Slashdot stories can be listened to in audio form via an RSS feed, as read by our own robotic overlord.

 



Forgot your password?
typodupeerror
Programming Technology

Grace Hopper, UNIVAC, and the First Programming Language 137

Posted by samzenpus
from the back-in-the-day dept.
M-Saunders (706738) writes "It weighed 13 tons, had 5,200 vacuum tubes, and took up a whole garage, but the UNIVAC I was an incredible machine for its time. Memory was provided by tanks of liquid mercury, while the clock speed was a whopping 2.25 MHz. The UNIVAC I was one of the first commercial general-purpose computers produced, with 46 shipped, and Linux Voice has taken an in-depth look at it. Learn its fascinating instruction set, and also check out FLOW-MATIC, the first English-language data processing language created by American computing pioneer Grace Hopper."
This discussion has been archived. No new comments can be posted.

Grace Hopper, UNIVAC, and the First Programming Language

Comments Filter:
  • Still faster than my first 8080....

    • by mcgrew (92797) *

      Faster than my first two computers, too, but neither of them weighed thirteen tons! Also, storage access would have been a much bigger problem than clock speed, seeing as how they used mercury switches to store bits.

      I found this article about Univac [usatoday.com] fascinating, an account of Univac vs. humans.

      ...Those circumstances set the stage for the election night dramatics of the Univac â" perhaps the most significant live TV performance ever by a computer. It might just be technology's equivalent of the first El

      • by Anonymous Coward

        Not mercury switches, mercury delay lines.

    • Your 8080 didn't spend most of its time waiting for instructions to pop out of the end of its delay line memory. (My first computer was also powered by an 8080, represent.)
    • by sjames (1099)

      Nevertheless, the 8080 probably ran faster due to lower memory latency.

      • Nevertheless, the 8080 probably ran faster due to lower memory latency.

        Also less down time to replace burnt out vacuum tubes.

        • by 91degrees (207121)
          Large scale valve/vacuum tube electronics were actually a lot more reliable that radios using the same technology. Heating and cooling does the damage. Keep the things running and they're more than good enough for the GPO's telephone exchanges in the 1930's. This was one of the arguments that had to be won for Colossus, but it was actually a lot more reliable than the bombes.
          • They were still quite unreliable. But it's my understanding that what they were doing was running them at reduced power for useful computations (which worked since even the tubes used in computers were always sort of high-power components, comparatively speaking, and you didn't actually need their full power to implement computer logic). Then, in maintenance periods, they'd run them on full power for a while, and replace those that burned out during that period. That is supposedly what actually made it poss
  • I would have loved to have one of her nanoseconds she use to hand out when asked how long was a nanosecond. I remember when she was on the tonight show with Johnny Carson and told that story. She use to keep a bag full of them with her all the time and would hand them out, when someone would ask how long is a nanosecond. One smart lady!
  • Grace put it so beautifully: "It is often easier to ask for forgiveness than to ask for permission".
  • Nanoseconds (Score:5, Interesting)

    by GlobalEcho (26240) on Sunday May 18, 2014 @10:19AM (#47031713)

    My mother was one of the first female programmers at Honeywell back in the `70s. Back then, IT companies recruited their programmers from the ranks of mathematicians (like mom).

    Grace Hopper was a big hero to her, and one of the things I remember best is mom coming home with a short length of wire given out by Adm. Hopper at a speech -- sized to represent the distance electricity would travel in a nanosecond.

    Mom is still coding, by the way, writing custom software for my dad's business in Python/Django/PostgreSQL. Dad complains that she's obsessed with the programming and won't do anything else. Sounds like me...thanks for the genes, mom!

    • Go Python/Django/PostgreSQL!
    • a short length of wire [...] sized to represent the distance electricity would travel in a nanosecond.

      You cannot see such a piece of wire. Electrons drift [wikipedia.org] at a speed in the order of 0.0002m/s, giving you a wire length in the order of 10^-13 meters.

      Electromagnetic waves "travel" roughly at the speed of light. But when someone talks about the travel of electricity, the thing that people think about is the flow of electrons, not the electromagnetic waves.

      • by tulcod (1056476)

        (on top of that, there are no electromagnetic waves travelling along a wire conducting DC current)

        • by rubycodez (864176)

          But there are no true DC currents, real current flow is not of constant amplitude and not of infinite duration in time. Therefore, real DC current in the real world always has EM waves associated with it.

        • (on top of that, there are no electromagnetic waves travelling along a wire conducting DC current)

          DC current is not used to transmit information. Even if your message is "00000000000000000..." you would use data compression, Manchester encoding, RS-232, or something else with an embedded clock or framing.

      • But when someone talks about the travel of electricity, the thing that people think about is the flow of electrons, not the electromagnetic waves.

        Speak for yourself, eh? I don't think it ever once even remotely occurred to me that someone meant the flow of electrons when they talked about the travel of electricity. I have always thought of the travel of electricity as the flow of the electromagnetic waves.

        (Note: I am not an electrical engineer and have not studied electricity intimately.)

        • by tulcod (1056476)

          I have always thought of the travel of electricity as the flow of the electromagnetic waves.

          Then how does DC electricity "travel" from your phone charger to your phone? (again, there are no electromagnetic waves, even though there may be fields. a wave is a changing field.)

      • They are 2x - 3x faster than copper signals. Those millisconds add up in financial trading.
        • not fiber. point to point laser and microwave links.

          I believe you are referring to ultra-low-latency trading.

          They prefer microwave links to fiber because the microwave signals propagate faster through air than light does through a glass fiber. Light travels through glass fiber at about 65% of c, which is also pretty comparable to the velocity of a electric signal in a transmission line (.65 to .75 c) (which is where Admiral Hopper ties in)

          Microwave signals propagate though air at damned close to the speed

      • by Anonymous Coward

        She knows that. See this:
        http://www.youtube.com/watch?v... [youtube.com]

    • Insurance companies and some science labs used clerks to make long calculations. The majority were woman. The "electronic computer" was a futuristic machine to emulate such people.
    • Or Grace hopper was wrong.
      Electric current moves in the range of a millimeters per second.
      A nanosecond long travel range would be invisible to the human eye.
      Perhaps you meant 'electric signal'?

  • by Anonymous Coward

    Long, interesting and informative and unapologetically technical essays like this are why I get up early and brew coffee in the morning,

  • by Anonymous Coward
    I love slashdot, and about every 10 postings there is someone ranting about "am I too old to be a programmer." Have some Grace, and do what you like to do. Grace Hopper is a real role model. Just because technology makes you feel like you are playing with toys, does not mean you have to grow up - just go out and play, and build something.
    • Easy to say. When Grace was around she wasn't competing against Indian, Chinese, Brazilian and Russian university students being pumped out by the thousands every year.

      • by Anonymous Coward

        Easy to say. When Grace was around she wasn't competing against Indian, Chinese, Brazilian and Russian university students being pumped out by the thousands every year.

        I really believe that coding is "in the blood." The problem with countries like India and China is that the economic rewards force people without the "knack" to go into the field -- and suck badly at it. So not only are you competing with someone who works for 1/10th of your salary but they suck at it but go to great lengths to hide that fact

        • by retchdog (1319261)

          you're not nearly as special as you think you are. if they are able to "hide that fact" successfully, it means that there is nothing remarkable about what you're doing.

          grow up, you pathetic fantasist. "coding" is not a particularly heritable property, and there is no fucking "knack". that was just something in a Dilbert cartoon.

  • not FLOWMATIC per se (Score:5, Interesting)

    by rubycodez (864176) on Sunday May 18, 2014 @10:27AM (#47031771)

    Grace's big contribution from the time wasn't the particular FLOWMATIC language but rather she conceived of the compiler. And note her languages were intended to be legible even to non-programmers, what an usual concept eh?

    • by Anonymous Coward

      Yes, regretted by programmers since, tasked with fixing the well-intentioned programs coded by non-programmers.

      • by Anonymous Coward

        Bah -- just remember who invented computer science: chemists, physicists and mathematicians!

        Unlike Richard Stallman, it was WWII heros like Grace Hopper that helped make us 'free' to make software!

        Thats right, computers were invented by 'squares' not by hippies and their 'information wants to be free' Gen X children ;)

    • And note her languages were intended to be legible even to non-programmers, what an usual concept eh?

      And that's how we ended up with COBOL.

  • by PPH (736903) on Sunday May 18, 2014 @10:54AM (#47031949)

    ... after finding an actual bug in a computer. Imagine how different [suvi.org] things might have turned out.

    • by rubycodez (864176)

      other critters have caused problems in electrical systems, we might be saying a snake or rat or spider.

      in not entirely unrelated concept, we have the molly-guard thanks to the toddler Molly who pushed the big red button on an IBM 4341 at UIUC twice in a day.

    • by ArsonSmith (13997)

      Not to mention all the work she did for Kung Fu.

      "When you snatch the pebble from my had you will be ready Grace Hopper."

    • by Coditor (2849497)
      The computer with a bug that was actually a fish is a new one to me. Imagine if debugging today was called fishing.
    • Re: (Score:3, Informative)

      by Anonymous Coward

      No she didn't.

      The journal entry reads "first example of an actual bug", which clearly indicates that the term was already in use.

      If you look in the OED, the first use of "bug" to mean a technical glitch of some sort dates from 1870-1899.

      To be fair, Hopper never claimed to have originated the term.

  • Yeah, but if you're wondering, it took about 1,000 clock cycles per instruction...

    • by rubycodez (864176)

      yeah they should have used red mercury instead of mercury in those delay line memory tubes!

  • "(1) COMPARE PRODUCT-NO (A) WITH PRODUCT-NO (B) ; IF GREATER GO TO OPERATION 10 ;
    IF EQUAL GO TO OPERATION 5 ; OTHERWISE GO TO OPERATION 2 .
    (2) TRANSFER A TO D ."

    What's wrong (if useless is wrong) with this code?

    Godz, I can't believe I'm trying to correct Hopper's code!

  • It weighed 13 tons, had 5,200 vacuum tubes, and took up a whole garage

    InB4 yo mamma.

  • I didn't know Grace Hopper had anything to do with Plankalkul.
  • How can you fit that much spaghetti in 17 lines??

    • by Tablizer (95088)

      Now a-days we use OOP to get similar bloat:

      NORMAL

      print(a + b)

      BLOATED

      am = new math.ArithmeticManager()
      opA = new math.Operand((float) a)
      opB = new math.Operand((float) b)
      am.addOperand(opA)
      am.addOperand(opB)
      am.operator = new math.operators.Addition()
      am.executeMathOperation()
      system.io.output.print(am.mathOperationResult())

  • While the section on Admiral Hopper looks correct to my knowledge, there were some hitorical flaws.

    The UNIVAC I was produced after Remington Rand purchased EMCC, though Grace Hopper did work for EMCC prior to its acquisition a year after she started work there. The UNIVAC I was built by Remington Rand. Four years later, Remington Rand merged all three of their computer related operations into the UNIVAC division. The following year Remington Rand merged with Sperry to become Sperry Rand and the UNIVAC divi

  • Konrad Zuse, who also built the first Turing-complete computer, designed the first high-level computer language, Plankalkul, in 1945 (though no compiler was implemented until 1998.)

  • It weighed 13 tons ...

    and smelled like a stake! UNIVAC-nero! (whip crack)

...there can be no public or private virtue unless the foundation of action is the practice of truth. - George Jacob Holyoake

Working...