when it was designed (codebreaking, ballistics, accounting etc) or was it purely a proof-of-concept device built in the hope that someone would later find a commercial use for it? – iridescent 21:15, 27 September 2010 (UTC)[reply]
  • The answer's in the lead: "The Mark 1 was initially developed to provide a computing resource within the university, to allow researchers to gain experience in the practical use of computers". Basically I don't think anyone was really sure what general-purpose computers would be useful for back then. Malleus Fatuorum 21:13, 28 September 2010 (UTC)[reply]
    • It may (emphasis may) be worth spelling that out explicitly. The likely audience for this article will presumably be people with a basic background knowledge of Turing's wartime proto-computers, who'll wonder if the military had their eyes on it. – iridescent 21:34, 28 September 2010 (UTC)[reply]
      • I'll give that some thought. The project was really just a few academics interested in trying to find out what computers might be useful for. A few of them had of course been involved in the development of radar and machines like Colossus, but that was then still secret, and Churchill had ordered that the machines be broken up. Certainly the military had no input into the development and no obvious interest in it. Malleus Fatuorum 22:10, 28 September 2010 (UTC)[reply]

The above discussion is preserved as an archive. Please do not modify it. No further edits should be made to this page.