Photos: 100 years of computing from punched cards to floppy disks

Photos: 100 years of computing from punched cards to floppy disks

Summary: From punched cards to electromechanical machines - a look inside the IBM museum charting the rise of the company and computing.

TOPICS: Hardware

 |  Image 13 of 15

  • Thumbnail 1
  • Thumbnail 2
  • Thumbnail 3
  • Thumbnail 4
  • Thumbnail 5
  • Thumbnail 6
  • Thumbnail 7
  • Thumbnail 8
  • Thumbnail 9
  • Thumbnail 10
  • Thumbnail 11
  • Thumbnail 12
  • Thumbnail 13
  • Thumbnail 14
  • Thumbnail 15
  • Dynamic RAM from 1966.

    Photo: Nick Heath / ZDNet

  • A floppy disk from 1971.

    Photo: Nick Heath / ZDNet

  • Solid logic technology from 1974.

    Photo: Nick Heath / ZDNet

Topic: Hardware


Nick Heath is chief reporter for TechRepublic UK. He writes about the technology that IT-decision makers need to know about, and the latest happenings in the European tech scene.

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.

Related Stories


Log in or register to join the discussion
  • "The System/360 family of computers was announced by IBM in 1967" - 1967??

    Try 1964 - the System/360-->zSeries is 50 years young in April 2014!
    Lord Minty
    • Thanks

      Thanks for the heads-up, the card in the museum said 1967 but elsewhere online it says 1964. I've changed to around the mid-1960s in the piece.
      Nick Heath
      • The release was 1964...

        The 1967 date you got likely refers to the purchase date.
        • 360/30 was my first mainframe with disk OS

          Ran a 1401 (or 1403) in 1963-64 I believe, it was a Tape OS at county govt center. Then a 360/30 at a bank, a 360/50 at a Navy Supply Depot, also a 1620 at a college once a year for high school class scheduling. My first "computing" was a 407 Accounting Machine, with a stand alone card duplicator, and interpreting keypunch machine, and a punch card sorter. I learned control panel wiring from the IBM field engineer when I was a high school student on a work/study program. Mid career we used Burroughs 3700 computers, then Unisys. When PC's became popular the mainframes went away as networks became stable and popular. Still working FT, about to retire. Hope to visit the Computing Museum this year in Santa Clara, Calif, adjacent to Moffet Filed and NASA. It's been a fun and interesting ride.
  • Supercomputer

    Hay that hardware with the tubes, would make a great Supercomputer for Windows 8.x.
  • It isn't IBM 29 card punch.

    The leading 0 was significant.

    It is an IBM 029 card punch.
    • If you mean the photo

      ... it's an 026, although as you can see there it was labeled "26" just above the punch station.
      • Different machine.

        The 029 had a print line (0), 2 coding lines (the unmarked top two), and 9 digits.
        • Actually, the 029 punched the same cards.

          The 024 (non printing) and 026 (printing; see the bulge over the column to be punched in the picture, containing the print mechanism) used old fashioned leaf spring relays and, for the heavy power surges, a vacuum tube of the kind used to drive radio speakers. The 029, with sleeker looking 1970s external design (and non-movable keyboard), used large printed circuits with lots of reed relays used as flip-flops, and a few transistors. The column-by-column programming, rather than using a program card wrapped around the drum (behind the flip-open lid in the center), was done by reading the card through the duplicating read station and storing its image in reed relays. It also had the ability to define, using the program card, a right justified numeric field with a fixed length of 3 to 8 columns. The digits entered by the operator were stored in reed relays, then punched with the correct number of leading zeros when the operator moved to the next field. No more counting leading zeros!

          There are no pictures of the 029 in this gallery. It was much better looking and easier to work with (and, in the purely mechanical sense, to work on).

          Incidentally, all 80-column keypunches punched the same format: the top two rows were usually (with general purpose preprinting) not labeled, and were used for "zone" or modifier punches to turn numbers into letters, or to add a sign to a number, depending on the application. The top (12 or Y) was an explicit positive sign on the low order position of a number. The second (11 or X) was an explicit negative sign on the low order position of a number. The "0" punch by itself was the unsigned digit zero, of course, and in the low order position could have a 12 or 11 added. Combinations of any one of 12, 11, and 0 with any one of 1 through 9 made the capital letters, and various other combinations represented punctuation symbols.
          • Now, about the sound...

            The 026 always sounded soooo much better than the 029 which sounded like some cheap apparatus made of stamped tin... The 026 on the other hand sounded heavy and solid while producing chad.
          • The keypunch "ladies" may have felt differently.

            You and I probably used the keypunches in a room with no more than three, student/programmer use on demand, and possibly in the computer room with lots of other noise, and sporadically throughout the day. The professional keypunch operators, however, spent 8 hour shifts in a room with 20 or 30 machines all being used at once (except when someone had a sick day). They probably preferred a quieter noise even if it didn't sound as "heavy." One summer I interned (OK, summer job home from college) working as an apprentice repair tech, and I worked on both kinds of machines, in the room where the full time people were using them. The gentle clatter of the reed relays was much more tolerable, even to me, than the loud banging of the heavy tube-driven solenoids.
          • Card punches drove us crazy

            We replaced the card punches for production data entry and eventually card decks to compile programs with a key to disk system (I forget the manufacturer). It could also write mini reel tapes for input as a card simulator on the mainframes. Key to disk was faster, quieter, more accurate and got rid of card jams! We had 10 keystations on one processor. This was in the 70-80's.
  • Hollerith machines

    Only real issue, is the failure to mention the boon of Hollerith machines to Nazi Germany and what this "census" information was used for in regard to the German Jews.

    It has been argued that Germany and WW2 saves IBM.
    • If anything, IBM was a "dupe" of the Nazis, not a true accomplice.

      The biography of Thomas Watson makes it clear that he had an idealistic "one world" belief that national rivalries would fade away with enough world commerce. Because of this, he could not bring himself to believe that Nazi Germany or any other regime was truly EVIL, and so overlooked (and at the corporate level, may have honestly been ignorant of) the kinds of uses to which his company's equipment was applied. It is said that RAF pilots drafted out of IBM loved their company (to which they expected to return after the war) so much that they "missed" assigned IBM factory targets and bombed other nearby targets instead.

      But we must also note that the same tabulating technology was used for good by others, such as the U.S. Social Security Administration. The technology came to be so vital for administering the pension program and other government functions that the IBM monopoly on purchasing the preprinted CARDS for use with the machines led the government to encourage patent-avoiding innovations by rival typewriter maker Remington Rand, which came out with machines that used two groups of six rows of 45 CIRCULAR holes to encode 90 columns with 6-bit binary codes rather than 12-bit 1/2/3-out-of 12 codes (NOT shown, of course, in the IBM museum). After the war, when IBM began selling commercial computers, Rand came out with the UNIVAC (for Universal Automatic Calculator), which at first was a more famous name than IBM. After the patent on the card format expired, Univac (no longer an acronym) made some equipment able to read and process both cards (the physical size common to both formats was derived from the size of U.S. currency in the 1890s). Most people in my generation are familiar with the prevalence of "IBM cards" as bills and other "turnaround" documents which would be returned to a company in order to save time and cost processing payments. Federal treasury disbursements such as tax refunds, Social Security checks, and government employee paychecks were printed on punch cards for years, along with U.S. Savings Bonds (an obsolete small-ticket loan to the government, sold to individuals), and even toll road entry booth tickets to be turned in at the exit booth to compute the correct toll. We all remember the warning not to "bend, fold, staple, spindle or mutilate" the cards.

      The function of "turnaround documents" today, when it is needed, is now performed by bar codes, QR codes, and other printable codes. The last public use for punch cards was their use in voting, which was spoiled by the inferior equipment used to punch and read them. Rather than using a keypunch as shown in the first picture, ballot cards were perforated at the subset of hole positions allowed for voting use (about a third of the 960 possible holes, in a staggered pattern) and punched by hand with a frame which aligned ballot questions and candidates with dies through which a pin could be punched. This led to the infamous "hanging chads" that plagued the 2000 Presidential election. And since cards were not the primary input medium for OTHER applications anymore, cheaper, slower, and rougher-handling equipment was used to read them than what was available in the heyday of punched cards (one of the classic machines was so gentle in handling the cards that operators said it could "read the rags they were made from"), resulting in "destructive" recounts and thus changing history.

      IBM continued to promote a "one world" philosophy, amended by the lessons learned from the Nazis, and in the early 1970s formed a division called World Trade Corporation, or WTC, which was the original owner of the WTC towers in New York. Ironically, a business venue originally dedicated to peaceful world trade became the target of fanatics who did not believe in one world, unless they controlled it.
      • thankyou

        well informed and an enjoyable read! why are you not writing for ZD !
        • Perhaps IBM was a dupe, Watson "was not".

          Well informed? You should be well informed in order to know that he is. Are you?

          Mr. Watson was awarded a medal by Hitler himself as an acknowledgment of IBM's services to the Third Reich. How many companies do you think received this award?

          He was really proud of this medal and only gave it back when strong pressure from the US public opinion and government forced him to.

          Do you really think he was ignorant of the Nazis’ ideology and practices? He just did not care as long as money came in his pocket.

          Of course Mr. Watson's position must not be viewed as IBM people's position.

          If you want to be well informed, read this book:

          José Fernando Blanco
          • Only one other of which I am aware: Ford

            Specifically, Henry Ford himself, but not for the car; for publishing an English translation of the fictitious "Protocols of the Elders of Zion" created in czarist Russia to justify pogroms.

            The biography which I read many years ago may have been published before the Holocaust connection came out. It is also possible that, like many other American industrialists and bankers, Watson admired the non-extermination parts of the Nazi ideology and was in a state of denial about the Holocaust part. That number would also include a Senator who was the father of a President and grandfather of another, banker Prescott Bush, who was STILL financing the Hitler regime AFTER the US went to war with them, forcing FDR to order his bank seized (one reason the family still hates Democrats, perhaps).

            Americans in general, except avowed socialists and communists, thought Hitler was generally OK but a little nuts, until he started the war. Even then they did not believe anyone could ACTUALLY want to kill all Jewish people until the evidence was undeniable, just as many Americans do not believe the conservatives today could ACTUALLY want to kill Social Security, Medicare, Medicaid, the minimum wage, all labor unions, OSHA, EPA, unemployment insurance, wage and hour laws, and even the direct election of US Senators.
      • RAF Bombing

        "It is said that RAF pilots drafted out of IBM loved their company (to which they expected to return after the war) so much that they "missed" assigned IBM factory targets and bombed other nearby targets instead."

        The RAF practice night-time saturation bombing. The weren't picking out targets, only areas.
        • RAF pilots darfted from IBM - it could not have happened.

          There is lot of the usual nonsense about IBM in these replies but that one about RAF pilots is new one on me. RAF pilots could not have been "drafted out of IBM" because IBM United Kingdom did not exist until 1951.
  • Fascinating

    It's simply incredible just to see how far technology has come. Fantastic post.
    James Stevenson