Talk:Computer/Archive 3

This is an archive of past discussions about Computer. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 1

Archive 2

Archive 3

Archive 4

Archive 5

Thoughts on Lovelace's Contribution

IMHO, I think that it needs to be added that not only was Babbage responsible, but Lovelace also (Even if she was Babbage's figurehead, it cannot be disputed that her name is part of computer history.)

Ada Lovelace's actual contribution to Babbage's work is disputed, with evidence to suggest that it was relatively minor. Even Babbage's work is only briefly mentioned in this article. We're trying to achieve a brief overview here. --Robert Merkel 02:55, 31 March 2006 (UTC)

I agree that Ada's work did not help Babbage very much - and one great difficulty is knowing how much of what she wrote came from Babbage and how much was her own, original research.

If you accept the view that she did much of the work herself: then her contribution was essentially to invent the entire art of programming. Her notes show things far more advanced than Babbage was ever planning to construct - so this is perhaps a tenable position. Her notes describe (for the first time) concepts such as the subroutine and the need to have loops that could be terminated at the start or at the end, and a bunch of other subtle details that we all take for granted. Anyway - she is generally credited with being the worlds first programmer - and as such needs to be described in some detail in the article - although arguably in her own right rather than as a footnote to Babbage's achievements.
There is a contrary view: that she merely transcribed what others gave her - which is a far less romantic view - which probably has a lot to do with why this is an unpopular position to take. In this view, she basically translated a paper into English, she took all of her programming ideas from Babbage - and was a complete klutz at mathematics. The latter isn't a problem for me - I'm a total klutz at math too - yet I've been a successful software and hardware designer and a programmer for 30 years and earn a six figure salary doing it. The kind of thinking it takes to come up with the ideas in her notes isn't of a mathematical nature.

Ada has a major programming language named after her - I believe that's an honor only ever bestowed on Ada Lovelace and Blaise Pascal! So, if we believe that her work was original then I would argue that Ada's contribution was possibly even larger than Babbages - since her ideas are still in active use today - and so she must have some mention here. If we don't believe she originated the ideas - then she is a mere footnote and already has more credit than she deserves. I have no idea which of those views is correct - and even if I did, it would be POV not fact. We probably need to adopt some of the language from the main Ada Lovelace article which discusses this in some detail with supporting evidence.

Ada's work (here: [1] and here: [2]) makes stunning reading for a modern programmer - the clarity with which she explains what programming is is quite wonderful given that computers didn't exist at the time she was writing it - but we'll never know whether she was merely good with words (as daughter of one of the worlds greatest poets - this would be unsuprising) or whether the ideas that she so clearly explains are also her own. I don't see how we will ever know - and it's likely that she will tend to be given the benefit of the doubt because it makes a better story that way! SteveBaker 12:59, 31 March 2006 (UTC)

Spelling...

Let's not get into revert wars over American or British spelling, please. --Robert Merkel 04:42, 5 December 2005 (UTC)

Opening text

It strikes me as somewhat disturbing that the first line of this article implies that a computer has to execute stored programs. Plenty of devices that are nearly universally called computers couldn't execute any instructions (like ENIAC). The text should be modified somewhat; especially since the next section does an adequate job of explaining that computers have been around far longer than the Von Neumann architecture. -- uberpenguin 02:00, 18 December 2005 (UTC)

ENIAC could be programmed - it had a plugboard kind of affair to do that. This has been described as 'rewiring the computer to reprogram it' - but that's a little unfair. This is no different (in principle) from a ROM-based computer that can only obey a single program. I don't think we'd exclude something like a mask-programmed ROM-based microcontroller just because it's difficult (or even impossible) to change its program - so long as ENIAC followed instructions, it's a computer. It's a grey area though. A Jaquard loom follows programmed instructions - but it's not considered a computer because it doesn't "process data" using its program. SteveBaker 16:26, 28 February 2006 (UTC)

There are the historical devices, that were called computers, that never executed stored programs, that today might not be called computers, such as the tabulating equipment, where the equivalent of programs were in wired boards that would be plugged into the hardware as needed to be used. Then there are contemporary alternatives to the general purpose computer such as supercomputer where the equivalent of software is moved to hardware. User:AlMac|^(talk) 10:28, 18 January 2006 (UTC)

Eckhart and Mauchly

Check a computer history text; Eckhart and Mauchly are probably responsible for the majority of ideas which von Neumann sketched out in his report. --Robert Merkel 02:08, 18 December 2005 (UTC)

I second that. This is a basic point which is taught in all decent computer history classes. Von Neumann's most important contribution was simply putting it all together in his famous report which then was circulated all over the place (and that's why we call them Von Neumann machines). --Coolcaesar 19:10, 18 December 2005 (UTC)

Picture of computer

Is there some reason for the free publicity Dell and IBM are getting as PC makers with the photo on this page? (That was a rhetorical question) --Eyal Rozenberg 10:21, 27 May 2006 (UTC)

I remember a few months ago, I made a minor edit to the definition of computer to include its communication abilities. I was welcomed with a lot of derision from Merkel et. al. saying that I do not understand computer because I am not a computer scientist and that a computer is much more than a PC. I am now indeed surprised that the new introduction is accompanied by a picture of a PC. Charlie 12:32, 31 December 2005 (UTC)

Fair point - I've removed the picture. Any suggestions for an appropriate replacement? --Robert Merkel 06:31, 16 January 2006 (UTC)

A PC is a computer. A computer isn't necessarily a PC. As a choice for a photo to put at the top of the article, a PC isn't necessarily the best choice - but it's not wrong either. IMHO, anyone who comes to Wikipedia is using a computer - and VERY likely knows what a PC is and looks like. In that regard, a photo of a PC isn't conveying much information beyond "Yes, this *is* the article about 'Computers' - you didn't mis-type it". It might be better to come up with a more thought-provoking or informative picture to head up the article. SteveBaker 21:54, 16 February 2006 (UTC)

Well, I personally think a picture of a PC is the best way to lead the article because it's the form that most people are at least somewhat familiar with (they have to be using one to read Wikipedia). Maybe an IBM/360 mainframe or a high-end workstation (like Sun or SGI) would also work. In contrast, I'm not so sure about the exotic computers that run on toilet paper rolls or Tinker-Toys, since (1) we would have to get clearance for a picture and (2) they're not instantly recognizable to most people as computers. --Coolcaesar 08:38, 17 February 2006 (UTC)

A picture should add something to the article though. If (as we seem to agree) anyone who is reading Wikipedia already has a computer (probably a PC) sitting in front of them then it's unlikely that including a picture of a generic PC would add anything. How about something recognisable as a PC - but with historical interest - an *original* IBM PC or an Apple ][, a Commadore PET or a Radio Shack TRS-80 Model-I? Maybe show the insides of a PC or pick some other kind of computer entirely - a washing machine controller or an ECU from a car? There has to be something we can put up there that simultaneously says "Computer" and imparts some more information that makes the typical reader say "Wow! I never knew that." SteveBaker 14:07, 17 February 2006 (UTC)

An introduction takes a topic, summarises it and relates it to what people know about the world. The rest of the article is what you use to expand people's minds. I think a PC is by far the best choice for the introduction picture as for many, many people a computer and a PC are the same thing. By putting a PC on the top it allows people to start the article at a comfortable level. The more novel pictures should be saved for the rest of the article, where the process of explaining the topic takes place. --FearedInLasVegas 19:05, 17 July 2006 (UTC)

I could not disagree more. Someone comes to the article with the (incorrect) idea that a computer is a PC (a very common view) - and the very first thing they see CONFIRMS that exact failure of imagination! That's ridiculous - we can do better. Furthermore, a photo of a PC has precisely zero information content since everyone who can reach Wikipedia has one literally at their fingertips. Finally, whenever we put up a photo of a current PC, someone replaces it with another PC from another manufacturer - so we are giving free advertising to Dell/Gateway/Apple/whoever. We can (and do) give them that initial 'comfort' factor by showing a Lego computer in a kid-friendly yellow case. At any rate - the main discussion of this is towards the bottom of this talk page - you should follow that discussion and post further remarks there. SteveBaker 19:44, 17 July 2006 (UTC)

I thought this was the best place to talk about the front picture. Do you recommend that I create a new heading at the bottom of the page about this? I think that when someone sees a picture of a PC on the computer page it tells them that a PC is a computer (note the order). The biggest reason I can see not to do this is because of advertising, although I can't see how that applies any more to this than pictures of Lego computers. On a slightly different point, there are no pictures of PCs on this page at all. I think this is necessary as, like you say, most people have PC on their minds when they come to this page. Not to acknowledge this association is missing something out of the article. --FearedInLasVegas 16:22, 18 July 2006 (UTC)

Still 100% with Steve on this one. It would take some mighty convincing arguments before I'd ever see a PC featured as the first image in this article. (Hint: "PCs are common forms of computers that people readily identify as computers" is not a convincing argument) -- uberpenguin @ 2006-07-17 22:30Z

"I think that when someone sees a picture of a PC on the computer page it tells them that a PC is a computer (note the order)." - but people reading Wikipedia already know that a PC is a computer - they have one right within a couple of feet of them. It's definitely not a high priority matter to tell them something they already know. The first photo in the article is by far the biggest bang for the buck as regards passing on information. We shouldn't waste that telling them something they already know and quite possibly make matters worse by reinforcing an unfortunate assumption. That photo isn't decoration - and it isn't there to give people the confidence that they found the right page - it's there to convey useful, encyclopeadic information. SteveBaker 20:16, 18 July 2006 (UTC)

Language is Software

From Daniel C. Dennett's Consciousness Explained 1991; ISBN: 0316180661 p. 302:

The philosopher Justin Leiber sums up the role of language in shaping our mental lives:

Looking at ourselves from the computer viewpoint, we cannot avoid seeing that natural language is our most important "programming language." This means that a vast portion of our knowledge and activity is, for us, best communicated and understood in our natural language... One could say that natural language was our first great original artifact and, since, as we increasingly realize, languages are machines, so natural language, with our brains to run it, was our primal invention of the universal computer. One could say this except for the sneaking suspicion that language isn't something we invented but something we became, not something we constructed but something in which we created, and recreated, ourselves. [Leiber, 1991, page 8]

Yesselman 21:40, 12 January 2006 (UTC)

That's no big deal. Leiber is simply putting a different spin on old ideas that were better articulated by Julian Jaynes, Richard Dawkins, Benjamin Lee Whorf, and Neal Stephenson (see Snow Crash). And yes, there are already some people in neuroscience who argue that the human brain is essentially an analog computer. Plus there's the sci-fi term wetware that's been around for a while. --Coolcaesar 08:28, 13 January 2006 (UTC)

I always liked "A Computer is a machine for following instructions" - but I don't recall who first said it. SteveBaker 21:48, 16 February 2006 (UTC)

External Links

the last edit changed the external link http://www.tech-forums.net/ with http://www.compuforums.org/ leaving the same description of the link, these two forums are not the same, and so I find the edit questionable. Rv?

It's probably someone promoting their favorite forum. If there is no solid reason to use one or the other, consider having either both or neither. SteveBaker 14:01, 17 February 2006 (UTC)

http://www.compuforums.org/ was first added on 14 Sept 2005 which is a long time ago, perhaps there's a reason for that link, so I will just rv. for now. Anyone could delete the link altogether, thanks for your answer. --A/B 'Shipper 女 ^(talk) 14:29, 17 February 2006 (UTC)

Are these really computers?

We've defined a computer (correctly IMHO) as:

 "A computer is a machine for manipulating data according to a list of instructions - a program."

...yet we talk about the Abacus, the Antikythera mechanism, and clockwork calculating devices as early computers. These are NOT computers according to our own definition - they are merely calculators. They can't follow a list of instructions - they are not "programmable". Even Babbages difference engine wasn't programmable - that honor goes to his analytical engine - which was never built and may not even have worked.

Furthermore, if we allow the Abacus to be counted as a computer (or even as a calculator) then human fingers get the credit for being the first computer since all an abacus is is a mechanical aid to counting. It doesn't automate any part of the process. Also, if the Abacus makes it, then so must the slide-rule - which is a VASTLY more sophisticated calculating machine because it can multiply and divide.

This is inconsistent. Either we must accept calculators (for want of a better term) as computers and change our headline definition to allow non-programmable machines to be computers - or we must rewrite the section that includes such mechanical wonders as the Antikythera and the Difference engine and explain that whilst they were antecedents of the computer - they are not computers because they are not programmable. I suggest the second course of action.

SteveBaker 14:26, 28 February 2006 (UTC)

Scroll up and you'll see me bringing up more or less the same issue. I don't really agree with the headline definition of a computer since it only allows inclusion of modern programmable computers, not many older devices that could still be considered computers but were not programmable. -- uberpenguin 15:11, 28 February 2006 (UTC)

Then you must also change the definition of 'Computer'...one or the other - you can't have both! The article clearly defines what a computer is - then goes on to list a bunch of things that are clearly not computers under that definition. We can argue about the definition - but no matter what, you have to admit that the article is inconsistent. My change (which someone just Rv'ed) fixed that - if you don't agree with my change (or something similar to it) - then you MUST change the definition in the headline. HOWEVER, if you do that then a difference engine is a computer, so is a slide-rule, an abacus and your fingers. Come to think of it, a pile of rocks is a computer under that kind of definition because a pile of rocks is a primitive abacus. I think the clear, sharp line that distinguishes a computer from a calculator is programmability - arguably it's a Church-Turing thing. Draw that line and you must rule out the difference engine, the abacus and it's ilk. We can still recognise the contribution of calculators to early computers - and perhaps even point out the grey areas - but there IS a bright line here. SteveBaker 16:10, 28 February 2006 (UTC)

We begin the history section with saying that "Originally, the term "computer" referred to" something different. Maybe we could include your argument later in that passage to pick up the thought that the definition of a computer has evolved to programmable computational devices, to clearly distinguish between early computational devices and a modern computer.--Johnnyw 20:35, 28 February 2006 (UTC)

I think the easiest way out of the situation is to have the opening definition apply to only programmable computers, describe the evolution of the term in the history section, and mainly concentrate on programmable computers in the rest of the article. That more or less follows the format of most other good computer hardware related articles here. -- uberpenguin 23:12, 28 February 2006 (UTC)

I disagree that the original meaning of the term "computer" referred to something different. Human 'computers' were used in ways precisely the same as modern silicon computers. The guy who ran the team "programmed" and "optimised" the use of the compters (people) in precisely the way a modern computer programmer does with his hardware. To understand how these human computers were used, read some of the biographies of Richard Feynman who (amongst other things) ran the 'computer' group during the Manhatten project. He organised the work-flow of his (human) computers - optimising their list of instuctions - making 'subroutines' that one group could calculate repetitively, using 'if' statements to skip over unneeded sections of his "program" depending on the results of previous calculations...it WAS programming. The team of humans who did the work were both called computers and were "manipulating data according to a list of instructions" - which fits the modern definition of the word computer. I don't think the meaning of the term has actually changed all that much. SteveBaker 00:51, 1 March 2006 (UTC)

I think you're blurring the line between definition and analogy If you have to use quotations around "program" and "optimize" for your statement to hold water, then perhaps your suggestion to include human computers in the current article's definition is a bit too metaphoric. Anyway, I'm not very interested in rewriting this article myself, so I won't really argue this much. -- uberpenguin 02:06, 3 March 2006 (UTC)

My use of quotations was erroneous - my apologies for misleading. I intended no analogy. Human computers had been used like this for lots of tasks in many situations - but the clearest explanation I've found is in the writings of Richard P.Feynman. There are many versions of this in different biographies - the one that I picked off my bookshelf first was Feynman's "The Pleasure of Finding Things Out". In the chapter entitled "Los Alamos from the Bottom" we hear of the times when he was running the computer center at Los Alamos - consisting of dozens of humal computers with Marchant mechanical calculators. He organised a system of colored index cards that were passed from person to person containing the data - with each person having a set of algorithmic steps to perform on each data element depending on the color of the file card. This included instructions on who to pass the results on to and on what color of index card. Any modern programmer would recognise what Feynman was doing. He had a parallel array of computers (human ones) - each of which had a written program, data packets flowing in and out, and an arithmetic unit for each human computer in the form of the Marchant calculator. He discusses optimising his procedures for doing this - and then eventually replacing his human calculators with IBM tabulators (which were evidently just non-programmable multipliers and adders). This isn't just an analogy - it's a process that is in essence indistiguishable from an electronic cluster computer - except that it uses human computers. The human computers with their index cards and Marchant calculators would (as I understand it) have no problem qualifying under the Church-Turing constraints. SteveBaker 04:32, 3 March 2006 (UTC)

Can I make a suggestion at this point? There's plenty of scope for an article on human computers. --Robert Merkel 04:49, 3 March 2006 (UTC)

Computational

Either the "Computer" article needs to explain "computational" as seen through the eyes of Douglas Hofstatder, Donald E. Knuth and Alan Turing or a separate article created anew. Arturo Ortiz Tapia 13:06 hrs, +6 GMT., February 28th 2006

Could you elaborate a little on what you mean by "computational"? --Robert Merkel 23:51, 28 February 2006 (UTC)

Criticism of the ABC Picture

I just removed the ABC picture from the article. Here are my reasons:

When shrunk to thumbnail size, all the labels on the picture turn into little dots that look like the image has noise all over it. At the very least we need a version without the labels and clutter to use as the thumbnail.
When the diagram is brought to full size, those same labels are fairly meaningless without some explanatory text surrounding them.
The image itself gives no idea of size or context. Is this the whole computer? Is it just a piece of it?
A photo would work better here - diagrams are fine for explaining the inner workings and component parts - but that's not what we're doing here. We are talking about the capabilities of the machine as a whole and for the image to earn it's price, it has to tell us something about this machine that the text doesn't do. What it looks like, how big it is...stuff like that.

(According to the history comments, Uberpenguin agrees with me) SteveBaker 17:31, 5 March 2006 (UTC)

Agree fully, but let me also give you some background on why I didn't delete it myself. There's currently an anonymous editor who is pretty ardent about including information about the ABC and the related 1970s court case in several computer related articles, regardless of whether the information is relevant to the article or not (see the recent history of CPU and UNIVAC for examples of what I mean). I've asked him to stop indiscriminately adding the text to articles and discuss it on the talk pages, so we'll see if he complies. He's accused me in his edit summaries of trying to obscure history by removing what I believe to be irrelevant text and pictures, so I didn't particularly want to start trouble by removing the image here where it could conceivably be relevant. -- uberpenguin 17:52, 5 March 2006 (UTC)

I also agree with both of you that the diagram should not be there. I remember studying the ABC in college (as well as the work of competitors like Zuse, Stibitz, Eckert & Mauchly, Shannon, etc.) and I always thought that Atanasoff's courtroom quest was rather quixotic and silly. I also agree that if the ABC should be shown at all, a photo would be more appropriate. --Coolcaesar 19:00, 5 March 2006 (UTC)

This second picture was added after considering that the previus one was ugly, u were not educated enought to understand what was writen on it and there is no people on it so you can say how big it is ... now this one had met all your criteria but still you've removed it just becouse four people can deside the censensorship. Now I've met all your criteria be good boys and put back the second picture yourself !!! P.S. It's not my and the rest of the readers problem that you've been told wrong in class just because of the cold war which is over BTW !!!71.99.137.20 00:21, 7 March 2006 (UTC)

I almost forgot I have writen to several institutions and medias (like NY Post, Newsweek and etc.) about your censorship, so from now on think twice what your sparing about history. 71.99.137.20 00:27, 7 March 2006 (UTC)

Re: Your peculiar comment about the educational level of people you've never met...it's not a matter of OUR education. I've designed computers vastly more complex than ABC from scratch at the ASIC chip level - I fully understand all of the points on the diagram. (You might maybe want to apologise - but that's OK). However, that's not the point. The point is whether the average Wikipedia reader who looked up 'Computer' would understand the labels without supporting text. IMHO, anyone who needs to look up computers in an encylopedia (as opposed to writing articles in it) should be expected to have a lower level of knowledge on the subject and therefore would NOT be expected to understand the diagram. People who are computer experts may be the ones to write this article - but it's 8 year old kids with a class assignment who will be reading it. As for censorship - if Wikipedia editors (that's all of us) can't edit the article once someone has put something into it without being accused of censorship then Wikipedia cannot exist because it's all about collaborative editing. Right here under the edit box it says "If you don't want your writing to be edited mercilessly or redistributed by others, do not submit it." - that applies to photos too. If I don't like what you put into the article - then I can take it out again. That's not censorship - it's me fixing a problem that I (rightly or wrongly) perceive. Going screaming to the media just because someone doesn't agree with your choice of photograph does not speak well of you - and if truth be told, I very much doubt that even one person who reads your inflamatory remarks believes for one moment that you actually did that - this does nothing for your reputation. SteveBaker 04:40, 7 March 2006 (UTC)

Please learn the difference between censorship and contextual relevance. You have yet to give a single comprehensible reason as to why we should go out of our way to display the ABC over one of the other plethora of contemporary computers. Perhaps I was a bit hasty to remove the picture because of your recent history of edits, so allow me to explain my position with a measure of lucidity. The abridged history section here only has enough text to merit perhaps one picture of a very old electronic computer. I'm not particularly opposed to it being of the ABC as opposed to ENIAC, but a concern is that the ABC image you uploaded has unknown copyright status. If that were fixed, I'd have no problems putting it to a vote as to whether the ENIAC image should be replaced with ABC.

Now, please take this opportunity to reflect on your unnecessary behavior; labeling and name-calling. I'm trying to be as accomodating as possible (especially since I see no particularly good reason to change an image that serves purely as an example), but your reverting and then attacking anybody who wishes to discuss your changes reasonably is unacceptable. Please read WP:NPA and try to behave in a civil fashion. -- uberpenguin 00:52, 7 March 2006 (UTC)

I would like to say, however, that if this WERE put to a vote I'd still oppose the ABC picture inclusion because it is of a replica, not the original. For brief informative purposes illustrating an early electronic computer, I believe it's much more valuable to use an image of the original. -- uberpenguin 01:17, 7 March 2006 (UTC)

Yeah - but from the point of view of telling our readers what the computer looked like - if it's a good replica then a photograph of it is a reasonable stand-in. The useful information it conveys is still there. It would have been dishonest to protray it as a photo of the real thing - but so long as it's clearly labelled (which it is) - I think that's OK with me. SteveBaker 04:43, 7 March 2006 (UTC)

Some of this debate is getting out of hand. Writing letters to the press is a ridiculous over-reaction to Wiki editing. If you can't stand this kind of thing - you shouldn't be here - your text and photo choices WILL be edited. Sometimes by people who know better than you - sometimes by random idiots - in neither case could it remotely be called "Censorship".

Now - we should have a rational debate about this image. Here are some facts:

The first ABC picture was unsuitable - this is widely agreed and I set out a comprehensive set of REASONS why I deleted it. I didn't do it on my own (although that would have been acceptable) - at least one other respected Wikipedian agreed with me - and I explained myself.
The second ABC picture would probably be suitable from a quality standpoint - but it has NO COPYRIGHT ATTRIBUTION. So we can't use it - period. It should never have been uploaded in the first place. We can't afford for Wikipedia to be sued by the copyright owner. So, NEVER, EVER post a photo without cast iron copyright permissions. So, a slap on the wrist goes to the person who put it up here.
Given the above, removing the second picture was 100% warranted - there should be no dispute over that - read the Wiki rules.
HOWEVER: When you do something like reverting someone else's work or a semi-reasonable photo, it's good manners to put a note here in talk to say why you did it. This engenders debate - which hopefully comes to some conclusion without a major bustup. So a slap on the wrist for the person who removed it because of the lack of a justification - not because of the removal of the image.

Those are the facts. Now my opionions:

If the person who posted the image can find another with SOLID copyright history (and given the grief about this, believe me, it'll be checked) - then we should have a debate over whether ENIAC or ABC is more suitable as the computer to illustrate this section - or whether both photos can be included. But that debate can't happen until we see this hypothetical ABC photograph.
The issue of this anonymous poster trying to get the ABC story into Wiki in a bunch of places is (IMHO) irrelevent here. If it belongs here - it should be here - if it doesn't, it doesn't and it's truly irrelevent to the quality of this page what the motivations of the poster are.

So, kiss & make up and let's get a nice article together here! SteveBaker 02:29, 7 March 2006 (UTC)

To be perfectly honest, I jumped the gun on reverting that image because I thought it was the same image as before. That was my mistake, and thus the reason for my lack of justification for its removal. That was, admittedly, wrong, but as we both have pointed out there are valid reasons not to use the new image either. As I stated above, I have nothing fundamentally against using a picture of the ABC here, but I do have concerns with the specifics of both proposed images and the worry of making the article cluttered by weighing the text down with too many images.

I'm perfectly willing to discuss the inclusion of an ABC picture here. Please excuse me if I seem a bit tense or borderline on ad hom, but perhaps you'll understand my irritation at being openly insulted without provocation in this and other articles that the anon editor has recently participated in. -- uberpenguin 02:43, 7 March 2006 (UTC)

I understand. But listen: In a recent study of online communications (email as it happens), 80% of writers believed that their readers would comprehend the 'tone' of their messages - where in fact, only 20% actually did. This is worse than chance! So we all have to take extra special precautions in all online dealings not to assume motives - one mans insult is another mans honest mistake. So - we still can't do anything either way about a better ABC photo. The one that was there last WILL be automatically deleted by the copyright police bot in just a few days. We can't/mustn't/won't use it. <shrug> If User:71.99.137.20 would care to provide either a different (and copyright-free) photo - or ascertain that in fact the photo that was posted is in fact legal - then we can discuss whether we want more photos or whatever. Right now, it's a moot issue because we can't argue about a photo that doesn't exist. QED SteveBaker 04:20, 7 March 2006 (UTC)

I agree with SteveBaker. Also, I believe the anonymous user is not editing in good faith, since all professional computer historians (Ceruzzi, Aspray, et al.) do not see the ABC as the only "first computer," regardless of what the Atanasoff family and their friends believe. I suggest that the anonymous user should be blocked if he or she fails to cooperate with Wikipedia policy. --Coolcaesar 06:39, 7 March 2006 (UTC)

I would not advocate blocking. If the Atanasoff's have evidence then let evidence be presented - lets discuss it's validity and document the results. We can't do original research here - but we most certainly can & should read all available sources and see things from all available points of view. If this debate has already been carried out on other Wikipedia pages - then let's hear about that too. SteveBaker 12:42, 7 March 2006 (UTC)

Steve this is verry good what you're saying but those are people who know the facts but are just prejudice and just whant to shut my mouth I've been blocked once already and denied the ability to create account so I can defend that I want to say. You don't need to believe me all I'm trying to write is a sworn testamony in US court, go to this adress and see for yourself [3] 71.99.137.20 21:11, 7 March 2006 (UTC)

Sir, nobody has blocked you from registering an account, and your IP was temporarily blocked for violating the WP:3RR (nobody — including registered users — may themselves revert any article more than three times in a 24 hour period). Instead of waving the prejudice finger again, why don't you either find a picture that can be legally used on Wikipedia or make valuable edits? -- uberpenguin 21:01, 7 March 2006 (UTC)

The Iova State University will be more than happy to give me whatever the licence is needed. And that's realy stange because in order to mail Voice_of_All I needed to be loged on, and up until I changed my IP there was no "create an accont" on the log on page. 71.99.137.20 21:11, 7 March 2006 (UTC)

Assuming that Iowa State owns the copyrights to that picture, you will need them to release (and have written proof of such release) it to either GFDL or public domain. -- uberpenguin 21:18, 7 March 2006 (UTC)

Incidentally, if you're going to go through the trouble of getting the school to release a picture, get this one since it's actually a picture of the ORIGINAL computer, and not a replica. -- uberpenguin 21:41, 7 March 2006 (UTC)

Hmmm - I've been reading up on this ABC gizmo. Once again - from the first line of our article:

  "A computer is a machine for manipulating data according to a list of instructions - a program."

(my emphasis)

The ABC is not (according to that definition) a computer at all. It's another fancy hardwired calculator. It's pretty impressive for it's time though - very small compared to the roomfulls of stuff you needed for ENIAC. But in the end, it's not a computer - it's an ALU and some memory - but it lacks any kind of sequencer/program-counter/program storage. If you allow non-programmable devices to be computers, how do you distinguish them from calculators? The single unique thing that makes a computer different is that it has a stored program that you can change in order to do anything your imagination can come up with. Sorry - but I now agree 100% with uberpenguin - there is no need to say much about this thing anymore than there is a need to go into the workings of the Antikythera. The sentence we have about the ABC right now is PLENTY. SteveBaker 04:59, 10 March 2006 (UTC)

I agree with all of this. It's a fascinating device, and a major step along the road to what we now call computers, but it was a special-purpose machine with no programmability. --Robert Merkel 05:13, 10 March 2006 (UTC)

I'm sorry Steve, but ABC not only does instructions, but it does 30 instructions at one time (the first parallel processing). Talking about sequencer it has system clock that's actually mechanical (the motor on the right side). The memory modules are made from capacitors and vacuum tube(exactly the way DRAM is constructed), although the access is mechanical(drums) instead of electronic. I can see that you jump to conclusions pretty easy. You have to understand ABC is not a machine or special-purpose computer it's the first "digital" computer prototype and because it has laid the "digital" fundamentals in just one version it's a GIANT step. You can't expect to have all the issues smoothed out with just the first version, but at least the bases.

I'm going to put this here so you can tell me what do you agree or not about ABC or has to be paraphrased "The first Digital computer to use binary base, logical operators instead of counters by using vacuum tube (transistors), memory based (capacitors - today's DRAM), separation of memory and computing functions, Parallel processing (it actually did 30 instructions at once), and system clock]]".

(RESTORED ORIGINAL CONVERSATIONAL ORDER TO THIS TALK PAGE SteveBaker 18:25, 10 March 2006 (UTC))

So are you now claiming that it was programmable? Doing 30 instructions at once is not programmability if they can't be sequenced automatically. Bottom line: Was the ABC Turing-complete? I'm pretty sure the answer is "no". If so, then this does not fit the common definition of what a 'computer' is. So in terms of your proposed wording: (a) Don't say it's a computer and (b) if it's not a computer do we really want to go into so much detail about Yet Another Calculator? SteveBaker 07:43, 10 March 2006 (UTC)

And Uberpenguin I can see that the histories has been sanitized very good, even where I have adding text has been cleared. On the question "Who am I to say that John Atanasoff is genius" - I didn'twrote anywhere that he is a genius, but for sure the President of the United States already did said that by giving him "National Medal of Technology" and the United States court had ruled that John Vincent Atanasoff and Clifford Berry hadconstructed the first electronic digital computer at Iowa State College. I what also to add here what John Mauchly confirmed under oath:

- He spent from 13 June 1941 to the morning of 18 June 1941 as a guest in Atanasoff's home in Ames. - During this period as Atanasoff's guest he spent uncounted hours in discussions of the Atanasoff Berry Computer and computer theory with John Atanasoff and Clifford Berry. - On three or four days he accompanied Atanasoff to his office in the Physics Building and observed the Atanasoff Berry Computer in the company of Atanasoff and Clifford Berry. - He had seen demonstrations of the operations or some phases of the functions of the Atanasoff Berry Computer and might have engaged in manipulation of some parts of the machine with Clifford Berry. - He was permitted to read Atanasoff's 35-page manuscript on the construction and operation of the Atanasoff Berry Computer from cover to cover and probably did read it. Atanasoff and Berry had willingly answered questions and entered intodiscussions with him about the machine and the booklet, but Atanasoff had refused to let him take a copy to Pennsylvania. - Immediately after his visit to Iowa State in June, Mauchly had written letters to Atanasoff and to his meteorologist friend, Helms Clayton, expressing enthusiasm about the Atanasoff Berry Computer and had taken a crash course in electronics at the University of Pennsylvania. - On 15 August 1941 he wrote a comprehensive memorandum on the difference between analog calculators and pulse devices that incorporated some ideas that were almost identical with those in Atanasoff's 35-page manuscript on the ABC. - On 30 September 1941 he had written to Atanasoff suggesting a cooperative effort to develop an Atanasoff computer and had asked if Atanasoff had any objection to him using some of the Atanasoff concepts in a computer machine that he was considering building. 213.222.54.133 08:25, 10 March 2006 (UTC)

(Cool! A new anonymous IP address has appeared!). Look - for the purposes of THIS article, we really don't care who invented what, who was smarter than who, or who had the best lawyers. The court case is irrelevent to THIS article. What matters is whether we consider the ABC to be a true computer. So far, all but one person here agrees that we don't. I took an informal poll at work today and asked my co-workers: "What single feature distinguishes a computer from a calculator?" - out of 18 people, I got UNANIMOUS agreement that general programmability is the sole, single feature. We agree that it doesn't matter whether it's an analog device, a base 10 or base 3 device, how it's memory works or whether it is a parallel machine or not - whilst all of those are important innovations along the route to making a practical computer, the one single defining feature is the ability to write a program, put it into the machine somehow and have the machine execute it as a sequence of steps. I'm convinced that the ABC couldn't do that - so by all practical definitions, it's a CALCULATOR...albeit a highly advanced one with all sorts of neat modern features and albeit that Atanasoff invented them all and was a really clever guy. I could care less about that at this point. The sole issue for THIS article is whether we tell our readers that the ABC was the worlds first computer - it seems utterly clear to me that is was not. Unless I see evidence to the contrary - I think the debate is over. SteveBaker 18:25, 10 March 2006 (UTC)

I agree that the argument should be over. When we ask for the anonymous editor(s) to justify that the ABC was a computer by modern standards, we get back rhetoric describing the ABC's features and the 1970s court case. I'm sorry, but US patent court is not any kind of recognized authority on technology or computer history. The ABC was not programmable, therefore it was not a computer by the standards of this article or most modern definitions. -- uberpenguin 20:36, 10 March 2006 (UTC)

What is a "computer"?

I've moved part of the discussion originally found on Talk:Atanasoff-Berry Computer here since it has become irrelevant to that article.

Here's the problem though. On this page of Wikipedia, we had said that the ABC is a computer - without adequately explaining that it is not a computer AT ALL in the modern sense of the word - it's essentially just an oversized pocket calculator. That allows this page (and certain anon users) to start claiming (without qualification) that the ABC was the world first digital computer - which is certainly not true in the modern sense of the word. IMHO, it wasn't even true in the 1940's sense of the word in which the human who OPERATED the ABC would be called a 'computer'. If you allow the claim of 'First ever digital computer' here - how can you deny it elsewhere (and I'm thinking of course of the Computer and CPU articles) without Wikipedia being inconsistant. So THIS article needs to explain - very carefully - the restrictions on that claim otherwise other pages will end up needing heavy revision in order to remove ENIAC and replace it with ABC as the first ever digital computer - and to do so would be exceedingly misleading. By all means, rephrase (or even revert) what I wrote - but you must do something because what was there before is WRONG. IMHO, you don't write a modern encyclopedia using words in the sense that they were meant in the 1940's (at least not without careful consideration and qualification)...otherwise we are going to end up calling a lost of merely happy people gay and I really don't think that's a good idea!! SteveBaker 18:49, 10 March 2006 (UTC)

If we're using the modern (i.e. most of the last century and this one) definition of computer, yes, I'd agree that programmability is the defining feature and therefore ABC does not qualify as a computer. I think the article shouldn't call it a computer, but should point out that by some definitions it would be called a computer, but by the modern definition it is not one. -- uberpenguin 19:06, 10 March 2006 (UTC)

Oh come on, a computer is one that computes. Whether it's a person determining ballistic trajectories or a machine determining ballistic trajectories or an electronic computer determining ballistic trajectories...it's all the same.

The ABC is not turing complete and nor did it have stored programs, but it computed. Use your qualifiers ("modern") if you wish but a stored program computer is a subset of all computers and such qualifiers are irrelevant. Just like a square has four equal sides (a "modern rectangle" if you will and get my point), but that doesn't make it any less of a rectangle. The ABC is a computer and to call it any less is pure rubbish just as if you said a square isn't a rectangle.

The ABC was the first electronic digital computer. The difference engine was not electronic, so it didn't beat the ABC to that one. Cburnett 01:44, 11 March 2006 (UTC)

Then we should change the definition of "Computer" on this page. Shall we also rope in calculators and abaci into the "computer" definition? I'm just playing devil's advocate here to point out that it's not easy to come up with a good definition of "computer" that will stand and be consistent across multiple articles. -- uberpenguin 01:58, 11 March 2006 (UTC)

Yes, a calculator is a computer. It is a specific purposed (possibly programmable) computer. An abucus is not a computer because it is an instrument and is no more of a computer than beans and wire. An abacus does no computation but the user performs computations by using it. Do you call an AND gate a computer? Certainly not, but it's definitely used by computers. An abacus is not a computer, but it is used by one.

I have no idea why this is so hard for you to comprehend. A computer is one that computes. It has several subcategories: human, mechanical, digital, electronic, and probably more. Under electronic there's analog and digital. Under digital there's programmable and non-programmable. There's an entire taxonomy to computers, but all that is required to be a computer is membership in this taxonomy. And an electronic, digital, non-stored program machine that computes is definitely in said taxonomy. Cburnett 02:11, 11 March 2006 (UTC)

But what does 'to compute' mean? How does it differ from 'to calculate'? Assuming we are writing Wikipedia in modern english - and not some older dialect - we have to ask what constitutes a computer and what a calculator? In every modern usage, a computer is a programmable machine and a calculator is not (except of course when we apply the qualifier "programmable calculator" - but that's WHY we say that rather than just "calculator"). We have desktop computers, there is an engine management computer in my car - both are programmable - but this little box with all the buttons on is a calculator we never call those little four function calculators "computers" - NEVER! We all understand the usage - where is the confusion? If you really truly believe that something that is not programmable is a computer - then where do we draw the line. It is PERFECTLY possible to design a programmable device that is Turing-complete yet has no hardware for arithmetic or numeric capabilities. This would clearly be a computer - because we could program it and have it implement arithmetic and such in software. But a calculator - without programmable features is just not something recognisable as a computer.

Where would you draw the line between what is a computer and what isn't? Is the Difference Engine a computer? How about one 'adder' circuit? Is an abacus a computer? A slide rule? A table of logarithms. Ten human fingers? A pile of rocks? ...because you can calculate with a pile of rocks. If the definition of computer is SO lax as to include anything that can perform any kind of arithmetic or logical operation - then a light switch is a digital electronic computer that pre-dates the ABC by a century or more.

But if you insist that a non-programmable device is a computer then you'll have to go through articles such as 'Computer' (which states that a computer is programmable in the very first sentence of the article) and find some better definition. To do otherwise would be to confuse the heck out of our readers. They read the 'Computer' page - it says computers are programmable data processors - then they see the reference to the ABC, click through to this page and it says something to the effect that this device is a computer - but that it's not programmable....now what? They've just read a flat out contradiction...how does this get resolved?

SteveBaker 02:59, 11 March 2006 (UTC)

Let me preface this by saying that your reply is so incoherent and jumpy that it's hard to form a good reply.

You need to read Von Neumann architecture. I quote:

The earliest computing machines had fixed programs. Some very simple computers still use this design, either for simplicity or training purposes. For example, a desk calculator (in principle) is a fixed program computer. It can do basic mathematics, but it cannot be used as a word processor or to run video games.

So the von neumann arch article recognizes fixed programs as computers. But ignoring semantics and etymology, just look at the number of recognitions that it's a computer. It's even in the name. Honestly, the IEEE recognizes it as a computer. So if you want to challenge the term, then wikipedia is not the forum for such a challenge. Go write the IEEE, or perhaps the US district court of Minnesota that named the ABC as the "first electronic digital computer".

I think the IEEE and US courts say enough to null your argument. But if you still feel you're right, then stop making WP your soap box and argue it in a proper forum.

All that said since you seem quite confused: rocks, fingers, abacus, etc. are instruments or aids for calculation and are not computers. No more than my shoes or rolls of toilet paper are computers. I can use any of them as aids to compute, which makes me a computer.

Semi-finally, you're using WP as an authoritative reference. As much as I like WP and have contributed to it, the word of IEEE and the US courts says more than the introductory sentence to "computer" on a web page.

Finally, a light switch? For serious? I can understand fleshing out an argument and seeing how far you can stretch something (and even being sarcastic to point out the flaws in the others argument).....but seriously: a light switch? Rocks? As computers. Come on. Surely you're more intelligent than that and can look up the word in the dictionary and understand how I'm using it to know that I wouldn't consider rocks or a light switch as a computer. I even stated an abacus is not a computer in that which you replied, so why would you go so far to saw a light switch. Come on! I haven't treated you with any disrespect, so why do you have to do so to me by using asinine examples that I already discounted? I only say this so that any further replies don't go so, for lack of a better word, wrong. Cburnett 05:13, 11 March 2006 (UTC)

His reply was in no way incoherent; please do not label a different point of view as invalid just because you don't agree with it. We are simply asking, at what clear point do you draw the line between computer and calculator? Steve is suggesting using the Church-Turing thesis as a benchmark; you are more or less expecting this to be evaluated on a per-case basis. I agree with Steve that the definition of "computer" has changed a good deal; in modern usage you never see a non-programmable device called a computer (find some common counterexamples and you've made your point). Forgiving Steve's hyperbole, would you consider an ALU to be a computer in its own right? A series of op-amp integrators? Both certainly are computing devices, but are they computers? I know that neither of us would answer "yes" to that question, but I mention them as examples because your proposed definition of "computer" is so broad that you could conceivably include these devices.

Please understand that we believe the definition of computer to no longer be as cut-and-dry as you understand it to be. We aren't treating you with disrespect in any way, just attempting to make the point that we believe your definition of "computer" to be far too inclusive to be useful on an encyclopedia. That being said, I'm not sure why the court case keeps coming up. If you consider patent court an authority on defining technology or language, you'd have a hard time supporting your argument. Furthermore, the IEEE has the luxury of not having to agree upon a formal definition of "Computer." We do, and thus this discussion. On your WP comments; nobody is claiming WP as a reliable resource, we only have noted the need for consistency here. -- uberpenguin 14:01, 11 March 2006 (UTC)

The IEEE doesn't have to define computer, but they clearly include the ABC in it by calling it the first electronic, digital computer. If they didn't believe in calling it a computer, then they wouldn't. The shear number of people that call it a computer is sufficient enough to make your argument that it's not a computer basically original research.

Computer Organization & Design by Patterson & Hennessy (ISBN 1-55860-490-X Parameter error in {{ISBN}}: checksum) calls the ABC "a small-scale electronic computer" and "a special-purpose computer". So the IEEE call it a computer, a district court calls it a computer, and a professor in comp arch from Berkeley and Stanford call it a computer. I did a google search and came up with no one discrediting the ABC as a computer.

Honestly, how many authoritative sources are needed? Why should I, and the readers of WP, take the argument of an undergrad (uberpenguin; I couldn't find much about Steve except that he has a son so he's probably not an undergrad) over than of professors, texts, and the IEEE and pretty much every source out there? I don't really care if you agree on my definition of computer — for another article — but everything I find calls it a computer. You are merely confusing yourself with the modern flavor of computers with the definition of "computer."

Again, if you desire to change the name of the ABC and the definition of "computer" then WP is far from the correct forum. If you want to make the turing test the de facto standard for labeling something a computer, then you are also on the wrong forum for such a change. I most certaily agree that a "modern computer" is one that does pass the test, but the modern computer is in a subcategory of "computer". Heck, to a lot of people "computer" means Windows but I'm nowhere near ready to acquiesce that definition. Cburnett 15:38, 11 March 2006 (UTC)

We really should be moving this discussion over to computer, because the crux of this is "how do you define computer", not "what is ABC's name". Incidentally, mentioning my educational status skirts the border of ad-hom. Questioning other authors' knowledge isn't a valid way to make your point here. -- uberpenguin 18:13, 11 March 2006 (UTC)

Clarification: I want to continue this discussion, but would rather do so at computer since it's become almost totally irrelevant to this article. -- uberpenguin 18:20, 11 March 2006 (UTC)

Once you start asking for people's credentials - you know things have degenerated too far for a reasonable conclusion. FWIW, I'm a professional computer/software designer. I've been programming computers since 1973. I've designed simple CPU's and I currently am the technical lead for a group of a half dozen programmers. I really want to know what YOU (Cburnett) would define a computer to be. Because if it's something that merely performs arithmetic - then we have a serious problem. The 'pile of rocks' and 'light switch' examples were (of course) merely an effort to have you tell me why they are excluded from the definition of a computer (I think we all agree that they are not) yet the ABC is. If you can clearly articulate what that definition is - and quote some references to back it up - then I'll be first in line to change the first sentence of the Computer article.

Here's the problem though. If doing arithmetic is sufficient then an abacus (presumably) counts as a computer. Lots of people use them for doing arithmetic - and in some cases are faster at doing some classes of operation than is a pocket calculator. So if you do accept an abacus - then I regret that a pile of rocks becomes hard to exclude. If I want to add 11 to 17 then I move 11 rocks off into a separate pile, then move 17 rocks off into another pile - then push the two piles together and count how many rocks I have. The pile of rocks just calculated 11+17=28. But as I explained, mere arithmetic is not enough to define a computer. Many years ago (when technology wasn't what it is now), I had to build a massively parallel (but pretty slow and stupid) computer array for doing graphics processing. Because the chip technology I was using wasn't all that great - and I needed to pack 128 computers onto a single chip, I build a ONE BIT computer - with the only operations being (from memory) NAND, SHIFT (which means 'copy this bit into carry and replace with zero), copy, load-with-1, load-with-0 and jump on carry. (Well, there was a little more than that - but not much). This computer could do NO ARITHMETIC in it's basic hardware. But it's a Church-Turing device - so it could be programmed to emulate arithmetic. Given that WAS a computer, you have to allow that performing arithmetic (in hardware) is not a requirement to make a computer. In this case, purely boolean logic operations were implemented in hardware.

So, I repeat - what is your definition of a computer? I don't think "Anything that some judge in a district court says is a computer" is a sufficient description. SteveBaker 18:30, 11 March 2006 (UTC)

Okay, now that this is on the correct page for this discussion, I'd like to try to point out that the issue is that the usage of the term "computer" is somewhat inconsistant on WP and needs to be discussed. User:Cburnett, how would you propose we define "computer" here, and how should we apply it to various early computational devices? User:SteveBaker has more or less suggested we use the Church-Turing thesis as the benchmark for what a computer is. While I personally believe this is pretty accurate for the modern usage of the word "computer," Cburnett has correctly pointed out that earlier "computers" fail this test but are still largely considered to be computers. I would like to add to this discussion that in every dictionary and encyclopedia I've looked up "computer" in, the introductory simple definition always includes programmability or program execution or instruction sequence execution as part of the fundamental definition for "computer."

Here's a list of points that we should try to reconcile:

The definition of "computer" has largely changed since the days of the ABC, ENIAC, etc; I believe we all agree on that.
In modern usage the term "computer" is never applied to any non-stored-program device. I can't think of any notable counterexamples, but if you have any they would be crucial to this discussion.
The former usage of "computer" is more nebulous. Programmability was not part of the definition and thus we have devices like the ABC that were called computers, but were neither general-purpose nor programmable by simple means (that is, those that don't involve rebuilding/redesigning the computer).
We ought to have some consensus on how we specifically apply the term "computer" to these early devices to avoid this sort of discussion in the future.

Please respond to these points, add more if you feel the need, and invite other users to discuss who might be interested (and competent). Please resist the urge to totally dismiss the other person's point of view, because we have (IMO) all made valid points and there is no need to slap labels on people. -- uberpenguin 18:32, 11 March 2006 (UTC)

Hi guys. Great discussion! My two-penn'th:

Language is a living thing that changes and evolves totally as societies change and evolve. Therefore we can clearly distinguish the 1940s common usage of a highly technical and socially relevant word like 'computer' from its modern common usage. Clearly we need to do so in articles about 1940s 'computational' technology (like the ABC one). We can also be wary of 1973 court judgements[4] over the meaning and applicability of this word now that another 33 years have passed, during which 'computers' (both the things and the word) have entered many more billion peoples' everyday lives, and evolved so much in themselves too.
In modern usage, there is no doubt - as per Uberpenguin's dictionaries and encycloediae, as well as the current article's first sentence - that today, both common and technical usage of the word implies, as he says, programmability or program execution or instruction sequence execution. Maybe we should warn readers of the Computer article that this was not always the case - then briefly mention the ABC, human computers etc. But that's as far as that goes - a warning to the reader not to let the evolution of our language confuse him or her.

Job done? I hope so. BTW, FWIW, I have quite a lot of academic qualification in this subject area and every day I develop software, self-employed, for a living. On the other hand, I have no connection with either the Atanasoff or the Berry families, and have never taken anyone to court over anything. --Nigelj 19:34, 11 March 2006 (UTC)

I have never disagreed or countered that the common usage of computer is that-which-sits-on-my-desk-and-runs-solitare. Not once. Do I have a problem with computer talking about the von neumann arch as "a computer"? Nope. What I would change is that the word "computer" means the modern computer of a stored-program machine and the whole turing test bit. I think it would be wholly negligent to ignore the historical defintions and usages of the word. WP doesn't exist to redefine words and since applying new definitions to words is considered original research: the practice is not allowed on WP.

What I have countered all along is renaming or redefining the AB Computer as something other than a computer. Every thing I have read that discusses anything about the ABC does not attempt to counter its name: AB machine, AB toaster, etc. etc. They all refer to it as the AB Computer. I have not seen one verifiable source that discusses the definition of computer and that discusses why the ABC is not a computer. A dictionary definition, in this case, does the discussion a disservice by using an extreme few words to define it. Java defines the language as "an object-oriented high-level programming language" but I would wholly reject such a simplified definition to describe what the Java programming language really is. It's too simplified, just like dictionary definitions of "computer".

Whatever this discussion leads to won't change the 1973 court result and it won't change many publications by the IEEE and various professors as calling the ABC a computer. In the end, Wikipedia:No original research demands verifiability and I believe I have pulled sufficient number of sources to show that the ABC is a computer. Only when a verifiable source can be shown (a dictionary is not a primary source) that the ABC is not a computer and that the turing test is required of a device to be labeled as a computer can WP redefine the definition to such. So: where are these verifiable sources? Cburnett 04:30, 12 March 2006 (UTC)

The problem with Cburnett's position is that the ABC was called a computer in the 1940's and none of us debate that - but if it were created as a small electronic gizmo with the sole function of solving equations today, it would not under any circumstances be called a computer. We'd treat it like we treat a fancy calculator. So in what modern sense can we call it "the first digital/electronic/whatever computer" - when we don't think it's a computer at all by the modern meaning of the word computer? Does this mean that something that back in the 1940's might reasonably have been called "the worlds first computer" has to lose it's title? Well, yes - I'm afraid it does. The meaning of the word has changed. We can't possibly keep on telling people something that is no longer true without heavy qualification. I've been wracking my brains for another word that's changed it's meaning - but this is still the clearest I can come up with: Back in the 1940's the word "Gay" meant "Happy" - but here in Wikipedia, we don't go around calling people of the 1940's who were happy heterosexuals "Gay" - because it would totally give the wrong impression. No matter that a book written in 1940 can be found as a concrete reference that 'proves' that John Q. Hetero was "Gay" - he's STILL not gay by 2006 standards. I think that's a reasonable analogy for the problem we have here. The 1973 court case was NOT about "Is the ABC a computer or not?" - it was about "Did some guy infringe on some other guy's patents?" --and he did - but that proves nothing about what the meaning of the word "computer" is in 2006. What further confuses matters is Cburnett's insistance that any device that 'computes' is a 'computer'. Well, OK - but that's just a circular argument - what does the verb "to compute" mean? Let's put up some new references - things on the web we can all read and compare:
- The 'Definitions' section of "www.computeruser.com" says: "Definition for: computer - An electronic device that has the ability to store, retrieve, and process data, and can be programmed with instructions that it remembers."
- "searchwinit.techtarget.com" -- "computer - A computer is a device that accepts information (in the form of digitalized data) and manipulates it for some result based on a program or sequence of instructions on how the data is to be processed."
- http://www.webopedia.com/TERM/C/computer.html - "Computer: A programmable machine. The two principal characteristics of a computer are:
  1. It responds to a specific set of instructions in a well-defined manner.
  2. It can execute a prerecorded list of instructions (a program).
- http://cancerweb.ncl.ac.uk/cgi-bin/omd?query=computer&action=Search+OMD: "computer - A programmable electronic device that can be used to store and manipulate data in order to carry out designated functions; the two fundamental components are hardware, i.e., the actual electronic device, and software, i.e., the instructions or program used to carry out the function."
- The Free On-line Dictionary of Computing - "<computer> A machine that can be programmed to manipulate symbols."
- The Computer desktop encyclopedia: "computer - A general-purpose machine that processes data according to a set of instructions that are stored internally either temporarily or permanently."
- University of Colombia Press Encyclopedia: "computer, device capable of performing a series of arithmetic or logical operations. A computer is distinguished from a calculating machine, such as an electronic calculator, by being able to store a computer program (so that it can repeat its operations and make logical decisions), by the number and complexity of the operations it can perform, and by its ability to process, store, and retrieve data without human intervention."
That's just the first half dozen Google hits...there are a bazillion of them and they pretty much all agree that PROGRAMMABILITY is a key thing.
Is a dictionary a primary source? Well, I wouldn't quote a dictionary as a source of information about the ABC (for example) - but when I'm asking about the definition of a word (which is truly what this debate is about) then where else but a dictionary?

SteveBaker 05:48, 12 March 2006 (UTC)

Just to add my 2 cents, nobody here would disagree that coverage of special-purpose computing devices is an important part of the Wikipedia. It is, and they are indeed covered in much detail. It is quite appropriate to refer to these devices as part of the historical context of the history of computing, which the article does. And the article discusses briefly (the history of computing hardware discusses in much more detail) about how the key feature - universal programmability - of the modern computer evolved gradually. However, for modern readers, we should reflect modern usage, and primarily discuss the digital, Turing-complete, programmable computer in this article. If that happens to mean the ABC falls outside that definition and is intead viewed as a fascinating step along the road to it, well so be it. --Robert Merkel 09:51, 12 March 2006 (UTC)

Here are a few other words that have changed their meaning over time. The word 'computer' is no different: In 1940, and maybe, in one judge's opinion, still in 1973, the word could reasonably be used to include the ABC. But not any more, apart from as a historical footnote to the modern concept, with a link to its own article.

link: chain link, conceptual link, hypertext link
server: one who serves (servant), waiter or waitress, centralised computing resource
nice: you have to look back a few centuries to see this one
automobile: steam, coal gas or petrol powered, sir? Very few cars from the 1900s and 1910s could legally be registered as such now.
gay: as above
net: fishing, internet
web: spiders, www
chat: natter to IRC
text: now a verb and applied to mobile phones
cool: chilly, good
mistress: householder's wife, long for Mrs, woman married man is having affair with

I'm not a linguist, and I'm sure there're plenty of better examples. I've also muddled words with changed meanings with words that currently have multiple meanings. This doesn't harm the argument as some of those multiple meanings will undoubtably fade into disuse within another generation or two.

The word computer began life with its literal meaning, 'one who computes', then began to get applied to various electronic inventions by analogy, but has now settled down to a precise and more restricted meaning - the digital, programmable thing you have on your desk, or in your server room - not your calculator, not your mobile phone (although these may also have tiny computers inside them, running some software to help them work), but the computer - the thing that runs your software. This is not pedantry, it's perfectly normal, everyday speech. To try to deny it or alter it is impossible, or at least is an attempt at social engineering - WP:OR doesn't even come close to describing such an attempt. --Nigelj 12:18, 12 March 2006 (UTC)

Tis funny, since my sources (the kind that give more than a definition) still label the ABC a computer. Here I thought I was the one using resources to keep Steve from redefining computer and doing original research. I haven't seen him produce anything much beyond a dictionary definition for his position (you know, professional organizations, professionals that deal with this stuff, federal courts, text books on the topic, etc. etc.). Nothing. And I'm the one doing social engineering and original reserach??? Holy crap it makes my brain hurt. Cburnett 15:11, 12 March 2006 (UTC)

How can you say I'm redefining computer? I'm just giving you it's modern definition. I didn't redefine anything - it redefined itself. This should come as no suprise because words do this all the time. I didn't do original research - I looked up the meaning of computer in a gazillion places and then I read everything I could find about the ABC. From those readings, I find that the modern definition that every reasonable resource provides for the word computer says - typically in the very first sentence - that a computer is a PROGRAMMABLE device. Now, I look at every reference for the ABC and find that it is most definitely not programmable. What other conclusion can I possibly come to? You have taken a different approach to answering the same question. You find some references that say that the ABC is indeed a computer - but one of those is a court case - and I don't think we regard Judges as experts on modern technology - the other is an IEEE publication - Computer Organization & Design by Patterson & Hennessy (ISBN 1-55860-490-X Parameter error in {{ISBN}}: checksum) ...unfortunately, none of the online ISBN lookup services can find this book (check, for example http://isbn.nu/155860490X ) SteveBaker 17:13, 12 March 2006 (UTC)

UPDATE: You have the wrong ISBN number - it's ISBN: 1558606041 - when I'm next at the library, I'll check it out. SteveBaker 17:54, 12 March 2006 (UTC)

We've been talking about this court case in 1973. 1973 was a long time ago in terms of the life of computers - and perhaps some of us here don't have a feel for the environment that Judge was making his decision in. I was in high school in 1973. Let me relate how the general public (including Judges) saw computers back then. We calculated using slide rules and thick books full of log tables. In the early 1970's (I don't know the exact year), my parents spent a LOT of money on buying me a TI pocket calculator - not only was I the first person in my entire school to have one - but nobody else had even seen a pocket calculator up close (I've always been an early adopter). It was a total marvel - my math teacher practically had tears in her eyes when she found she could show a mathematical series converging by pushing four buttons per step and seeing the numbers get closer and closer to the ideal value. We learned Fortran in 1973 (we were probably the first school in England to teach computers) - writing our programs out on 'coding forms', mailing them to the regional computer center where they punched them onto cards and slipped our programs in at the end of the payroll calculations on the midnight shift. We got our printouts back two weeks later in the mail. That was as close as 99.99% of the population ever got to a computer. In 1974, the computer we used in college was a Singer mainframe - it had no concept of subroutine calls - there was no stack, it had conditional jump instructions - but they could only jump forwards down the instruction stream, it had a 24 bit word and packed characters into 6 bits (no lowercase!). We stored programs on papertape and used Teletypes to enter them. Things were primitive back then - everyday people had NO IDEA what computers were. Language was different - Judges were old guys who had only gotten where they were by being on the legal circuits for decades. To regard a judge (ruling in 1973) to be an expert witness about what a computer is considered to be in 2006 is sheer lunacy. This is why that court case is worth nothing in deciding what is or is not a computer in a modern context. SteveBaker 17:49, 12 March 2006 (UTC)

OK, Steve. I think we have a clear consensus here, including everyone except Cburnett, and I for one am starting to doubt his good faith and agenda. Have you seen how long this discussion has dragged on for? Over 53 KB in the last two sections here! I think if Cburnett keeps on, ignoring our points and repeating his same intransigent position, now with mounting aggression ('And '''''I'm''''' the one doing social engineering and original reserach??? Holy crap it makes my brain hurt'), we may begin to regard it as trolling (WP:DFTT). Notice how he totally ignores my posts, even my existence, and keeps going for you by name? He must reckon you're more fun than me! The WP policy is 'Don't feed the trolls', so let's just get the article sorted out, IMHO. If we have trouble with endless reverts, we can escalate that. There's more important things to do than endlessly argue this dead point, I think. --Nigelj 20:18, 12 March 2006 (UTC)

That's fair comment. I will cease to feed the troll. SteveBaker 20:59, 12 March 2006 (UTC)

I think the label "troll" is far to harsh and is unnecessary. While I disagree with ~~Cbaker's~~ Cburnett's position and his defense of it, he certainly has some valid points. I do have to agree that the consensus here is against Cburnett, however, and that there is little point in debating this further amongst ourselves. I think we should move on to proposals for reworking the intro of this article somewhat to explain that the term's meaning has changed since the advent of the stored program computer. -- uberpenguin 00:56, 13 March 2006 (UTC)

(Cbaker??!) Actually, I read the Wikipedia page about "Don't feed the troll" - and you are quite correct, Cburnett's comments are not trolling according to the Wiki definition. But still - I think we are better to just fix the problems and end the debate here. SteveBaker 16:45, 13 March 2006 (UTC)

Whoops. My mind must've been somewhere else, I fixed my mistake. -- uberpenguin 16:57, 13 March 2006 (UTC)

I agree. The general field of devices that perform computation was and still is broad. After decades of use, we now know what "A Computer" is. The ABC was not a computer, in spite of being named one. Most people are unwilling to take a stand against the ABC marketing campaign. I applaud your effort. --Zebbie 05:43, 14 March 2006 (UTC)

Featured Article Candidate?!?

I see someone has nominated this article for FAC. It would have been nice to have discussed it here first. I have added a strong oppose comment in the FAC discussion page - this article is a million miles away from being a good article on the general subject of "Computer". I can pick almost any sentence out of it and find a reason why it's too specific, inaccurate or just downright wrong.

We have a VERY long way to go to reach the dizzy heights of FA.

Sorry.

SteveBaker 03:16, 7 April 2006 (UTC)

Example: I'll take one of the smallest sections of the article: "Instructions" - let's go through claim by claim:

The instructions interpreted by the control unit, and executed by the ALU, are not nearly as rich as a human language.
By what measure? Are you comparing a single machine code instruction with a word? a sentence? a phoneme? What on earth do you do to measure the "richness" of a language. Saying "The cow in that field is happy" in x86 machine code is indeed very tricky - but transcribe any C++ function into English - and if it involves abstract base classes, templates, exceptions, etc, you'll end up with a 20 page English description of a one page function. So which language is "richer"? This sentence is bullshit.

A computer responds only to a limited number of instructions, which are precisely defined, simple, and unambiguous.
Could the computer respond to an infinite set of instructions? No - of course not. Are they "simple" - well, by what standards? Yeah - they are probably not ambiguous...but the rest of this sentence has zero information content.

Typical sorts of instructions supported by most computers are "copy the contents of memory cell 5 and place the copy in cell 10", "add the contents of cell 7 to the contents of cell 13 and place the result in cell 20", "if the contents of cell 999 are 0, the next instruction is at cell 30".
This gives the strong impression that the instruction contains the addresses of the memory cells...which is true only of a very limited class of computers. In many simpler computers, you can only do register-to-register addition - in others, you can't have two addresses in one instruction - so copy cell 5 into cell 10 needs two instructions. Do "most" computers support those instructions...well, maybe. I could maybe live with this sentence - but I don't like it.

All computer instructions fall into one of four categories: 1) moving data from one location to another; 2) executing arithmetic and logical processes on data; 3) testing the condition of data; and 4) altering the sequence of operations.
Sorry - but what about resetting a watchdog timer or switching an interrupt priority, putting itself into a SLEEP mode to save power...there are LOTS of instructions that fall beyond that set of four things. There is no clear distinction between doing arithmetic and doing logic - that's an arbitary distinction based on the fact that 'arithmetic' was invented before binary logic operations...and comparisons are just arithmetic by another name...not a separate class of operation. This sounds like it was written by someone who has a week of machine code classes - not someone who actually knows what makes up a computer.

Instructions are represented within the computer as binary code — a base two system of counting.
Nonsense - many older computers used base 3 or base 10.

For example, the code for one kind of "copy" operation in the Intel line of microprocessors is 10110000.
"The Intel line of microprocessors" - Intel make MANY kinds of microprocessor. The instruction for COPY on an 8048 is different from a Pentium.

The particular instruction set that a specific computer supports is known as that computer's machine language.
...or it's "Microcode".

To slightly oversimplify, if two computers have CPUs that respond to the same set of instructions identically, software from one can run on the other without modification.
slightly oversimplify?!? - try running your copy of WORD on your Linux x86 PC - or on an x86 Mac! More fundamentally, the memory configuration, I/O locations, firmware...lots of other things determine software portability.

This easy portability of existing software creates a great incentive to stick with existing designs, only switching for the most compelling of reasons, and has gradually narrowed the number of distinct instruction set architectures in the marketplace.
I disagree. Many video games have been ported between Playstation, GameCube and Xbox. Those three systems have TOTALLY different CPU architectures. What makes the software portable is that the programs are written in a high level language. If you write in JAVA (say) your program will run OK on a huge range of CPU's. Portability isn't much to do with low level machinecode any more.

So - in just one section of this article, I can find very good reasons to disagree with every single sentence. This is true of nearly all of this article. That might be good enough for a run-of-the-mill Wiki article - but it's nowhere close to the standard of a Featured Article.

SteveBaker 03:38, 7 April 2006 (UTC)

Fft... You didn't even need to justify why this article isn't nearly ready for FA. I had the exact same reaction as you upon learning this. I'll add my oppose as well. This article needs a lot of work before it can be called "good". -- uberpenguin @ 2006-04-07 03:40Z

This just underlies the bigger problem - not enough references. The section you use as an example above contains no references - if someone were to start adding references to it, they would likely see the problems you mention. Pagra shtak 03:44, 7 April 2006 (UTC)

Which in turn underlies a yet bigger problem. This article is a 'big topic' article - like Physics or Mathematics. It can't be written by just talking about the PC on your desk at highschool level. That stuff can go down in subservient articles (such as Central processing unit. This article needs to make statements that are equally true of ENIAC, the Intel 4004, the Pentium, the Cell processor in the PS3, the pneumatic computer made out of Lego, the experimental DNA-based computers and the Quantum computers. It has to think about gigantic supercomputers, multiple-CPU machines...all the way down to the lowliest microcontroller. It can't assume that all computers are binary - or even that they all do arithmetic. It has to be very careful when talking about memory - are we talking about registers, cache, RAM, ROM, Flash, Magnetic disks? It mainly has to speak in extremely broad generalities and point the reader to more detailed articles on history, modern computers, quantum computing. What we have here needs to be completely thrown away and we need a clean start. SteveBaker 03:55, 7 April 2006 (UTC)

Go at it, then. Computer/Rewrite. Call us when you've got something better. You'll find it a lot harder than you presently think.

Just to take up one specific point that was raised, you seem to be exhibiting the classic computer scientist's nitpickery. Whether instructions are register-to-register, register-to-memory, memory-to-memory, stack based, or using some other kind of addressing scheme is really irrelevant to the point that was being made, which was to give an approximate insight into what individual machine language instructions are like for somebody who knows SFA about computers. Which is, of course, who this article is for, not for demonstrating how much detail the writers happen to know about the inner workings of CPUs.--Robert Merkel 04:13, 7 April 2006 (UTC)

Sorry - but at what point did I say it would be easy? How can you possibly presume to know how hard I personally think it would be? To the contrary - I've been lurking around (mostly fixing the eternal vandalism that plagues this article) trying to figure out what to do about it. I think it's going to be very hard indeed to rewrite this in the manner it deserves...which is why I havn't done it (yet!). However, difficulty-of-fixing does not a featured article make. If it were merely adequate to write a half-assed article because a good one was too hard - then Wikipedia would be virtually useless. Remember - I didn't start complaining until someone pushed it into the FAC limelight.

As for nit-picking. There is a really good reason why computer scientists are nit pickers. We live and die by writing things extremely concisely. When you are in the middle of a million lines of C++ code (as I am in my daily job), you simply cannot afford vagueness - that one stoopid little 'nit' that didn't get picked is the one that'll crash the computer the very moment a customer steps in front of it! However, having seen how the English majors here on Wikipedia can pick an article to death - I'd say that we geeks are not the worst culprits! :-) Precision is very important in an encyclopedia - and I make no apologies for demanding it. Besides, the point of my previous tirade was not to attempt to find specific things that need fixing. I was attempting to convey the magnitude of the problem by picking one section more or less at random and examining it critically. The vast majority of the statements made are false - or at least strongly debatable. We could argue about that one specific point out of the NINE statements I analysed. But even if I concede that I was wrong about it - what remains is that 8 out of 9 sentences I examined were faulty - you can probably talk me out of one or two more if you work at it (please don't!) - but still, that's a mighty pile of incorrect statements for an encyclopedia entry. Heck, if even 10% of the statements were faulty we'd have a major problem! SteveBaker 04:55, 7 April 2006 (UTC)

I agree that a couple of statements there are flat-out wrong; the Intel one is a good example. Most, however, are hand-waves that could be debated endlessly in isolation but serve to illustrate a broader point in an accessible manner. And, frankly, I don't believe anyone can write an acceptably short and accessible article about such a broad topic without using such handwaves. It's exactly the same as an article on, say, "the history of the world" - there are inevitably going to be one-sentence summaries of things that somebody with expert knowledge of that area will vigorously dispute.

If you can come up with an article that uses a better set of handwaves, I will be very impressed; wiki barnstars all round. But what I will *strongly* object to is what I fear you're likely to come up with; in your efforts to avoid handwaves you end up with an article that doesn't actually convey any useful information to somebody unfamiliar with the topic. --Robert Merkel 05:08, 7 April 2006 (UTC)

This dialogue is irrelevant. The point is that the article is nowhere near FA status. End of story. Continue this conversation again when someone gets around to actually rewriting the thing. -- uberpenguin @ 2006-04-07 05:12Z

Also, regarding the references concept. Refs aren't quite as necessary when most of the people watching the article have a good idea of what they are talking about. For the general structure and scope of the article it's fairly impossible to find a few references that magically justify what you are asserting. References really become useful on contentious points or specific details. -- uberpenguin @ 2006-04-07 05:15Z

I'm merely trying to point out some considerations if SteveBaker wants to attempt a rewrite. By the way, looking at that specific section of the present article, it's probably the worst of the lot. That specific section could do with some attention. --Robert Merkel 08:28, 7 April 2006 (UTC)

The attention it just got is making matters worse. Whoever wrote the sentences on Neural Networks and Quantum computing clearly doesn't know anything about either of them - and didn't even bother to read the Wikipedia articles before writing about these topics! I can't imagine why you'd possibly want custom hardware neural nets for spam filtering - that's the kind of thing that you can do with neural network software (but that's not how it's currently done - Baysian filter is nothing to do with neural nets). As for Quantum computing processing information that can only be described in higher mathematics?!? Where did that come from?! One excellent (theoretical) application of Quantum computation is in factoring large numbers efficiently. The data is simple integers and the process is just factorization - no 'higher mathematics' in sight there! Quantum computers (if they ever become practical) will be mostly distinguished by the fact that they can use superposition and entanglement to perform a vast number of calculations in parallel. Please - if you don't know a lot about a subject, don't write about it - at least until you've researched enough to know the basics. SteveBaker 17:23, 7 April 2006 (UTC)

If a contribution doesn't improve the article, just revert it. And if you leave this article for a week and it's in a worse state than it was, revert to a version from a week ago if necessary. Don't tolerate poor-quality contributions. -Robert Merkel 00:42, 9 April 2006 (UTC)

Removal of "good article"

seeing as many of us have reached the conclusion this is no longer a "good article", perhaps we should remove the tag? Vulcanstar6 04:09, 7 April 2006 (UTC)

i figure i should ask you all this time, :P Vulcanstar6 04:14, 7 April 2006 (UTC)

Well, GA doesn't actually mean a whole lot - all it takes is one person to promote the article or one person to delete it. The tag says that the article achived GA status - which it did - it's a fact and it bears asserting...even if the presence of the tag merely confirms some people's opinions of the worthlessness GA process. So - I say leave the tag there as a warning to those who think it's worth something!...but YMMV. SteveBaker 04:34, 7 April 2006 (UTC)

ENIAC == decimal

I added a (differently worded) comment about ENIAC as decimal. I am in the "ENIAC as first (insert definition) computer" camp, but I think that since the argument is often over "first computer" vs. "first reprogrammable computer" vs. "first electronic computer" vs. "first elecronic digital computer" vs. "first binary electronic reprogrammable digital computer", a brief mention (not as a limitation, but simply a distinction to the systems that used binary) is in order. -- Gnetwerker 17:29, 7 April 2006 (UTC)

Without trying to stir up the ABC hornet's nest again, the problem here is that any computer you can think of can be awarded the prestigious title "First (insert definition) computer" for some definition of the word "computer". However, in the context of this article, we have a definition of the word that describes what this article is about - and for that definition of the word (which I believe is the true, modern definition), it doesn't matter whether the computer is binary or decimal. Given that, I think it's VERY important that we mention that the ENIAC was a decimal machine because it's something that Joe Public may not be expecting - many people (I'm sure) belive that all computers are binary and digital - they are suprised to hear that there were decimal computers - and computers which didn't use two's complement for negative numbers - or which had 6 bit bytes...so pointing out those weirdnesses is a vital thing to do. I am coming to the view that we should perhaps move all of this "First computer" stuff off into another article. In the Wikiproject Automobiles (which has a lot of similar "First" problems), we have List of automotive superlatives. That article has the space and organisation to list all of the things that might make a claim to be "First" or "Best" or "Most" at something - it also has "Honorable mention" catagories for thing like "First ever mass-produced car" which is generally considered to be the Model T Ford - but which honor probably belongs to an Oldsmobile. If we had that for computers - and referred to it from the main articles whenever a "First" type claim needs to be explained - then we can get rid of all of this clutter and get down to the important issues of describing what a computer actually IS. SteveBaker 14:45, 9 April 2006 (UTC)

That's sounding good, Steve. Now I don't want to stir things up either, but can we just forget analogue computers here? These use(d) op-amps and capacitors directly to model and solve differential equations. They were very important in helping engineers understand all kinds of damped oscillatory systems like shock-absorbers in car suspension systems and automatic control loops for heavy things like gun turrets and slow things like furnaces. Maybe, as is currently going on in solar panel, solar cell etc, this, top level page could become more of a disambiguation page with a very general overview illustrating the range of meanings the word currently has, and has had in the past - then links to all the specific pages, like history of computers, computer architectures, personal computer, mainframe computer, supercomputer, analog computer, embedded computer, microcomputer etc. (Interestingly, I just typed all those article names off the top of my head, and now I see they're all blue. I haven't gone to read them all yet!) --Nigelj 15:40, 9 April 2006 (UTC)

- Yes - analog computers don't fit our definition because they aren't programmable in the normal sense of the term - although there were hybrid computers (used extensively in my field of Flight Simulation) that most definitely were programmable. However, I would be happy to accept an argument that those were basically digital computers with an analog "computer" as a peripheral. Those were still in service at NASA as recently as 8 years ago. SteveBaker 15:44, 9 April 2006 (UTC)

Move page

Hi all. It seems to me this page is slightly off topic. It describes digital computers. I propose to move this page to digital computer, and to but a disambiguation page called computer pointing to stuff like:

Analog computer
Digital computer
DNA computer
Turing machine (or perhaps a page listing all main mathematicals models of computers)
Molecular computer
Quantum computer

Powo

I concur that the article is too tilted towards the digital computer design, which then means that WP is contributing to the common misconception that digital computers are the only kind of computer. But I also think a mere disambiguation page would be too intimidating for WP users who do not understand computer science, especially young children and senior citizens.

Looking at the article, it seems to me that the first third of it is already written in a hardware-neutral fashion, or could be easily adjusted to meet that standard. It is the second half that probably could be split off into a separate digital computer article. Then the first one could end in a list of See also links, with digital computer at the top of the list as the most familiar kind of computer. --Coolcaesar 15:35, 23 April 2006 (UTC)

Not strictly so, I don't think, Coolcaesar. As far as analog computers go it falls down in the first sentence and continues to do so all the way - there is no 'list of instructions'. I don't know enough about DNA, molecular and quantum computers to comment, but I think 'stored programs' are generally a bit thin on the ground outide of the digital/Turing model. On the other hand, I agree that, judging from the vandalism and graffiti, most kids and their grannies think that this article is (or should be) about the thing on their desks and nothing else. But then isn't that the purpose of an encyclopedia, to help enlighten them? Maybe the best answer is a new introduction about how complex the concept is, followed by brief but descriptive link-sections to all the options? --Nigelj 16:09, 23 April 2006 (UTC)

I think that there are two issues here:

Any article with the name 'Computer' covers a VAST amount of terratory. I think we should treat it like the page called 'Physics' (PLEASE go read that to see what I mean). It should be a broad-brush introduction to the subject in general and should consist mostly of a description of the topic leading to a well-organised set pointers out into the rest of Wikipedia. The present article is about a million miles from being that.
An article about digital/electronic computers such as are on people's desks is needed - and this one is a good starting point.

So, yes - I guess I agree with moving this article out of the way and getting started on a proper "Computer" article. We really need to ask why someone would come to Wikipedia and type 'Computer' into the search box. Anyone who does this is (by definition) using a computer to do so - and knows (at least) how to use the most basic functions. So they CERTAINLY aren't coming here to find out what a computer *is*. In all likelyhood they are after more detailed information about a tiny, specialised area of computing and simply assumed they could find it by going to 'Computer' and following links from there. That's why we need to do this like the Physics article does it. SteveBaker 23:47, 23 April 2006 (UTC)

Analog computers are an interesting historical artifact. Molecular computers (as I understand the concept) are just a reimplementation of Turing machines on a smaller, faster scale, and are research projects. Quantum computers and DNA computers are moderately interesting research projects. Why complicate matters unnecessarily? A paragraph headed "alternative computing models" is all that's required for the last three. --Robert Merkel 11:43, 24 April 2006 (UTC)

Minor remark: I am not a specialist, but molecular computers are not at all a reimplementation of a Turing Machine (are you thinking of nano-computers?). A molecular computer is a massiviely parallel computer taking advantage of the intrinsect computational power of individual molecules. A "computation" of such a molecular computer could, therefore, happen in a test tube! Not much of a Turing machine (although not more powerfull from the computability point of view, of course...) Regards, --Powo 13:03, 24 April 2006 (UTC)

I mostly agree with Steve and Robert. I think that the label "analog computer" is somewhat odd and subjective in the first place. One could easily argue that an (analog) oscilloscope is a type of "analog computer" when you consider exactly what devices explicitly called analog computers did.

While I certainly understand the difficulty with placing a precise and useful definition of computer (that is, not fun rhetoric like "a computer is that which computes"), I find myself in the position that the usage of the term computer has come to very nearly exclusively mean "stored program computer" in modern usage. I think this article should reflect modern usage and primarily focus on this type of computer with just minor mention of historical computers and links to appropriate articles.

In all honesty, there is a lot more to be said about digital electronic computers any way. Since the advent of the stored program computer in the 1940s, there hasn't been much significant effort to design computers outside this paradigm. As Robert said, quantum, DNA, and molecular computers are all research topics and not yet things that inundate every facet of modern society as stored program (micro)computers are. -- uberpenguin @ 2006-04-24 15:27Z

Analog computers per-se have very little to do with our definition of 'computer' as something programmable (although they are typically reconfigurable...which is not quite the same thing). What DOES deserve further discussion is the hybrid computer - which was a mix of digital and analog circuits. The programmability of these machines is just like a modern digital computer - but the actual calculations would typically be done with analog circuitry. These are genuinely programmable 'true' computers - but with analog ALU's. SteveBaker 03:33, 25 April 2006 (UTC)

Alternative computing models...

So does anybody want to write a paragraph or two about the alternative computing models discussed above? --Robert Merkel 00:15, 27 April 2006 (UTC)

computer

i have read this article and found the information very usefull thank you for this has it helped me with understanding the computer industry more

Jacquard

I was surprised to see the Jacquard loom not mentioned in the history section of the computer article. I don't know if it was left out by design or overlooked. I've always had the impression that the Jacquard loom was considered a key development in digital computers. I added a quick mention just before Babbage. Rsduhamel 03:28, 7 May 2006 (UTC)

Good addition; it's a bit of a judgement call on what things to include and what to leave on in that section, but you're right that the Jacquard loom was a pretty important example of precursor technology. --Robert Merkel 07:10, 7 May 2006 (UTC)

The three pieces of the puzzle of modern computers are the arithmetic part, the programmability part and the storage part. We have lots of stuff about calculators and other arithmetic devices. The Jaquard loom (which is undoubtedly an important step on the programmability side) is a valuable addition. Thanks! What do we need to say about storage? We don't mention mercury-acoustic memory, drum storage, the Hollerith card punch system...there is a lot left to say! SteveBaker 15:24, 7 May 2006 (UTC)

Illustration

The "illustration of a modern personal computer" looks nice, but does not add any information to the article. Either an actual picture or a diagram with more information (such as text labels that identify the different parts of the machine) would be more useful. -- Beland 04:59, 7 May 2006 (UTC)

It's an introductory image. It really doesn't have to do much other than provide a common example of the subject. Besides, I much prefer a nice diagram to one of the umpteen pictures of peoples' PC desks that creep into this article from time to time. -- uberpenguin @ 2006-05-07 21:27Z

I don't really like it either - it's a zero-information-content image. We should find a photo of a more unusual computer - or perhaps a diagram with the parts labelled or something. But I really don't see the point of showing a picture of a PC because it's almost 100% certain that the person reading the article is sitting right in front of a PC at the time they are reading it! We need a photo that screams "Computer" without being a PC. SteveBaker 01:12, 8 May 2006 (UTC)

I'm definitely not liking having (copyrighted) images of two PCs in the intro. If we MUST include a PC, at least show some other form of a computer as well. How about a mainframe like a S/360? A midrange like a PDP or AS/400? An embedded computer? Perhaps something older like the Harvard Mark I? There are far more interesting and unusual (and free) pictures of all of those systems available on Wikipedia. I hate the idea of both intro images being PCs, partially because they are so ubiquitous and I think we could do better on the 'interesting' front than a picture of a box+monitor+keyboard. -- uberpenguin @ 2006-05-25 17:20Z

I agree. It's certainly ridiculous to use a copyrighted image (and especially one that has Dell promotional stuff on the monitor screen) - when it's so easy to take a nice photo of a PC that doesn't carry any copyright or advertising. But still (as I've said before) BY DEFINITION: EVERYONE WHO CAN READ WIKIPEDIA KNOWS WHAT A COMPUTER LOOKS LIKE BECAUSE THEY ARE USING ONE RIGHT NOW!!! So this is a pointless choice of image. It's not informative (to our readership) and it's not particularly interesting or beautiful - so we might as well pick something else. I think we need to carefully explain that not all computers look like PC's. We should pick the computer from a washing machine or a microwave oven or the one in your car or an old Sinclair Spectrum or a PDP-11 or a Lego RCX computer...anything other than a modern laptop or deskside computer because that is totally uniformative. SteveBaker 18:55, 25 May 2006 (UTC)

There's always this interesting image of a PDP-8/I's guts. I doubt that one will go over well, though, since I imagine we want the intro photo to be of the "outside" of a computer (course, using a picture of an embedded computer trumps that...). I'm open to suggestions, but please, let's not keep the PC double-whammy in the intro. -- uberpenguin @ 2006-05-25 19:35Z

I'd suggest a picture of a microcontroller on the basis that they are a complete computer, but not as the reader may know it. --Robert Merkel 07:21, 27 May 2006 (UTC)

A microcontroller or SoC would be fine by me, but can we do better than just showing an IC's packaging? I think that might be a tad bland for an intro picture (though not as bland as a desktop PC). -- uberpenguin @ 2006-05-27 07:38Z

How about a Lego Mindstorms controller? --Robert Merkel 07:42, 27 May 2006 (UTC)

I'm for it... It's a fairly identifiable object that may not normally be thought of as a computer. Seems appropriate. Do you have a particular picture in mind? I think it would be best to have a disassembled view of it so it becomes apparent on first glance that there are some microelectronic guts in the thing. -- uberpenguin @ 2006-06-04 23:30Z

Your wish is my command! I took a photo of my RCX 1.0 computer - then dismantled it and took another - mushed the two photo's together so you can see the insides and the outside. Took me 10 times longer to get the darned thing back together again - but after three tries, it works again! Hope this one is to everyone's final satisfaction. SteveBaker 01:21, 5 June 2006 (UTC)

Lovely. I think that's a great picture for this article's intro. -- uberpenguin @ 2006-06-08 05:23Z

I disagree. This article is about the computer: arguably the most important technological achievement of the 20th century. Is a Lego toy really an appropriate ambassador of this? (It's better than the Furby, granted). There are computers out there doing all sorts of amazing things, surely we can come up with something photogenic that does them justice? 213.38.7.224 13:38, 3 July 2006 (UTC)

"Is a Lego toy really an appropriate ambassador of this?" -- Yes, totally appropriate. What better way to illustrate how computers have infiltrated almost every facet of our lives? I think having a full microcomputer in a toy brick is every bit as impressive as any picture of a SoC die we could slap up there. -- uberpenguin @ 2006-07-03 14:07Z

The Lego computer is a complete computer - an SoC die really isn't so it should be excluded on that basis (also, the current photo has an SoC die slap in the middle of the circuit board). If someone has a better photo of a complete computer (NOT A PC OR A MAC!) that has as more to tell the reader than the Lego RCX photo - then I'm happy to consider it. But three out of the four people who contributed to this discussion so far liked the idea of the RCX photo (one person proposed the idea, a second person agreed and took the photo and a third person said it was a great choice, the fourth person didn't say they don't like it) - so far, we seem to be keeping everyone but 213.38.7.224 happy. I believe it is a mark of just how ubiquitous computers have become that they show up in places like this. We need a picture that informs people. It has to say "Computers are everywhere - Computers are NOT just PC's - This isn't an article about PC's - Computers are in every corner of our lives - Things that you might not think of a computers are in fact computers - Computers look like this inside - Computers control things". I think the Lego picture along with it's caption does exactly that. A picture of a PC sucks in every way imaginable - it says nothing to the reader (who is almost certainly sitting at a PC as he/she reads the article!) - it promotes commercial rivalry as every PC maker seeks to have THEIR PC be the one in the photo - it's also obvious - and making a non-obvious choice here is what'll keep people interested in reading the article. I'm happy to entertain alternatives but I'm not seeing any offered. If you have a better photo - bring it here and let's discuss it. SteveBaker 14:54, 3 July 2006 (UTC)

Agreed. (as an aside, some SoCs are pretty much complete computers with CPU, memory, I/O and peripheral structures; they just require power supplies and a clock signal to work, and little more to be useful. Still, the Lego brick is a solid choice for the intro photograph). -- uberpenguin @ 2006-07-03 16:21Z

Well, I guess you could count an SoC chip by itself - but without a power supply + clock...hmmm - dunno - is a car still a car if you take the wheels off? I guess so. But anyway, the H8 chip on the bottom-right corner of RCX board is an SoC (System-on-a-Chip) and showing it in an appropriate context (ie with clock, etc) seems worthwhile. SteveBaker 18:50, 3 July 2006 (UTC)

The Lego RCX illustrates the ubiquity of computers today, but I think it makes their use seem rather limited, failing to show the impact of computers on communication, science, manufacturing, culture. But I suppose what wouldn't? A cellphone, a supercomputer, a computerized production line, or a PC would be misleading too. So I don't really have any better suggestions! 213.38.7.224 16:38, 4 July 2006 (UTC)

We can always have more photos in order to cover all of those other things - the question here (I presume) is which of those things should be at the top of the article. There is no doubt in my mind that pictures like the Furby and the RCX are important to telling the story - the issue for some people isn't whether these pictures belong in the article - it's a question of what should be at the top. SteveBaker 17:04, 4 July 2006 (UTC)

Dispute over example in "Programs" section

I've reverted back and forth now with User:81.179.195.75 here, here, here, here and here. I can't figure out where the "40 million lines of code" figure is coming from, but the 2 million lines of code figure is coming from the Robert Lemos article. What's going on, User:81.179.195.75 ? -GTBacchus^(talk) 16:53, 19 May 2006 (UTC)

I support your reverting him for your given reasons. If the anon user doesn't provide any evidence and reason to use his text but continues to revert, we ought to report him on WP:VIP. -- uberpenguin @ 2006-05-19 17:54Z

It makes more sense now. The 40M figure is coming from Source lines of code, where it's quoted as coming from a book, referenced by ISBN. I think it would actually be interesting to use both figures, to see the difference in scale between a web browser and a behemoth OS. -GTBacchus^(talk) 18:40, 19 May 2006 (UTC)

Allright, though if you want striking code base comparisons, try an office suite or a RDBMS like DB2 or Oracle... 40 million lines pales in comparison. -- uberpenguin @ 2006-05-19 20:04Z

Edit to Program section requested

"A typical example is the Firefox web browser, created from roughly 2 million lines of computer code in the C++ programming language;[7]"

Firefox is not a typical example, typical would infer common or usual Firefox is not yet either of these. 2 million lines of code appears invalid, the foot note link forwards to an external site that does not mention that Firefox is written in C++. The external link does mention 2 million lines of code however the roughly statement makes it appear to be that authors best guess.

There is no proof that 2 million lines of code was actually counted using SLOC (source lines of code) method like the lines of computer code link is alluring to. The SLOC article does contain several examples all validated in that article, Windows XP would provide a better typical example because of the readers familiarity with Windows XP (perhaps unfortunately everyone knows XP). Placing a Windows XP example in the programs section is perhaps out of place, placing it within Operating systems section would be more appropriate.

Equally, we could point to the SLOCs in the Linux kernel, which is trivially verifiable because it's open source; or, as an example that people might not be familiar with, the fact that the Joint Strike Fighter will apparently have roughly 11 million lines of source code in its associated software. But we're kind of missing the point here, which is to demonstrate that software applications represent a huge intellectual effort. --Robert Merkel 14:46, 20 May 2006 (UTC)

Linux and to some extent Firefox have, by being open source have a potentially accurate but also more variable (depending on distrib, optional extras etc) SLOC figure. Windows being closed source ought to have a less variable but, more difficult to prove figure. The size of the intellectual effort is important, a contrast between an applications lines of code and a Operating Systems lines of code could be informative to the readership. Part of the point is also that edits are occuring because the facts are disputed by both parties, the SLOC article contains verifiable data provided by published authors about the subject matter. The “roughly 2 million” in the cnet article to me at any rate carries less gravitas. Windows XP cited as an example is one that the majority of readers could relate to.

The trouble with picking Windows XP is in knowing what exactly is being counted. Windows is not just one program - it's DOZENS of programs kinda mushed in together. Firefox is a good example because it's a single, standalone chunk of software and you can verify by counting the lines yourself. SteveBaker 18:59, 25 May 2006 (UTC)

I think "dozens" is a severe understatement. I agree, however, that the Windows XP source code reference is very very weak. I'd suggest following Robert's suggestion by counting the lines of code in some readily recognizable piece of open source software. Perhaps a Linux or BSD kernel or even OpenOffice.org (if someone wants to have a script churning away at their hard disk for an hour :). I'd be glad to do the count myself if someone would indicate which piece of software might be suitable to count... -- uberpenguin @ 2006-05-27 03:45Z

OOO might be a good illustration, as the average non-technical reader might have a good idea what functionality is provided through that 20 bajillion lines of code...---Robert Merkel 07:20, 27 May 2006 (UTC)

Technically, we can't just count the lines in some software package and report the result since that would be a violation of WP:NOR. (I personally find WP:NOR to be a real pain to adhere to - but it's a rule) SteveBaker 12:09, 27 May 2006 (UTC)

Personally I don't consider that to be original research; it's just a quick calculation to find an undeniable, easily verifiable fact. Is it original research, for example, to perform calculations for the maximum addressing space under LBA in the ATA article? I'm sure with enough searching you could find a citation for the given calculations, but why bother if the majority of the editors of the article know the numbers to be correct and can verify them with a calculator? The NOR rule was designed to prevent inclusion of information that is questionable (which of course is ridiculous since plenty of published papers contain questionable to downright false assertation). If you were to apply NOR unilaterally, most good articles would be downright impossible to keep around since it's impossible to cite every single factual statement made in a text. One's own wording of historical events or the usefulness of a certain device could very well be interpreted as 'original research' if you take things far enough. Anyway, I wouldn't worry about NOR in this case unless someone here actually has an objection to the proposed citation and method. All that I see as necessary is a footnote to the effect of exactly how the lines of code were counted. -- uberpenguin @ 2006-05-27 16:44Z

Russian computers section

I've removed this section because it contained numerous overtly POV statements and several rather wild claims that were unsourced. What's more, the information probably doesn't require an entirely separate section within this article when any significant developments in Russian computer technology should be seamlessly integrated with the rest of the article. I ask the author of that paragraph to reference some of his claims and give more detail about which developments he feels are most important to mention in this article. -- uberpenguin @ 2006-05-27 03:40Z

This has nothing to do with POV statements or wild claims. I was simply quoting an article from a program known for its journatlistic integrity. The reason why so little is known is because of the cold war. Back then everything that was going on in the Sovjet Union was a secret for the west, and when the war was over the whole thing started to fade into oblivion when it was shut down. That's why there is so little information about it. If you can find a way to integrate it with the rest og the article, that's fine. If you don't read Norwegian, maybe someone can translate it for you; http://www.nrk.no/kanal/nrk_p2/verdt_a_vite/1956658.html And a short interview partly in English; http://www.nrk.no/dynasx?p_lenke_id=244350&p_artikkel_id=1956658&mswmext=.asx

Your contribution was also far too detailed for an overview article; it probably belongs more to history of computing hardware. Frankly, I'm not sure it's even significant enough to mention here. None of the really big developments in computing happened first in the Warsaw Pact; they managed to reverse engineer or independently invent most things the West came up with, but they didn't beat the West to the punch on anything major. In fact, as our article on the ES EVM suggests, it's even possible that the political leadership's emphasis on cloning Western designs rather than developing their own technology screwed up their own computer industry. --Robert Merkel 05:16, 29 May 2006 (UTC)

It's good stuff - it just doesn't belong in this article. I suggest you create a new article and put it in there. An article on Computing in Russia could certainly be linked to from here. SteveBaker 18:08, 30 May 2006 (UTC)

Thanks. In my opinion, it is irrelevant if it had any impact in the present technology or where the original inspiration came from. It is still a part of computer history. And yes, it was a bad idea to include mix western and russian computer technology. Maybe someone could write an article about, but that would have to be someone who knows how to write articles about computers.

For those who are interested in what was deleted:

Russian computers

"Parallell with what was going on in the west, USSR had their own computer evolution and technology. The first was made in Akademgorodok at the end of the 50's and named M20, and the last one at the end of the 60's which was called Big Electronic Computational Machine, or BESM 6. Measured in qaulity, these were just as good or even better than those in the west. Later generations were also made, but these were influenced by western technology, or by IBM 360 to be more precise. USSR even created their own personal computer in 1983, called Chronos. Its operative system got the name Exelcior. This program was later improved and ended up reminding about Windows in some areas, which was an example of parallel development. This personal computers were more modern and had the same speed as the PCs that existed in the west at the same time. The russian scientists were working on and making prototypes of advanced integrated circuits for their Chronos, such as special circuits for signal processing which had the same function as present graphics cards and sound cards, but the impact of perestroika would put a permanent end of this alterantive direction in computer science and technology. If the evolution had continuted, it would probably have ended up as a good alterantive to the present and dominant standards." Rhynchosaur 16:39, 27 August 2006 (UTC)

Oldest computer in the world

I leave it up to far more knowledgable editors than myself to judge whether this item [5] is appropriate for inclusion in the history section, or with some other related article. Politis 18:11, 7 June 2006 (UTC)

Thanks for the link. That article refers to the Antikythera Mechanism - which is mentioned in the second sentence of the history section - so we've got it covered. The difficulty in assigning the term "Oldest Computer" relates to the very definition of computer. We have chosen to apply a modern definition which distinguishes between 'calculators' and 'computers' by requiring that a computer be programmable. By this definition, the Antikythera Mechanism is not a computer - it's a calculator. You could use it to calculate the position of stars, sun, moon and planets - but that's all it could do. It could not be reprogrammed to balance your checkbook or to do any operation other than that one specific thing. If we were to consider primitive calculating engines as 'computers' (which we don't) then the first device like that is probably the South Pointing Chariot which effectively performed the calculation of subtracting one number from another and multiplying by a third number using gear wheels. It's not impressive - but it's just as much a calculator as Antikythera, Truly, the concept of using artificial aids to doing arithmetic is as old as the tally stick or counting on ones fingers...I doubt you could definitively come up with a "first ever single-purpose arithmetic calculator". Antikythera is an important milestone along the way - but I still think ENIAC wins the title "first computer". SteveBaker 19:47, 7 June 2006 (UTC)

I concur with SteveBaker. ENIAC was obviously the first programmable electronic computer and programmability (as well as the concept of the state of a variable) are what make computers different from calculators. --Coolcaesar 20:13, 7 June 2006 (UTC)

I don't know that it's so obvious. If it were, then we probably wouldn't have to give due respect to Konrad Zuse. Anyway, I'm fully supportive of our continuing to define a computer by the trait of programmability. It makes defining the scope of these articles much much easier. -- uberpenguin @ 2006-06-07 22:25Z

Not wanting to stir up that particular hornet's nest again - I think we give due credit to people who made things that helped to make computers possibe. We wouldn't have computers without all the work on what we are currently calling calculators - and all of the things like the Jacquard loom that contributed so much to programmability - and the guys who designed stuff like punched cards, random access memory, the transistor, the Teletype, the cathode ray tube, boolean logic, binary numbers...the people who worked on those things (including Konrad Zuse) deserve mention for the work they did - even if the things they built don't fit our definition of computer. SteveBaker 18:28, 8 June 2006 (UTC)

Computer != PC (again!)

<rant> Far too much of the content of this article makes the horrific mistake of equating the term "Computer" with "Personal Computer". We really, really need to stop doing that. From the guy who repeatedly insists on putting a photo of a PC at the top of the page because that's what computers look like to people who say stuff like the arrival of the mouse and the GUI was what made computers what they are today...this is WRONG, WRONG, WRONG.

There are something like 100,000,000 PC's in the USA today. How many computers are there? Well, in my house there are 5 PC's - and my son did a school project which entailed counting how many computers there are. You wanna guess? The answer was 87 - and I think that was an underestimate because he decided that things like the chip that plays "Happy Birthday" inside one of those horrible birthday cards was probably a custom music chip and not some kind of tiny computer...I'm not so sure. However, I think it's pretty safe to say that there are at LEAST a dozen computers for every PC (heck - there's likely to be a dozen computers inside a PC). So PC's are a relatively small fraction of the computers out there - almost too insignificant to mention - let alone dominate the article.

Let's talk about washing machine controllers, telephones, the computer inside your TV remote, the ones inside toys, the industrial controllers, the car engine management computers, your wristwatch...the PC comes *WAY* down the list. It's worthy of a mention - but it shouldn't be the majority of the article.

<end of rant> SteveBaker 21:16, 8 June 2006 (UTC)

Calm down. It's okay. The most verbal editors to this article all agree with you and recognize the problem. Just remove incorrect text; rants aren't needed. -- uberpenguin @ 2006-06-08 21:53Z

Comparable dates

The article includes the following:

"Notable achievements include the Atanasoff-Berry Computer (1937)... the secret British Colossus computer (1944)... the Harvard Mark I (1944)... the decimal-based American ENIAC (1946) and Konrad Zuse's electromechanical Z3 (1941)..."

To the new reader the dates are presented as though they are directly comparable. I don't have enough information to be sure but I have the impression that this is not so.

These are dates I've found in various articles. (Please correct where they're wrong.)

Machine	Concept	Demonstrated working	Operational use
ABC	1937-38	1941?	N/A
Colossus	1943?	1943	1944
Harvard Mark I	1939	1944?	1944
ENIAC	1943?	1944	1946
Z3	1939	1941	?

It looks from this table as if the ABC's concept date is being presented as comparable to ENIAC's "shown working" date or Harvard Mark I's operational date.

It would be more useful for the reader to be given dates which show the same stage of development for each machine. The concept date doesn't look very useful for this purpose. Equally, the operational date would be unfair to the ABC and possibly the Z3 if they were never in operational use. The date when the machine was first shown working looks like the one that is best for this purpose.

Can anyone contribute further dates or any other thoughts? Adrian Robson 19:19, 22 June 2006 (UTC)

I am a bit rusty on the history of the early computers but I think you are right that the dates need to be fixed for consistency. We have a big problem on Wikipedia with Iowa State University graduates trying to elevate the prestige of their lousy fourth-tier university (and awful fifth-tier state) by pushing a POV favoring Atanasoff's work over all others in many computer-related articles. What does everyone else think? --Coolcaesar 20:42, 22 June 2006 (UTC)

I think the attack on IASU folks is unnecessary. The persons responsible numbered in the two or three range. I also don't think that all those dates are even necessarily needed in this article; wikilinks to the respective computers' pages should be adequate. -- uberpenguin @ 2006-06-22 21:26Z

Let sleeping trolls lie. SteveBaker 23:53, 22 June 2006 (UTC)

Many thanks for the comments. As there don't seem to be any improvements on the dates at the moment, I'll update the article with the dates in the Demonstrated Working column, unless someone has any better ideas in the next few days. Adrian Robson 17:11, 4 July 2006 (UTC)

Depending on which side of Honeywell v. Sperry Rand you're on, you might argue the dates for ENIAC were 1943-1944-1945. (The middle date was especially important for determining the critical date of its patentability; the judge in the case said that the two-accumulator version of the ENIAC, ready in 1944, was as good as the whole thing, citing this as a reason for invalidating the patent.) ENIAC was certainly in "operational use" throughout 1946 prior to its move to the BRL. Robert K S 14:36, 31 August 2006 (UTC)

OK, I've amended the table above. I'll amend the text of this article and I would think the template table of dates should be amended in line with this, too. Adrian Robson 08:40, 13 October 2006 (UTC)

Featured article

It's my desire to get this article to featured status, but that's not something I feel I can accomplish on my own. My expertise lies in digital computers and their architecture, so while I can address the bulk of the article myself, I'll fall short on some of the other critical areas (such as history, computational theory, and alternative computing models). Here are my notes upon surveying the article regarding some of the things it needs before it can attain featured status:

Intro - The intro is pretty solid. It could probably use a little refinement once the bulk of the article is brought up to a high standard, but in general I'm pretty happy with how it reads.
History - Also a good general overview. It would probably be good to carefully check all the dates and facts just to make sure it's accurate. I think the paragraph on stored program machines could stand for a little bit of improvement. There certainly should be a little more text on modern computer developments. The history section should at least briefly mention the advent of the PC and how it has helped computers become ubiquitous in industrialized society. I think we should also include a few pivotal architectural improvements that were made feasible by transistors. Increasing levels of parallelism and complex superscalar CPUs are worth brief note, I think.
Stored program architecture - Needs improvement. I think in the first few sentences it is guilty of missing the point of what is significant about stored programs in the first place. Furthermore, the logical separation of control units, execution units, memory, and I/O isn't mandated by the stored program design, nor is it always the case (I'm thinking of some DEC machines in which the line between memory and I/O was very thin because I/O was abstracted AS addressable memory). This section rambles a bit, goes into some unnecessary diversions, and probably needs to just be rewritten. It's extremely important to talk about the stored program architecture in clear simple terms without getting distracted by implementation details or an explanation of the mechanics of memory storage systems, etc. I think this is also the place to talk briefly about usage of the term "computer" and how stored program architecture is an important, if not the important, defining point of what a modern "computer" is.
Digital circuits - Okay, but could be trimmed down. I think there's too many specifics in there for a general article on computers.
I/O devices - Hmm... Well something should be said about I/O in this article, but where this section is situated seems very out of place. I'm open to suggestions here.
Programs - Crucial section to have. Probably should be renamed to "Software". Needs to have a short paragraph about ISAs since they are the interface point between software and hardware.
Libraries and operating systems - Probably should stay, but needs a little reorganization and flow improvement.
Computer applications - Okay, maybe reorganize a bit and come up some major lines across which to divide various computer uses into classes.
Alternative computing models - Needs to be expanded a little. For some reason I feel that it isn't representing various theoretical and experimental computing models well enough. However, my knowledge on the subject is limited, so perhaps someone else could take an interest here?
Computing professions and disciplines - Do we really need this? It's definitely related, but I don't find it of significant direct importance to this article. Maybe it can be moved to another article or removed altogether and an appropriate article linked instead.
See also - Trim this way down. It's overwhelming as is.

What's missing:

Computers in society/popular culture - Someone (preferably a sociologist :) really ought to write something about how computers have gradually become an integral part of society and the way people live and communicate. Smatterings of this topic are dispersed through the article, but I think it's important enough to merit its own section.

Please discuss what you think I have right/wrong about the direction this article needs to go in. I'll start improving along these lines where I can. If you'd like to work on some of the above topics, please let me know here. Recruiting other knowledgable editors would also be a big plus since this is a broad topic that requires the viewpoints of multiple people. -- uberpenguin @ 2006-07-18 22:52Z

I still believe that the entire article is The Wrong Thing - not that it's inaccurate or badly written or anything - it's just not what should be here. 'Computer' should be a very abstract top level article like Physics that surveys the field and points you off to other articles for almost all of the content. Beyond the introduction, Physics has almost one word in four linked off to other articles and tables to help organise the material..that's what we need here.

Computer should organise the field and impart structure onto the articles we already have. We should do a comprehensive sweep of all of the Wikipedia for articles on stuff like computer games, logic design, programming libraries - and organise it into tables like the ones in Physics. The history stuff should be in a history of computing article - the alternative computing models belongs in an article of its own...a "How Computers Work" article would be worthwhile. Trying to wedge too many fragments of information on this wide field into a single article just isn't working...IMHO of course. SteveBaker 23:15, 18 July 2006 (UTC)

To an extent I agree, though I don't believe Computers to be nearly as broad a topic as Physics. Even if we do turn this into a hub article (which I'm not at all opposed to), it still should (and can) have a very solid intro and probably a brief history section. So do you want to propose a structure for this article? It's been nagging me for awhile that the entry article for everything computer related on Wikipedia is in a relatively poor state, and I'd like to get something done about it that everyone will be happy with. -- uberpenguin @ 2006-07-18 23:39Z

Well, I really do like the way Physics does this - so I think we should organise like this:

Intro paragraph - what we have now is OK.
Introduction section - explain that we have hardware and we have software - we have operating systems and applications - we have the basic hardware CPU/RAM/ALU/whatever and we have peripherals.
Tables - structure the subject - list the Wikipedia articles that fall into each catagory. Do this heirarchically - hardware breaks down into architecture, electronics, other technologies (quantum, bio, nanotech, etc). Software breaks down into OS's, applications, libraries, languages. Then break those down - languages breaks into C, C++, Fortran, Pascal...etc. OS's breaks into DOS, Windows, Linux, UNIX, MacOS. Libraries breaks into DirectX, OpenGL, OpenAL, etc. Applications breaks into video games, office applications, etc.
Summarize the key articles - history, computer architecture, software. Have those be rich with links.
A really comprehensive 'Further Reading' section.

SteveBaker 03:04, 19 July 2006 (UTC)

I like it. Do you want to work on any particular part of that? I'll take whatever you don't want. I think the tables section will take a lot of thought since beyond the major classes of OS (DOS, Win32/NT, Unix/BSD-derivative), there are a plethora of other miscellaneous designs. Same goes for programming languages. Anyway, let me know what section(s) you want to take and we can start working on this on Computer/Temp or the like. -- uberpenguin @ 2006-07-19 03:18Z

I guess I'll take the tables part if you like. But before we dive in and start work, let's at least get some kind of consensus from the community here. I'd hate to spend a month honing and polishing a replacement article only to end up with open warfare about whether it's the kind of thing we should have here. Let's leave it a few days to see what comments we get from other contributors. SteveBaker 04:04, 19 July 2006 (UTC)

You could spend a month waiting for anyone else to comment... However, a few days is reasonable to wait, I suppose. -- uberpenguin @ 2006-07-19 13:43Z

I think this is a potentially reasonable plan. My only concern with this plan is that the subsidiary articles that it would link to sometimes offer relatively few concessions to readability for the non-expert reader, and do little to place these topics in context, which was my goal with the present version of the article. So this means that it will be important to place the links in context within this article, but also to improve the readability of some of the other computing related articles. In fact, surprisingly enough, the computing-related articles now generally lag in quality behind some other aspects of the Wikipedia.

Secondly, we should really think about how such a top-end article belongs in the context of other such articles, such as "computing", and computer science.

I would further request that any reorganization take place on a temporary page until it is to a satisfactory standard.

I'll try my best to help out when I have time, but I'm travelling at the moment so I don't have a lot of it. --Robert Merkel 13:58, 19 July 2006 (UTC)

I agree that a few days waiting for comments is reasonable - we don't need to wait a month. I was merely commenting that if we worked on it for a month and then dropped the present article in favor of the new one - then the howls of complaint might be terrifying! If we can at least get a reasonable amount of consensus before we start then hopefully we can say "You had the chance to discuss this and you didn't" and have people such as yourself on record as agreeing that this is a good idea so it's not just a minority going off at a tangent. I recommend we wait until Monday in order that 'weekend Wikipedians' get a chance to comment - but if we don't see too many howls of derision we could at least get started sooner than that.

I also agree that some of the subsidiary articles aren't up to scratch yet - but the solution is to fix them and not to try to cover up their deficiencies with an uber-article that trys to cover everything (after all - not everyone will come here first). I agree that a significant part of what's proposed here is to move sections of the current text out into the subsidiary articles and to improve them as needed.

We certainly need to coordinate what we propose with computing and computer science (and perhaps also electronics and software). There is a strong case for merging computing with this article (or vice-versa).

We obviously need to do this work off in a subsidiary page. uberpenguin already put a copy of this article into Computer/Temp - and we should certainly work there until we can say that the new page is better than this one and then throw the big red switch. SteveBaker 17:12, 19 July 2006 (UTC)

You're preaching to the choir here, Robert. Seems like nearly every time I look up a digital electronics or computer architecture related article here, it's either a mess, too convoluted to understand, or downright wrong. Your concern is valid, but the more I think about it, the more I agree with Steve that this article should be a hub like Physics. Once this article is up to snuff perhaps we can slowly focus our attention on other major articles that need some love. Regarding merging this article with Computing, I'm not sure how well that will go over since a convincing argument can be made (even based upon what we've discussed in the past) that a computer is a specific kind of computing machine, and "Computing" can include all kinds of devices that we would no longer term "computers". Whether that alone is meritous of separate articles I don't know, but I'm inclined to keep them separate just for simplicity's sake. The scope of one hub article about computing and computers would be fairly enormous. -- uberpenguin @ 2006-07-19 21:38Z

I went ahead and posted messages on the talk pages for the electronics and computer science wiki projects asking for their insights, suggestions, and help. Also, Computer/Temp currently just redirects to Computer. It was used in an earlier revamping effort, so whenever we get underway with the new article, we can just start putting content there. -- uberpenguin @ 2006-07-19 21:49Z

So, after a week and a half with no comment from third parties, do you think it's safe to start rewriting this article on Computer/Temp? I'll copy the current article text there to work from. -- uberpenguin @ 2006-07-31 18:53Z

Though I am not involved in this project, after reading your proposals and discussion, it sounds reasonable to create somewhat of a hub article. It would be a shame though, if only tables and links remained, a decent prosa embedding and/or introducing the links and tables would be lovely (and necessary, too.) Best wishes, I'll provide help if needed and if time permits. --Johnnyw talk 21:40, 31 July 2006 (UTC)

Yeah - let's get started. We aren't talking about only tables and links - but rather putting a strong emphasis on pointing people to other articles than trying to answer everything in-situ. This will give us the freedom to provide a nice overview without the need to dive into details that are covered elsewhere. Once again - I'd like to recommend the Physics article as the model we should shoot for. SteveBaker 22:10, 31 July 2006 (UTC)

Yes, and as it is I think we should retain most of the intro of this article and a fixed-up history section in the redesigned hub article. Anyway, let's keep all further discussion of the content of the new page on Talk:Computer/Temp. -- uberpenguin @ 2006-07-31 22:53Z

User:Coolcaesar and Wikipedia:Requests for arbitration

Since User:Coolcaesar frequents this page, I request that any of the editors on this page that have an opinion of User:Coolcaesar to express it on a recently filed Wikipedia:Requests for arbitration. You may also add further evidence there to support your view, as well as explain all situations/attitudes/etc. about the user. I urge anyone that has any sort of opinion about this user leave a comment, and comments cannot be used against you in any way on the arbitration page in the future. Thanks for your time. --Mr.Executive 08:46, 20 July 2006 (UTC)

(I restored this edit because it is un-cool to remove other people's stuff from Talk pages) SteveBaker 22:18, 20 July 2006 (UTC)

Though in all honesty Coolcaesar does NOT really edit this page much and this is a rather inappropriate place to advertise his RFAr. -- uberpenguin @ 2006-07-21 02:26Z

I agree - but removing posts (particularly controversial ones) from Talk pages is a worse thing - so I put it back. SteveBaker 13:35, 21 July 2006 (UTC)

wtf...why is the article like repeating?

computer networks

The article mentioned the SAGE computor. A link showed a 86000 based computer from a colorado company.

The SAGE AN/FSQ7 Tube Computer At NYADS McGuire AFB N.J. was made by IBM at Kingston New York. I was a Blue Suiter who attended the Computer School at Kingston New York in 1961. Air Force Blue Suiter. It was a lot of fun working on that old beast. Later I worked at McChord AFB Washington, on the CC/DC Sage complex.I rarely ever see the name SAGE mentioned.75.20.220.32 03:42, 31 August 2006 (UTC)David Stewart

Additional note

Networking and the Internet

Computers have been used to coordinate information in multiple locations since the 1950s, with the US military's SAGE system the first large-scale example of such a system, which led to a number of special-purpose commercial systems like Sabre.

I am a newbie. I didn't wish to disturb the written text refering to the sage link. The link goes to a computer co in Colorado. My comment refers to the Military (USAF) AN/FSQ7 Sage Computer. A second generation Tube (valve) Machine manufactured by IBM at Kingston, New York.Dcstu41 20:17, 31 August 2006 (UTC) dcstu41

Dcstu, next time, just be bold and fix the link. I've made the correction. --Robert Merkel 01:55, 6 September 2006 (UTC)

Image

Is realy the lego thing the best we can find for the first image of the article?Why not an open casing?--Pixel ;-) 17:24, 19 September 2006 (UTC)

Open casing of what? We've had several discussions of the image before, and it's always been agreed that a picture of a PC is about the most boring and useless thing that could be in the lead of this article. -- mattb @ 2006-09-19T19:18Z

questions

what bit patterns are represented by the following hexadecimal notations? i)BC ii)9A

How many cells can be in a computers main memory if each address can be represented by three hexadecimal digits?

This isn't really the place to seek help with questions like that. -- mattb @ 2006-09-28T14:21Z

BC=10111100, 9A=10011010, Three hex digits is 12 bits so 4096 addresses are available.

...and yes, this is the wrong place to ask. SteveBaker 19:56, 29 September 2006 (UTC)

Who made?

Hey who was the first inventor of the computer?

That's practically impossible to provide a single answer for since there are varying definitions of "computer". Read history of computing hardware. -- mattb @ 2006-09-29T15:39Z

Yes - the definition of what a computer is defined as is the tricky part of that question. Charles Babbage is generally credited with the first design for a programmable machine (The 'Analytical engine') - but he never managed to build it. Arguably, before that, the Jacquard loom was a programmable machine - and it worked and was widely used - but it didn't do any arithmetic - so I guess that's not a computer either. Then you get into various machines that could perform calculations - but which weren't programmable. If you count those as computers then can you count a mechanical adding machine as a computer? What about a slide-rule - or an abacus? If you allow an abacus then does counting on one's fingers count as 'computation'? So nowadays we generally call those things 'calculators' and reserve the term 'computer' for programmable things. So that leads us to obscure machines such as the cryptographic contraptions that the British built during WWII to crack German codes - but those machines were only just barely programmable - and there is some debate about whether they can be counted as the first computers. For my money:

The later model of the ENIAC was the first working, practical, useful, programmable computer that could do math - which would give the credit to John William Mauchly and J. Presper Eckert of the University of Pennsylvania.
If you allow for theoretical designs that were never built then Charles Babbage easily gets the credit.
If you allow non-programmable calculating machines then the first human to count using his/her fingers gets the credit - the unknown inventor of the abacus might arguably have built the first thing that could add and subtract (but only multiply and divide indirectly) - Edmund Gunter invented the slide rule which could multiply and divide (but not add or subtract) but Blaise Pascal, and Gottfried Leibniz built the first mechanical calculators that could do all four arithmetic operations without much human intervention - so one or other of them wins in my opinion (they invented these machines at about the same time - it's hard to say who was first).
If you allow programmable but non-calculating machines then Joseph Marie Jacquard gets the prize for his programmable weaving machine that could weave complex patterns from 'programs' set up on punched cards.

I worked on the team that built the first ever CD-ROM (which contained all of the dictionary entries for the letter 'O' with text, pictures and sounds) - does that count? No - I thought not. SteveBaker 19:51, 29 September 2006 (UTC)

True Gottoupload 22:31, 4 October 2006 (UTC)

New note by David Stewart Just a comment on this undertaking. I am one of those dinosauers who worked on computers in the early sixties. The SAGE AN/FSQ7. The work here on sage was pretty good. The work on the overall effort is . Nuts work of this scope just doesn't fit proper discriptors. Great Job.I read it with interest. Mr Baker sir, My computer is beige.I am now retired. Work of this nature is vital to the future generations. I salute you folks.It will take years.People ask me about this stuff. How do you explain a lifetime playing with these things. You folks are doing just that. Don't worry about things just keep putting in the information. There is tonnes of it. For starters what are you going to use as archival devises. The state of the art outruns the archival process. Plann ahead.for when memory goes holographic or goes organic whatever. 10-23-2006 Dcstu41 03:36, 24 October 2006 (UTC)Dcstu41

Firstly David, I would strongly encourage you to contribute to Wikipedia. If you were around in the early days of computers, every single thing you can remember will be of huge interest to future generations. The best way to preserve that information for future generations is to stuff it into Wikipedia.

I think the concerns over archival matters largely goes away with the advent of OpenSourced documents and the Internet. Wikipedia can't get 'stuck' on obsolete media so long as people are still using it. It gets mirrored onto lots and lots of other sites - and as new media comes along, it'll just naturally get copied from one to another. Concerns that ancient historical versions might get 'left behind' is unwarranted too because Wikipedia has integral version control that means that you can easily get back to the state of any article at any time in the past. The thing to be concerned about is closed-source material. SteveBaker 04:20, 24 October 2006 (UTC)

Computer/Temp

The major rework of this article has been proceeding slowly (see Computer/Temp for the progress so far. I think we are close to the point where we can remove the present Computer article and move Computer/Temp up to replace it.

The changes are drastic - so don't be surprised to see vast chunks of the current article simply vanishing. The intention is most definitely NOT to try to say everything there is to say about computers in one gargantuan text. The intent is to provide a simple introduction to each of the major subject areas and to defer to the many excellent daughter articles that are out there.

I'd like the change over to go smoothly and without major upset from people who might feel that a lot of their work is being destroyed. So I'd ask for the following:

Just so we keep things together, please discuss any problems you have with Computer/Temp on it's own talk page - not here.
There is definitely information in the current Computer article that isn't in Computer/Temp - that is 100% deliberate. However - if there is information in Computer that isn't in Computer/Temp and it's not in any of the articles mentioned as a "Main article: ..." in Computer/Temp - then we need to know about that. Probably, that means expanding the referenced article because even Computer/Temp is awfully long already. Anyway - if you find something like that, please let's discuss it on Computer/Temp's talk page.
Obviously we need facts to be checked, prose to be polished and references to be added.
The new Computer article needs to be a featured article - so as soon as we move Computer/Temp up there, we need to start seriously attacking it as FA reviewers will.

Thanks in advance. SteveBaker 05:57, 1 November 2006 (UTC)

Last Warning!!!

We are very close to moving Computer/Temp to Computer - the article as it is now will abruptly cease to exist and be replaced by the new version... SteveBaker 22:57, 10 November 2006 (UTC)