Now that they've covered the future of computing, the TechStuff guys are ready to tackle computers of the past. Tune in as they explore the origins of computational devices, from the abacus to punch-card computers, in this podcast from HowStuffWorks.com.
Learn more about your ad-choices at https://www.iheartpodcastnetwork.com
Brought to you by the reinvented two thousand twelve Camray. It's ready. Are you get in touch with technology? With tech Stuff from how stuff works dot com. Hello there, everybody, and welcome to tech stuff. My name is Chris Polette, and I'm an editor here at how stuff works dot com. And sitting next to me, as always, with a really enigmatic smile on his face, is senior writer Jonathan Strickland. Hey there, having a Mona Lisa moment. Oh okay, alright, Well, you know the vacuum tubes are all humming and everything is ready to go. Let's not get ahead of ourselves. We have to start this this podcast off a very specific way. We do yes with listener mail, I swear you moved that on purpose always. So today's listener mail comes from Jason from Fort Wayne, Indiana, and Jason says, hey, guys, I was having a conversation about computer technology the other day and the people I was talking to mention something about old computers where you had to carry around a box of punch cards that served as the program. However, if you happen to drop the box and got them out of order, you essentially lost your program. I'm only twenty one and was wondering if this technology was a bit before my time, because I don't remember a thing about it. I love hearing about these older technologies and how they work, and would be grateful if you could enlighten me on this subject. I enjoy listening to you guys, and congratulations again on your one episode and too many more to come. Well, thanks, Jason. We're going to tackle that a little bit today because we're going to actually talk about computers from the past nifty You know, it's actually it's more than a bit before your time. It's actually sixty four bits before your time. Uh that you Oh, I see what you did there? So um, yeah, I upgraded your Internet. But let's let's let's let's let's take a step in the way back machine, shall we. Sherman and um try and come back all the way to the like the earliest computational sort of engines, and then we'll work our way forward. I've got some stuff I can talk about as far as the punch cards go, because I came across a great paper written by a former engineer at IBM who has uh he kind of shares his experience of using punch Cards for the first time, and it's it really is a great read. I'll just read a couple of excerpts from it to kind of give you guys an idea. But before we do that, let's, um, let's let's think back a bit. Now, computers really are number crunching machines. You know, everything you see and do on a computer is the result of different numbers being crunched in different ways. You've got you've got two different things going on. You've got a data stream coming in and you've got a set of instructions going on. The instructions tell the computer what to do to the data stream, and then you get your result. I was wondering what what all those crumbs were on my desk. Now I know they're the crumbs numbers, numbers. Yeah, this is gonna be a hell of a show, guys. I I would like to apologize on the behalf of Pallette. I mean, even I have my limits, folks. Um. So so let's let's go back to about five BC. All right, So that's that. That's what that was a PC junior, Right, Yeah, that's yes, exactly. Now that was the first Apple No, no, no, five hundred BC that's it's well beyond the first apple. Did you get where I'm going with that? Okay? Anyway, so five n b C that's when you start seeing counting boards show up. This was a way for people to keep track of numbers that were bigger than ten, because you know, we have ten fingers, most of us anyway, and uh, you know, if you're lucky and then um eight fingers and two thumbs. Alright, fine, fine, but if you want, if you wanted to try it exactly, if you wanted to track numbers bigger than that, then uh, you know, you you had to store it in your head or if you were talking about really big numbers, you would want something to help you out. So that's where counting boards came in. A couple hundred years later that kind of evolved into the abbacus, and then we can skip ahead quite a bit. Okay, how far I was gonna go to sixteen seventy nine? You got something before that? Uh yeah, actually, uh h, I had a sixteen twenty three, which was actually there. It was before that because Leonardo da Vinci had actually drawn up plans for a calculator, a mechanical calculator. But William schickerd had created the first one in sixteen twenty three where actually built the thing. And then you know in Blaze, Pascal had the Pascalene, which was an arithmetic machine, and that was the first volume. You know, one that was the first calculator was actually created in volume around sixteen six. Actually built fifty of them. But I'm assuming that was all by hand, so we're not talking mass production. Yeah, no, I don't think they had assembly lines back in sixteen Yes, but you were, I'm betting based on your data, I'm betting you're thinking, Uh, that guy who invented calculus so got free liveness, who I would like to jump up and down upon his grave for inventing calculus. Well, you know, for those of you who are going to write us and tell us that he didn't invite invent calculus a calculus, well he he independently came up with calculus at the same time, around the same time as Isaac Newton, which I always thinks, so yeah, it's it's it's one of those one things where two different brilliant thinkers come up with the same sort of system um independently and it does happen. It's pretty neat stuff when it happens. But yeah, Newton is the one you should be yelling at though right, well, I'm gonna yell linus anyway. So he also but because he also introduced binary mathematics, and now binary mathematics, we're talking about using either of the numbers. Now people are going to write in because I'm calling it a number using either zero or one as the your digits. And uh, that's what all of modern computing is based off of. Our these binary digits, or bits as they are also known. And uh, I'm that suppose the guy who came up with this idea of using bits um Now, it would be a while before anyone had come up with a practical purpose for them as far as computing goes. But that's where it got start. What's the next date you wanted to throw out there? Because I'm gonna I want to. I've got a big list of dates here, and I have another big gap. Okay, Uh, the next date that I wanted to bring up is eighteen o four, eighteen o five. Yes, that's uh, that has nothing to do with computers, or does it, Because that thing you're having to do with weaving, Yes, it's unbelievable. Joseph Murray chaquad yes, yes, yeah, this is a guy who um yes, Jacques Jacques, Yes, you shouldn't say. Alright, so what did you said that? Right? What did Joe Mury do? Well? Joe Mury, he was he invented an automated loom. And you might say, okay, well, was an automated loom have to do with computers? Well, he did it. He used method that had already been used before, but he perfected it. It was a method where he used a punched cards to allow the needles of the loom to pass through to create specific patterns weaving patterns. Um. So it's sort of like you would see on a player piano, because that is something you might I'm guessing that those of you listing might have actually seen with your own eyes how the digits pop up and down with the holes in the paper. Right. So the yeah, the holes in these cards were what allowed certain needles to go through at certain times to create these patterns in in uh, when he was weaving. And why would this be important? Well, it would allow you to create patterns that only a master weaver could really make on his or her own. And then automated in such a way that anyone running this loom could create the same pattern, which, by the way, tended to tick off the master weavers a lot. Yeah, they actually burned quite a few of these automated looms and stop him. But but they definitely protested and as a result, textiles in France, Uh, the prices dropped quite a bit because it turned up there was a much cheaper way of producing these really nice weaves. Um. I'm surprised that you weren't the person to drop the trivia effect here, But since you left the opening for me, I'm going to point out that they also, in addition to setting fire to these limbs, like to drop wooden shoes in there, you know, the sabo that's a bot, which gave rise to the term sabotage and later would produce an awesome Beastie Boys song. So thank you, Joe Marie. But yeah, the getting back to it, the whole reason we're talking about is because the punched cards, those of course come into play later on. So the next the next day I have is one. Yeah. Yeah, that's when Babbage, Charles Babbage first invented the idea of the difference engine. Yes, he would later go on in four to come up with the analytical engine. Um, neither of which he ever actually saw a maid. No, he made bits of them, but never created an entire working engine. Now this was a shame. This was an incredible idea, and more so, the more we look back on it, the more we realize how incredible this machine would have been had he been able to create it at the time, because it took into account many of the things that modern computers actually do. I mean, it would have done these things on a different scale, but that's it's and on a mechanical scale, not an electrical scale. But it was still really cool. And the idea here was that, um, you know, if if he, if he could have created it would have been the steam powered, massive thing made out of brass and steel and you could run computations through it. And um, it's just a shame that he never you know, he he realized during his lifetime that he would never be able to actually build it. Well he uh, he had a problem. There was no the venture capitalist had not been invented yet, being on the government to uh to fund him, and they basically figured out that rather than finding navigation for ships, which is what he said he wanted to do with this machine. He wanted to be able to to uh, to run calculations to help people travel over the sea more safely and efficiently. Basically, he was an inventor who wanted to build better computational machines and they said, yeah, we're not giving you money for that. Dude. Yeah, he wanted to rush the Singularity on a lot earlier than we would have seen otherwise. Um. But despite the fact that no one built this machine for for a couple hundred years actually because um, well a little less than two hundred, I guess, uh, because it has been built uh in London. Um was that since then? And since then? Yes, yes, I think two thousand two is when when they built that that model. So that was a long time between design and production. But despite the fact that was never built, it didn't stop people from writing programs for it. Well, in know, give a hobbyist to a project. One of those would have been the Countess of Lovelace. Yes, not that they ever met, no, and but she created programs for Babbage's theoretical engine using the binary system. Hey, so there you go. You've got your first binary coded program for a computer that doesn't exist. And now we can say that about lots of different things in modern times, to build lots of programs for computers that can't run them yet. Well, you know, it's uh. With Babbage's machine finally being built in the two, you know, the twenty first century, it's very much like every other machine that's being built in the twenty first century. It was obsolete as soon as it was exactly so, setting a precedent there. Skipping out a few more years. In eighteen fifty, you've got George Boole who invent and logic that provides the basis of of computer logic and is far too complex for me to go into here. Let's see what what what's the next thing you want to hit? I mean there, I'm jumping ahead quite a bit from here. Otherwise, Yeah, it became obvious when I was doing the research for this that there's far too much to cover in our normal time frame, so I cut out some dates here and there. Okay, So what's the next thing you have? Um, I'm I'm getting towards the uh, the early early twentieth century in my notes around nineteen eleven, actually, when a bunch of you know, basically, these machines that existed at this point where advanced calculators. If if that, I mean they were used for business purposes, and um, you know, business calculators were pretty popular. They had they had reason to, uh to do these kinds of calculations. I mean, the the census needed advanced math, um, and I mean they weren't really used for science, so you were just tabulating numbers. Um. Well, nineteen eleven, a bunch of business machine companies decided to get together and they created a new company called this is a really catchy name, Computing, tabulating Recording Company. It was so catchy that they decided to change it in to International Business Machines. Hey, one of the initials of that IBM. I think I've heard that before. Yes, it's a it's a relatively small company, but they were they were to go on to be one of the first well you know this business machine manufacturing, but they uh sort of set a precedent later on. Right. Well, well, before we get further in there, I had one date before that, nineteen that was when William DeForest invented the vacuum tube, Yes, which provided the basis for what would eventually the thing that would eventually replace the vacuum tube was the transistor, but that wasn't around yet. So we had these these machines using enormous vacuum tubes instead of tiny, relatively tiny transistors, and those are going to be important in some of the machines, were getting ready to talk about so UM. That was when IBM adopted the eighty column pun card UH format. The so the eighty columns was your typical punch card. You were asking us Jason about the punch cards. Um, this was the way that you could design programs for early computers. Now when we're talking, we're really talking about running specific um functions. Were not really talking about programs yet. That's that's kind of a little early for that. But the punched cards that IBM settled on that became the standard for them, and a lot of the machines could only read the first seventy two columns, so you'd say, well, what would you use the last eight columns for a lot of programmers would use those columns in order to create a serial number so that you could keep the punch cards in the correct order. Because anyone who has worked with a punched card system would tell you if you drop those cards and they all fall out and get out of place. Yeah, that's a bad thing. It's even worse than slides. Yeah, because you know, at least with slides you can look at them and go, hey, that one goes here. So then we've got six. We have Alan Turing doing the publishing is on computational numbers, which was also very important for theoretical computer principles. Um. And then I've got the the first what what the United States anyway recognizes as the the first electronic digital computer. Do you know what this one is? Um? I what sort of depends on which one is what a judge decided was the first one. So because because there is some debate about that exactly so legally in the United States anyway, the first computer, electronic digital computer was invented by John Vincent A, Tennis Offt and Clifford Berry, and it was called the A B C. Isn't that clever? Yes? For theisoft Berry computer. YEA. So that was that was But in started in the late nineteen thirties, it was completed in the early nineteen forties, and it wasn't until nineteen seventy three that the US judge said that that was, in fact the first digital electronic digital computer, as opposed to one that was had an interesting acronym for a name, Yeah, eniac. That was the that was the competing computer in the court system for the first electronic digital computer, although they were sort of different. Yeah, oh no, they were very different. So so all right, so that was nineteen round eight. You've got Conrad Zeus who builds his Z one calculating machine, which was the first programmable calculating machine. Some people call that the first computer. Um, you've got You've got Touring himself working on a system that was directed by m Dr Tommy Flowers. It was a computer that no one outside of a certain group of people really knew about. It was called Colossus. Colossus, and that that was. It was so secret that that there were no contemporary reports on Colossus, and all the documentation to show how to operate Colossus was destroyed by the end of World War two. They didn't want anyone to find out how this machine worked, and it was used as a way of cracking codes and things like that. Yes, uh, I actually wrote a blog post about that in the tech Stuff blog not terribly long ago. Although you may not recognize it by its title. Um, somebody who Uh. They actually rebuilt Colossus in uh in the UK at Bletchley Park a few years ago. And uh, there's a musician who writes what they call chip tunes, which is using electronic equipment basically gaming equipment mostly, uh to write electronic music. Well, he actually went in and sampled sounds that Colossus made to make a record from it. And uh, you know, I happened to hear this BBC podcast all about it, and I learned that Colossus actually didn't crack the codes itself. It was calculating the positions of the wheels on the German code machines. It imagining that it took a room full of computer to calculate possible positions for the for the wheel, so that cryptographers could take that information and then use it to crack the codes. You know, I'm pretty sure you know an iPhone could well yeah, but at the time, but at the time, you know, it's very very important. In three, you've got the Harvard Mark one digital sequence controlled computer and the first computer bug. It was a real bug. Yes, it was a moth. There was a moth inside the Harvard Mark one and it touched a relay and it made the computer fail, and thus we get bug. Yes we didn't, you know, not a features in. That's when we finally have the invention of the transistor. Now it's gonna be a while before the transistors make their way into computers, but this is what eventually allows us to create computers that are not the size of your average gym. These these old computers were enormous. They took up huge amount of space. And this is all a very formative time in computers because it's a time when people are starting to rethink what computers could be used for. Um X, Arthur Burkes, Herman Goldstein, and John von Newman came up with a paper called Preliminary Discussion of the Logical Design of Electronic Computing Instrument uh and Jonathan's looking at me like, why are you bringing this up? Well, it's because a lot of people consider this the basically the document that started computer science, because it was discussing things like how data and instructions should be stored together and how instructions could be uh changed by other instructions, which is something that Conrade Zeusa didn't believe. He said that that was a horrible horrible thing and it shouldn't be allowed to happen otherwise. Uh, dogs and cats living together are um. But actually this is what allowed advanced computer programming to to come about, was people started thinking about computers and how they could be used, you know, with the concepts, and as the forties roll on and and people start thinking about what computers really could be used to handle. Uh, they they that was really the end of it being a computational machine, in the beginning of it being used for you know, other kinds of of uses. So let's say go up to that's when Touring creates the a C, which was the first programmable digital computer, the ACE. He also publishes the Turing test. So the Turing test is of course the famous test of finding out, you know, are you human or are you are you computer? Right? If if a computer is able to pass the Touring test, it means that it is able to fool a certain percentage of people into thinking it's another person. Mm hmm. Touring tests are also what we used to base the Capture system on. That's the system where if you have to you know, when you're creating a profile on a on a site and you see the jefas put in a certain sequence of letters and members, like the jumbled up type written looking letters and numbers with a line through it. That's a captured test, and that's designed so that it, theoretically anyway, is easy for humans to complete, but difficult for computers to complete. That's not always the case because people, you know, there are hackers out there who are clever enough to make computers that are clever enough to pass it. And then of course you have to come up with a new test, but that's all based off the touring test. Uh. In fifty six, IBM built the first hard drive it had had a five megabyte capacity and it cost one million dollars. Yeah. Yeah, my iPod has I think eighty gigabit gigabyte capzi. Yeah, it's a it's a lot less than a million anyway. Uh. And then by fifty nine, that's when you start seeing computers using transistors instead of vacuum tubes. So the yeah, transistor was invented in forty seven, and then in fifty nine, this is is when you finally start seeing them show up in computers. Um. But you know, the ones with the vacuum tubes just sound better, man, I think you're thinking about something else. So sixty one that's when you get the first integrated circuits commercially produced, And in sixty three this is this one. I like, I mean, you may have some more dates to fill in through here. Actually I skipped ahead again because I want to I want to read about this in more detail. Yeah, we would love to be able to do a series of these. Actually, that cat should be a spoken word album right there, because the early days of computing. We'll just we'll we'll pitch that. If you guys all write in and say we really want to hear that, then we get to do it. If you write it and say we really don't want to hear that, then I guess we'll move on to something else. But in sixty three Douglas Inglebart gets a patent for the computer mouse, and sixty four he builds it, and uh so that was the first appearance of the computer mouse. Actually, a few years later, um he shows up and uh shows a a not just a computer mouse and a keyboard, but a graphic user interface which kind of forms the basis of some operating systems we know and love today. At least. No, I actually don't actually don't know the answer to this, I'm asking you this question is was he involved with Xerox? Um. I believe he did play a part in that. I mean, in nineteen sixty eight he introduced the mouse, keyboard and Windows Interface system UM at the Joint Computer Conference in San Francisco. It wasn't until let's see, um, I got it written down here, let me chick. Seventy three was when Xerox developed Alto, which was the computer that had the graphics user interface. But then they decided there's no market for this. People aren't gonna want a computer in their homes, you know, trash it. So I'm not sure if Inglebard himself actually was involved in the in the development of Alto, but it was pretty clear that his inventions at least played a part in the production of Alto. Well again, never really went anywhere, but I would eventually show up in other products like I don't know, Windows and and Mac operating system. Well, it's uh, it's funny you would mention nineteen seventy three and how people wouldn't be interested in that, because that was an unintentional segue. UM. I noticed that there was a magazine called Radio Electronics that in three published uh information about something called the TV typewriter. It's basically a terminal you could use to uh connect to a mainframe and uh more or less a hobbyist project. But you know, these are the kinds of things that get people who like to tinker, the kinds of people who might read MAKE magazine. Now. Uh, you know, they were reading Radio Electronics back then, and and they were saying, mmm, I could use that. And uh, you know, in France right about the same time, uh, you know, there was the R two E made. Uh. They had their own that was the name of the company. They had their own micro computer that used an Intel eight O eight processor and they sold five hundred of them in France. Of course, nobody heard of them in the United States, So in the States that would be It wouldn't be until about nineteen seventy five when Ed Roberts does signs the Altar hundred yes for a micro instrumentation telemetry systems and fifty name for why are the computer manufact So he introduces this kit for three seven dollars. You can you can buy the kit and build your own personal computer. And it is wildly successful within the computer world. Well yeah, I mean they were on the cover of Popular Electronics. Saved the company they were they were basically a custom chip manufacturer, and they were they were tanking. Now what's interesting is back in uh you technically had the first personal computer. It was like the ken Back one, but that machine didn't really take off and it wasn't called a personal computer. Actually, Ed Roberts was the one who kind of coined that phrase the personal computer, and PC has been around ever since. Um. Also one that was when Pong was invented. Oh yeah, didn't have anything to play it on, but they had the game. Now, if you wanted to use the al Tear, you had to program it in assembly code. And there were a bunch of switches on the front and a bunch of lights and basically you'd flip switches and watch the output of the lights and uh yeah, exactly, not exactly know, uh you know, that was how you would get an output on the on the machine. And you might say, who would care about something like that, Well, the Homebrew Computer Club would care. There were a bunch of people in California who were interested in programming, and a couple of guys named Steve were members of that club. Yeah, Steve Steve Wozniak and Steve Jobs, who later went on to actually were Atari employees at one point but later went on to UH start Apple Computer and and in in v they were working on they were starting to build their own computers within UH in a garage, which is the way almost every technology company in California started. And there were a couple of guys out east to who were looking at the Altare for programming, UM, Paul and Bill Billy Billy Billy Gates and Paul Allen. I assume you're talking about. Yeah, so the Altaire actually has a place in the pantheons of computer history for both Apple and Microsoft. So, yeah, Jobs in Bosnia they're building computers in the garage, Gates and Allan they are building basic compilers, and that was kind of laying the foundation of Microsoft. I'll talk about compilers in a second when I do a quick rundown of the punch card system. UM. And then in nineteen seventy six, you have the Cray one super computer. Yes, actually the Cray. Uh it's it's named after was it was it? Seymour Cray, Yes, Seymour Crazy, I was right, Cray And no, he's but but Seymour Cray, it actually developed another computer, the world's first supercomputer actually way back in nineteen sixty four, um, and that was the c d C sixty six hundred. But the Cray one, of course left in the dust. That was, you know, twelve years later. So it's much more advanced computer, especially if you follow the logect of Gordon Moore, and it can leap tall buildings in a single bound. So that that's where I cut off. I'm like, all right, seventy six, we've got that. By then, you've got the Apple computer coming out. Um, you've got the genesis of Microsoft. We're really entering the modern era of computers. But um, that's where I was like, okay, I'm not going any further than seventy six. That could be a different podcast. Computers with the modern era. Yeah, that's that's really the transition because right around then and through their early es, IBM released the first personal computer, and uh, you know, the other the Windows system of using the graphical Union user interface and the mouse, you know in the the Macintosh and the Amiga and ther St and Windows and all the other computers that that started out with that, and uh, you know, that's sort of up to where we are today, more or less with refinements. Yep, well, let me hit this punch cards thing really quickly. We're running kind of along on this podcast problem. Let's obviously we're both passionate about this historical stuff. Um, who knows, maybe stuff you missed in history class will try and cover computing at some point and we'll be like, ha ha, beat you to it. So this I'm I'm pulling this information from a paper called Programming with Punched Cards, which was written by Dale Fisk, who was an engineer at IBM. This was talking about his his experiences back in the early seventies and if you wanted to write a program using punch cards, here's kind of how he lays it out. Writing a program began with a paper tablet of coding forms. Each page of the tablet had about fifty lines on it, and each line on the form would eventually be converted into a punched card and stowed away in a box with a bunch of other cards. So you would fill out one of these forms, you take it to a key punch operator, or if you knew how to use the key punch machine yourself, you'd use it yourself. You would encode each line into a card, and then you would have a deck which was called your source deck. Now, for programmers, this is the essentially the same as talking about source code. This is oh that humans can easily interpret. Machines aren't very aren't good at that they can interpret source code, but people can. And it's very important because the next step is you'd feed the source code into a computer that would translate it into object code. You would get an object deck of cards. Now that process, that's what you're putting your your source code through a compiler. That's what compilers do. They translate from source code into object code. So you're compiling your program. Now you have an object code. You have an object deck. This is what allows you to create to run a program on a computer. And he talked about the process of running this object deck through a computer, and then he got this big stack of paper and it essentially told him that everything went pear shaped. Shortly after the first couple of cards went through and it turned out that his source code, one of the source code cards was missing a comma, so he had to go back and insert the comma, redo the punch card, replace the one that had been put in there. Run the source deck through the compiler again, get a new object deck, try it again, same mistakes, another mistake rather, and this is the way they debugged programs in the early days. Like if you had not written it out properly, or if you had key punched it in incorrectly, then your program wouldn't work and you would have to locate the problem and put build a new punch card, or several new punch cards if if it were was a problem that that spanned more than one, and then run the object or the source deck rather through the compiler again. Get an object deck that, run that through the computer and see if your program worked. This is why people to write software for me. So yeah, and of course keep in mind there are no monitors. You're getting printed information here. You know, you're not seeing it in in the sense of the way we see things on our computers today. So that kind of gives you the rundown on what it was like to do to create a program on the punch card. Now, the paper that Dale Fisk wrote, it's about eighteen nineteen pages long, and it's a great read. It's not written in in super technical language, a lay person could easily follow it, and so I highly recommend hounting that paper down again. It's called programming with punched cards, so you can google that and find the pdf online. Now, um, that pretty much wraps up our our discussion about computers from the past, which is a good thing because we're running over thirty minutes now, all right, so let's wrap this up. Guys, if you have anything you want to say to us, if you have any comments or corrections or compliments or suggestions, you can write us. Our email address is tech stuff at how stuff works dot com. You can learn all about these computing systems on our website how stuff works dot com. And hey, if you if you have technical questions that you might need answers to very quickly, you might ask someone who can do more tech support type answers, because not that we don't want to get those questions, but we might be less helpful and it might take us longer to get to them. Yeah, it turns out we have a lot of stuff going on, so if your computer is acting funny, we're probably not the best people to talk to, but we do appreciate your email. You guys are swell, So with that said, I guess this wraps up another episode and we will talk to again really soon. For more on this and thousands of other topics, visit how stuff Works dot com and be sure to check out the new tech stuff blog now on the house Stuff Works homepage, brought to you by the reinvented two thousand twelve camera. It's ready, are you