The Sword and Laser discussion

Part of the fun of old sci-fi is spotting all the places where reality has surpassed the technology in the story. For example, at one point Mike claims he has 10^8 bits in free memory for recording conversations. If you plug that into Google, you'll find it comes to a bit less than 12 megabytes. I suppose if Mike is transcribing the conversations into plain-text, that's plenty, but it does seem a bit paltry for a super computer of the future.

Sean wrote: "Part of the fun of old sci-fi is spotting all the places where reality has surpassed the technology in the story. For example, at one point Mike claims he has 10^8 bits in free memory for recording..."
Yeah it's interesting how the SciFi authors of that time missed the boat as far as computer technology. Heinlein underestimates the power, memory and ubiquity of computers today and overestimates what could be done with that power. Even the pre-awake Mike can do more intelligent decision making tasks than computers today. Of course we still have another 65 years to get there.

I think it shows how the future is difficult to predict. In the 1960's the computers were the tools of universities and big businesses. They were huge boxes tucked away behind doors. You often never interacted with them directly but instead submitted your code to the pseudo preist hood of computer operators who ran it and handed you back the result. This type of usage constrained the uses to which computers were put and therefore constrained peoples imagination. So what if they were getting smaller and cheeper? They are still just big calculating machines. Why would average joe want a big calculating machine in his house?

Changes in technology have unforeseen results that's what so exciting about it.

However I think the larger truth is that Heinlein really just needs a plot device to deal with the mundane details of the revolution. Mike is this plot device. Heinlein is really more interested in the political and social philosophy he is advancing in the book. So having a super computer that can easily solve a lot of the details of the revolution is convenant. The fact that Mike is also a major and interesting character acts to distract the reader from this.

Actually, I don't think it should at all be difficult to anticipate future computer memory sizes or file sizes for a certain medium, but unfortunately writers are not not engineers or scientists.

The author would merely have to conceive of a manner of digitally encoding video of certain resolution with a certain number of bits dedicated to RGB recorded at a certain bitrate for a certain length of time. And do the math.

... just for the sake of argument: maybe Heinlein thought that code had to be optimized?! (Instead of bloated auto-generated overhead-heavy useless... etc etc - *five-minutes-later* ... from 'developers' who are fine with copying and clicking their noodles of code-objects and pasting them to these gigantic bowls of steaming program goo, everybody today seems to call 'apps' for no real reason? Remember, those were the days of ALGOL, COBOL, FORTRAN, everybody understood ASSEMBLER, hehe...

Alex, The days of Heinlein had COBOL and probably FORTRAN as the highest level languages. In the sixties, you did Assembler and machine language. Then in the seventies came Basic on the Altair and many stories used its structures in stories. It certainly was much easier.

Magazines like Creative Computing let people know of Basic, and David Ahl (editor) still has some of the best robot cartoon vehicles. I still remember the robot chiding another robot saying: "Oh yeah?, Well your mother has nobby tires!"

I remember programs where I had a total of 4K total adress space. And a 256 byte page size.

Thank you, for the insight, Louis. I am not old enough (42) to know first hand (something I rarely say these days).



Maybe this is a good moment to plug the ACM SIGPLAN "History of Programming Languages" Conferences (HOPL) book versions to fellow Sword and Laser readers? Fascinating stories, about how the pioneers brought technology and concepts to light, we, or some of our machines, still use. Big companies are still hiring COBOL programmers (... if they can find any).

Sure more fascinating read than some big novels, IMHO :)

Lepton wrote: "Actually, I don't think it should at all be difficult to anticipate future computer memory sizes or file sizes for a certain medium, but unfortunately writers are not not engineers or scientists. ..."

Lepton, There was no concept of "video" or "digital encoding" in 1966. Main frame "super" computers of that time had less processing power than your iphone does.

In 1966 the CDC 6600 super computer had 65,000 60-bit words of memory (less than half a gig), with a clock speed of 100 nanoseconds (1 cycle every 100 nanoseconds).

The cheapest iPhone today comes with 8 gig of memory with a CPU speed of 412mHz (approx 0.412 cycles per nanosecond or 41.2 cycles per every 100 nanoseconds)

http://www.cisl.ucar.edu/computers/ga...

P.S.

Just smack me in the head if I missed your sarcasm tag :)

Isn't all data digital where computers are concerned?

In 1945, Arthur C. Clarke anticipated a mass communications network based on satellites.

There is no excuse for lack of imagination where science fiction is concerned.

The computer you reference ran FORTRAN 66 which explicitly supported these data types:

INTEGER, REAL, DOUBLE PRECISION, COMPLEX, and LOGICAL

which by definition are represented in bits. Ultimately, the continuous numerical types are discretized into bits.

Sounds digital to me. Sounds encoded to me.

John wrote: "10^8 must have sounded absurdly humongous when this novel was written, and I'd guess that sense of "wow" was what Heinlein was after with that number.

It's fun to compare their envisionings with how things actually worked out. SovUnion still around, but United States not so much. Moon bases. Sentient computers with less memory than an iPhone. Trajectory computers only available at Peipeng or Princeton. No global data network. No cellphones.

Oh, yes...and all the most efficient engineering and manufacture being done in the Chinese markets.

NASA's trajectory computer used for the Apollo program had 4K of memory. It gets featured in history write-ups on occasion. One of those seat-and-desk half-a-kitchen jobs. And TARDIS blue.

Something fun:

Clarke didn't just anticipate communication satellites, he created their principals. He worked out that when a satellite is 22,000 miles up, the orbit is geosynchronous--and if the inclination is zero (directly over the equator) the orbit is geostationary--always in the same relative place in the sky. More importantly, he's the one who suggested that might be a useful thing. So we call that doughnut region of space the Clarke Belt.

Heinlein revisits computational power and memory capacity in other books and radically underestimates them there, too. He also liked throwing in gee-whiz big numbers in, too. Things that would take weeks or months by hand; something someone with a computer could work out and send him.

The thing is...if he'd been right on the money, I don't think people would have geeked out or gee-whizzed. They'd have Comic-Book-Guyed the story to death with, "*ACTUALLY* I am assured that is impossible since we can't possibly fabricate that quantity of core memory in a space that small."

Realize also: In the Harsh Mistress days, analogue computers were still widely of interest, and digital computers were not necessarily binary. You might recall Mannie referencing Prof's cell model as resembling "a ternary logic." Computer Science was playing with tri-state logics in those days, too: Positive, negative, and ground states.

Can't accuse Heinlein for lack of imagination. Just being off by a few orders of magnitude. Remember Clarke, who had all that vision of orbital mechanics, had us well established on the moon and heading to Jupiter by 2001.

Gotta cut these guys some slack. They may not have foreseen the future to the last decimal point, but they did better with the future than some of us do with the past.