• JJ (unregistered)

    I always thought Thinking Machines were supposed to be really great computers. Guess not.

  • (cs)

    It's a good thing the machines could think, since apparently none of the humans could.

  • my name is missing (unregistered)

    I think the machines were smarter than the people.

  • (cs)

    So, what's the actual WTF here? That the original contributor went to a research lab and expected the projects to have real world relevance? Or that Thinking Machines was apparently trying to be a genuine commercial company rather than just admitting it was a purely research-driven organisation? Or that the government were dumb enough to fund the organisation? Or that the government were dumb enough to pull the plug rather than continue funding Thinking Machines but making it work on its own projects? Or is this one of those posts that isn't meant to lnclude a WTF, and is just an interesting story from real-world IT (or, in the case of Thinking Machines, theoretical-research-world IT!)?

    Whatever it is, it's a nice, if unsurprising, story. I'm ashamed to say that if I'd had that interview, i think I probably would've been a Believer - the chance to work with technology that cool without any real world connections? Show me the money...

  • ratis (unregistered)

    (print "Who in their right mind would use LISP?")

  • (cs)

    So, this is how the Butlerian Jihad started.

  • (cs) in reply to JimM
    JimM:
    So, what's the actual WTF here? {list}
    None of the above. Clearly, the WTF is the absurd interview procedure.
  • (cs) in reply to JimM
    JimM:
    So, what's the actual WTF here?
    The fact that the article doesn't give us any of the recipes from the cookbook.
  • ratis (unregistered) in reply to FredSaw
    cook book:
    Open box of {name of recipe}, add water to form sloush mixture. Blob on plate and claim it is {name of recipe} to petty employees.
  • SomeCoder (unregistered) in reply to JimM
    JimM:
    So, what's the actual WTF here?...

    Probably all of the above, I'd say.

  • (cs) in reply to NaN
    NaN:
    So, this is how the Butlerian Jihad started.

    lol... I was thinking 'where have I heard the term 'thinking machines' before...

  • The Spirit of the Electoral College (unregistered)

    It reminds me of Google interview stories.

  • (cs) in reply to dpm
    dpm:
    JimM:
    So, what's the actual WTF here? {list}
    None of the above. Clearly, the WTF is the absurd interview procedure.
    Actually, if someone asked me for interview and when I got there I was told that I'd get to spend five days talking to existing developers, eating my lunch in their gourmet restaurant of a canteen each day, I don't think I'd be thinking "WTF?". Although I'll give you that Having to be approved by all the head developers and the CEO to get a job is a bit of an arse...
  • Some Guy (unregistered)

    ... and nothing of value was lost.

  • Rune (unregistered) in reply to JimM

    Well, I should expect some real world relevance in there, else there would be no research labs whatsoever, cause their duty is to research stuff that can be used in 'the real world'.

    Furthermore, I would decline, I like cool technology, but playing around with cool stuff gets boring after a while when you don't have any appliance for it.

    Picture yourself in a really cool car, a big lambo or something, and you sit there, flashing the lights, honking the horn, perhaps rev up a bit. But then, after 5 minutes or so, you get bored. Without your drivers license it's really no fun at all.

  • CEO of Doom (unregistered)

    At first I thought that someone really had an ax to grind with the CEO. Then I found this article:

    While the company was sinking, she focused her attention on putting out a cookbook with recipes from the company's now-infamous cafeteria. Increasingly paranoid, she had a video camera aimed at her personal parking spot and, by some accounts, made people take meetings with her in her parked car. She hired a bodyguard, telling her colleagues that she had received death threats.

    Some members of Thinking Machines' board suddenly seemed to realize that the person who had been running the company all those years had no business skills. The board discussed dumping Handler, but she managed to get her biggest enemies there kicked off.

    -- http://www.inc.com/magazine/19950915/2622_pagen_5.html

  • (cs) in reply to dpm
    dpm:
    JimM:
    So, what's the actual WTF here? {list}
    None of the above. Clearly, the WTF is the absurd interview procedure.
    About 20 years ago, I interviewed at ATT/Bell Labs. It involved meeting 16 groups in 3 locations over 2 days; 1 hour with each group. In each group, you'd meet four people for 15 minutes each. Every single person spent the first 13 minutes asking you to essentially verify everything on your resume, and virtually no time seeing if you actually knew anything, and the remaining few seconds walking you to the next office.

    By the end of the second day, I had repeated my life story 64 times and just didn't give a flying f--- any more.

    After all of that, HR asks you which of the 16 positions you want. Then you go home, and they see which, if any, of the 16 groups would like to hire you.

    If the job you picked was offered, you're in. If you didn't pick one of the jobs that were offered (eg: if you picked # 1 (not offered) but 2-15 were offered), you're out.

    Somehow, I think I dodged a bullet on that one...

  • vlad (unregistered)

    This would be a massively-parallel interview process, no?

  • Doc Monster (unregistered)

    Reminds me of the saga of Systems Concepts and the Mars project, as told in the Jargon File:

    http://www.catb.org/jargon/html/M/Mars.html

    Particularly the epilogue: If you want to play in the Real World, learn Real World moves.

  • NeoMojo (unregistered) in reply to vlad
    vlad:
    This would be a massively-parallel interview process, no?
    only if they were interviewing 25 candidates at once.

    As it reads, this would be a extensively thorough, serial interview process.

    I'd be interested to know the motivations of the developers, if it wasn't for Science or Business. Maybe it's for religious reasons, or for candy.

  • Anonymous Scheme weenie, thinking of registering... (unregistered) in reply to NeoMojo
    NeoMojo:
    vlad:
    This would be a massively-parallel interview process, no?
    only if they were interviewing 25 candidates at once.

    Beat me to it. As it is, this sounds more like a cult-of-groupthink recruitment and less like a job hire. The Bell Labs one, a few comments back, sounds like an actual job hiring system with a terribly cargo-cult interview process.

  • Ozz (unregistered) in reply to Rune
    Rune:
    Picture yourself in a really cool car, a big lambo or something, and you sit there, flashing the lights, honking the horn, perhaps rev up a bit. But then, after 5 minutes or so, you get bored. Without your drivers license it's really no fun at all.
    But that's when the fun REALLY starts. Without your drivers license you have nothing to lose #;-D
  • (cs) in reply to Anonymous Scheme weenie, thinking of registering...
    Anonymous Scheme weenie...:
    As it is, this sounds more like a cult-of-groupthink recruitment...
    1. Forgot to log in.
    2. I meant that the WTF sounds like a cult-of-groupthink recruitment, not the GP comment.
  • David C. (unregistered)

    Sounds more like they're running a cult than a computer company.

    I've worked at startups where everybody wants to interview everybody, and where there isn't a very good business plan, but when the primary focus of the interview is to impress the candidate with existing products instead of trying to determine if he has useful skills, then something very screwy is obviously going on.

    Of course, given the fact that the CEO was a paranoid lunatic (and probably in need of psychiatric treatment), this doesn't surprise me much.

  • (cs) in reply to JimM
    JimM:
    So, what's the actual WTF here? That the original contributor went to a research lab and expected the projects to have real world relevance? Or that Thinking Machines was apparently trying to be a genuine commercial company rather than just admitting it was a purely research-driven organisation? Or that the government were dumb enough to fund the organisation? Or that the government were dumb enough to pull the plug rather than continue funding Thinking Machines but making it work on its own projects? Or is this one of those posts that isn't meant to lnclude a WTF, and is just an interesting story from real-world IT (or, in the case of Thinking Machines, theoretical-research-world IT!)?

    Whatever it is, it's a nice, if unsurprising, story. I'm ashamed to say that if I'd had that interview, i think I probably would've been a Believer - the chance to work with technology that cool without any real world connections? Show me the money...

    Except that Thinking Machines really wasn't a research lab-- it was a real company laughingly trying to make money while everyone treated it like a university research lab. Sure, I'd jump at an opportunity like that too, but it sounds like the well dried up pretty quickly for them. So in retrospect, fleeing was a good move for Andrew G.

    But I agree that the real wtf was the interview process. How did they ever manage to hire anyone?

  • (cs) in reply to FredSaw
    FredSaw:
    It's a good thing the machines could think, since apparently none of the humans could.

    I just had a funny mental image of a bunch of inept goof-offs sadly unplugging a self-aware super computer "when the bubble burst". "Sorry CONMACH, we have to put you down"

  • i wanted some (unregistered)

    I was involved in some theoretical physics research at the time.

    We could have really made use of some hardware from Thinking Machines but could never get the funding to afford it.

  • (cs) in reply to Rune
    Rune:
    Well, I should expect some real world relevance in there, else there would be no research labs whatsoever, cause their duty is to research stuff that can be used in 'the real world'.
    Some labs have the duty of satisfying our curiosity (which only kills the cat if you're curious enough to open the box.) The fact that they happen to invent useful things along the way is a bonus. It's called basic research. Of course, you need money from Daddy, and it's nice if your daddy isn't a military agency.
    Rune:
    Picture yourself in a really cool car, a big lambo or something, and you sit there, flashing the lights, honking the horn, perhaps rev up a bit. But then, after 5 minutes or so, you get bored. Without your drivers license it's really no fun at all.
    Some people would find it fun to whiz around a race track really really fast (but much slower than the speed of light) and winning trophies, even if they were prohibited from using the car to actually get anywhere.
  • Captaffy (unregistered)

    Who on earth would go back for a second day of interviews upon learning about their hiring process? Did anyone actually get hired at this company?

  • Anonymous (unregistered) in reply to dpm
    dpm:
    JimM:
    So, what's the actual WTF here? {list}
    None of the above. Clearly, the WTF is the absurd interview procedure.
    I had an interview with UPS a year ago. I had applied for an IT position. I talked to about 3 people for 30 mins each, the first 2 people seemed flustered when I went on about my experience with computers and programming, etc, but didn't say anything about why. The last interviewer was nice enough to stop me and tell me that they were looking to fill a management position. I BS'd that software engineering skills can apply to management, but I was happy to leave that interview. What can brown do for you? Well, they can't match up resumes to positions.
  • Jim (unregistered)

    Being at a school in the Boston/Cambridge area, I can add that a Thinking Machine was a large black cube covered with red blinky lights/LEDs evenly spread over each of the four sides. It was an nice thing to look at. About 8 x 8 x 8 feet square. The front-end machines on the one I knew of were a Sun Microsystems 4/280 running Sun/OS and some DEC model running Ultrix. Both were in standard 19-inch racks. Biology related scientists used it.

  • (cs) in reply to NSCoder
    NSCoder:
    Of course, you need money from Daddy, and it's nice if your daddy isn't a military agency.

    Most of the time it's DoD-dy who's paying.

  • Andrew (unregistered) in reply to ratis
    ratis:
    (print "Who in their right mind would use LISP?")

    LISP can be executed in parallel very easily. There are often independent sub-expressions to hand-off to separate processors.

    For example, an AI "knowledge tree" (written in LISP or Prolog) spans several competing goals. The "first goal" is (probably) the simplest computation. A parallel system can try each goal simutaneously, and return the first goal to succeed.

    The vector "dot product" is another good parallel example. For vectors a & b, a "dot" b = SUM(a[k]*b[k]). Each product term can happen at the same time. a = (1 4) b = (2 5)

    LISP: (+ (* 1 2) (* 4 5)) CPU1: (* 1 2) => 2 CPU2: (* 4 5) => 20 CPU3: (+ 2 20) => 22

    Of course, modern Fortan has the "dot product" vector function built-in. On multi-core hardware, Fortan should run in parallel with no code changes.

  • Steve-o (unregistered)

    It is my understanding that, as part of the agreement, the administration sold off the assets to recoup a portion of the losses. Those same hardware assets later resurfaced and are now sold under the new "eMachines" brand name. Brilliant!

  • Just Some Guy (unregistered) in reply to Captaffy
    Captaffy:
    Who on earth would go back for a second day of interviews upon learning about their hiring process? Did anyone actually get hired at this company?

    I guess that depends on how badly you want to work on one of the coolest pieces of computing hardware in the world at that time. If it's your dream job to crank out business logic, then no, you probably wouldn't go back for the second day.

  • (cs) in reply to Rune
    Rune:
    Picture yourself in a really cool car, a big lambo or something, and you sit there, flashing the lights, honking the horn, perhaps rev up a bit. But then, after 5 minutes or so, you get bored. Without your drivers license it's really no fun at all.

    Actually it would be more like driving a Lambo around on a race track all day long, going as fast as you want and driving however you want to. Except the chicks couldn't see you. No, the chicks could not see.

  • Anon Fred (unregistered) in reply to David C.
    David C.:
    Of course, given the fact that the CEO was a paranoid lunatic (and probably in need of psychiatric treatment), this doesn't surprise me much.
    How often did Sheryl threaten to quit?
  • Captaffy (unregistered) in reply to Just Some Guy

    A smart person would have to want it pretty badly in order to look past the fact that they basically have no chance of getting the job.

  • Saaid (unregistered) in reply to vlad
    vlad:
    This would be a massively-parallel interview process, no?

    Massively-Serial. They were held one at a time, one on one.

  • Steve (unregistered)

    Okay, a lot of that interview story was pretty boguscomputing. The interviewee probably was lucky that he didn't get the job. He definitely would have been a square peg in the canonical round hole.

    I'd've dug working there immensely. My recollection of the mid to late 1980s was that it was an exciting time to be in computing.

    People were trying new things, often totally off the wall, wacky things, with respect to architecture and (gasp!!) operating systems (anyone remember Very Long Instruction Word architectures?).

    A lot of it didn't work, didn't work well, or solved non-existent problems, but still, it was a whole load of fun to get to play in and with all sorts of bizarre environments.

    I worked briefly at NASA Ames Research Center where we had a 8K node CM-1 and I know that some interesting and perhaps useful work was done on it. I recall attending a two-day conference on the CM where all sorts of applications, both computer science theoretical and real world, were discussed.

    Yes, the CM-1 was rather limited in the languages you could use (I believe that in addition to *Lisp or "StarLisp", there might have been an experimental C language -- I seem to recall talking to someone about it from TMC) but it was a visionary architecture and the contortions one would have to go through to map FORTRAN onto it would have been remarkable.

    The later systems, starting with the CM-2, by the way, did have floating point units, though they were Weitek "strap ons".

    I'm only sorry that I didn't get to play more on the machine than the brief introduction I had.

    Beats the hell out of the Windows/*NIX/Mac world in which we're living today.

  • Mumbly Joe (unregistered)

    If my memory is correct, the CM-5 could most certainly run FORTRAN programs. The only other language I remember was C* (C-star) which was like a parallel version of C.

  • Anon (unregistered)

    So did they actually tell the candidate before hand that the interview process would (or at least could) last 2 days? The article seemed to suggest that they hadn't. What if you had something else to do the next day?

    "Oh we'd like you to come back tomorrow for more interviews"

    "Sorry I can't, it's my mothers funeral"

    Very inconsiderate.

  • Chris (unregistered) in reply to Steve

    Yes, there was a C dialect that ran on the CM's - C*. I've worked with the team that developed it, and its actually an interesting, if cumbersome way of dealing with the matrix computing problem.

  • Paul Steckler (unregistered)

    In grad school, I took a parallel programming course where we used a Thinking Machines CM2. My course project was written in C*.

    I used one of their visualization packages to display results and, as I recall, I found a bug in the line-drawing primitive (or something like it).

    -- Paul

  • CoderDevo (unregistered)

    I was a system operator at a site that had a Thinking Machines CM-2 followed by a large CM-5.

    These systems did not have any interactive login nodes. So, you had to submit jobs to it using HP 9000 HP-UX hosts as a front end. (Neat mice had two wheels under them instead of a ball or laser.)

    The CM-5 had 896 air-cooled Sparc CPUs inside, partitioned into 4 system images. Each image would run a single job at a time. So the job submitter had to decide whether they needed 64, 256 or 512 CPUs for their MPP program.

    The heat sinks on each CPU were fanless and cylinder shaped, each fin being a disc above the other. There were large fans at the top of the cabinets. There were 7 cabinets in all, each 7 feet tall, arranged in a zigzag pattern in groups of 4 and 3 with a data bridge connecting them at the top.

    I would walk between the cabinets with the data center lights off. At the time it felt very ST:NG. The fan hum and thousands of blinken-lights made it a very cool experience. The lights were connected to the system boards and effectively showed activity. They also showed status codes when the system was in diagnostic mode. Mostly, they were there to look cool.

    The system was owned by a Darpa organization and eventually replaced by a Cray T3E. It stayed busy with a good sized research staff feeding it long running jobs every day. We had 3 full time Thinking Machines engineers on site to keep it running. They performed hardware fixes, software upgrades and provided system support services.

    After the company folded, the CM-5 engineers became employees of a new company that took over the TMC support contracts. It was a sad day when we finally turned the system off. The floor space remained open for many years, but not in memorium. All of our new, larger Cray systems were liquid cooled and therefore took less space.

    And hey, what's wrong with Lisp?

  • CoderDevo (unregistered) in reply to Steve

    Jim is right. It was not an HP front end, it had a Digital (DEC) front end Unix system running DG-UX. Jim's physical description was of the older CM-2 hardware, which was a black cube.

  • (cs) in reply to David C.
    David C.:
    Sounds more like they're running a cult than a computer company.

    I've worked at startups where everybody wants to interview everybody, and where there isn't a very good business plan, but when the primary focus of the interview is to impress the candidate with existing products instead of trying to determine if he has useful skills, then something very screwy is obviously going on.

    Of course, given the fact that the CEO was a paranoid lunatic (and probably in need of psychiatric treatment), this doesn't surprise me much.

    I've always enjoyed reading stories about Thinking Machines (and btw, Fortran would be a pretty bloody stupid choice for a machine targeted at the AI market. And every dialect of Lisp is "specialized." At least it wasn't Prolog).

    The interesting thing about Shirley, to me, is not that she might have been a hopelessly incompetent and under-qualified and paranoid lunatic -- note to any lawyers out there: I said might have been -- but that the DoD went ahead and splurged tens of millions anyway.

    Does this reflect even more poorly on the competition?

    Does it mean that the DoD only funds gibbering inadequates?

    How palpably insane do you have to be in order for the DoD to turn your application down? Would you have to pull a gun on ... no, wait a minute, that would be an instant "Funding approved," wouldn't it?

  • John (unregistered) in reply to David C.

    I read something about this company some time ago -- if I remember correctly it was started by a couple of grad students and they actually ended up employing Richard Feynman.

    Yes, here it is; pretty interesting read. [url]http://www.longnow.org/views/essays/articles/ArtFeynman.php[url

  • CoderDevo (unregistered) in reply to Steve-o
    Steve-o:
    Those same hardware assets later resurfaced and are now sold under the new "eMachines" brand name.

    I don't see how that is possible as either accurate history or even as a bad joke. Thinking Machines had nothing to do with commodoty PCs or even Intel x86 architecture. Brilliant!

  • http://jobs.thinkaloud.in (unregistered)

    Rise and Fall of Thinking Machines is a WTF. I understood.______. :|

Leave a comment on “Thinking Machines”

Log In or post as a guest

Replying to comment #:

Log Out

« Return to Article