Slashdot Mirror

← Back to Stories (view on slashdot.org)

Researchers Are Reconstructing Babbage's Analytical Engine (plan28.org)

Posted by EditorDavid on from the legacy-code dept.

Slashdot reader RockDoctor brings an update on a project to build Babbage's Analytical Engine: Between 1822 and 1847, Charles Babbage worked on a number of designs for general-purpose programmable computing engines, some parts of which were built during his lifetime and after. Since 2011 a group under the name of "Plan-28" have been working towards building a full version of the machine known as the Analytical Engine. (The group's name refers to the series of Babbage's plans which they are working to -- versions 1 to 27 obviously having problems.) This week, they've released some updates on progress on their blog. Significant progress includes working on the machine's "internal microcode" (in today's terminology; remember, this is a machine of brass cogs and punched cards!) [and] archive work to bring the Science Museum's material into a releasable form (the material is already scanned, but the metadata is causing eyestrain). "One of the difficulties in understanding the designs is the need to reverse engineer logical function from mechanical drawings of mechanisms -- this without textual explanation of purpose or intention..." Progress is slow, but real.

Last year marked the bicentennial of Ada Lovelace, who wrote programs for the Analytical Engine and it's predecessor, the Difference Engine, and whose position as "the world's first programmer" is celebrated in the name of the programming language Ada.

76 comments

  1. Re: Question for Slashdot by Anonymous Coward · · Score: 0

    They died you insensitive clod!

  2. Re: Question for Slashdot by Anonymous Coward · · Score: 0

    They both passed away in a car accident.

  3. Oblig by Tablizer · · Score: 2

    And it does run Linux, because it's Turing Complete. Just....very....slowly

    --
    Table-ized A.I.
    1. Re:Oblig by NotInHere · · Score: 0

      No machine buildable by man is turing complete, as our universe has finite mass, and it is impossible to build an infinite machine that can simulate a full turing machine. In this particular case it means that the analytical engine probably has not enough RAM or whatever its name for it is.

    2. Re:Oblig by Tablizer · · Score: 2

      TC is usually interpreted as "approaching" infinite storage, if given lots and lots of time.

      Also note I should have said "could run" instead of "does run". One would need to code up some adapters/emulators first.

      --
      Table-ized A.I.
    3. Re:Oblig by Anonymous Coward · · Score: 0

      Imagine a Beowulf cluster of them! ... now imagine the floor falling out from underneath, due to the weight.

    4. Re:Oblig by Anonymous Coward · · Score: 1

      Technically, any computer that can display prompts to the operator to "insert next disc" and "insert previous disc" has access to "infinite" memory in compliance with the rules for a Turing machine.

    5. Re:Oblig by PPH · · Score: 1

      And it does run Linux

      Pretty good change that, when complete and after the first turn of the crank, it will try to install Windows 10.

      --
      Have gnu, will travel.
    6. Re:Oblig by jeremyp · · Score: 1

      The Earth is finite.

      --
      All I want is a secure system where it's easy to do anything I want. Is that too much to ask ~~ Randall Munroe
    7. Re:Oblig by Tablizer · · Score: 2

      Use turtles, dummy

      --
      Table-ized A.I.
    8. Re:Oblig by RabidReindeer · · Score: 1

      Technically, any computer that can display prompts to the operator to "insert next disc" and "insert previous disc" has access to "infinite" memory in compliance with the rules for a Turing machine.

      Sounds like an install for OS/2.

    9. Re:Oblig by Anonymous Coward · · Score: 0

      Of course! Build a virtual engine with LOGO!

    10. Re: Oblig by Anonymous Coward · · Score: 0

      Oh god that brings back memories!

    11. Re:Oblig by NotInHere · · Score: 1

      Downmodding me doesn't change the truth!

  4. raspberry pi by Anonymous Coward · · Score: 0

    But... why? A raspberry pi is cheaper and much more powerful...

    1. Re:raspberry pi by MightyMartian · · Score: 3, Insightful

      But not nearly as historically significant. Babbage designed the world's first Turing complete computer, and if he could have had the parts machined in his day, the computer revolution would have kicked off a half a century earlier. Just imagine the Internet in WWII, all those Nazi hackers and astroturfers,

      --
      The world's burning. Moped Jesus spotted on I50. Details at 11.
    2. Re:raspberry pi by DarkOx · · Score: 4, Interesting

      I don't know about the prospects for the Internet. Mechanical computers and later electro-mechanical computers are likely as far as anyone would have got. Its really the transistor that gives you computation that is fast enough to run complex packet switch protocols and gives you signaling properties that are conducive to long ranger "high speed" communications.

      I am not sure having computer technology sooner would have accelerated the development of the transistor or digital logic gates. There simply were not enough people running around with a deep enough understanding of physics to work out semiconductors.

      That said we saw how instrumental even "slow" electro-mechanical machines were in applications like code braking, as well as sorting and cataloging. Certainly teletext was an important form of communication and operated on the same principles. I am sure machines would have been applied to artilary targeting, more communications, more ciphering and encryption. It would no doubt have been a very different war, but I don't think a pre-WWII Internet could have been possible.

      --
      Repeal the 17th Amendment TODAY! Also Please Read http://www.gnu.org/philosophy/right-to-read.html
    3. Re: raspberry pi by prefec2 · · Score: 1

      No it would not because the problems where different at that time. Turing took off as there was the need to crack enigma and Zuse needed to caclulate differential equations for planes. Later we needed computers to go to the moon and hit the Russians.

    4. Re:raspberry pi by Anonymous Coward · · Score: 0

      I dunno...imagine web pages being stored physically in thin planes with slightly upraised font akin to printing presses, stacks of them. When a query is made, it is scanned by running a reading head over the text (or physical characters have distinct conduction points), and the characters sent over teletype...might have driven miniaturization, maybe all the way to nanotechnology!
      LOL

    5. Re:raspberry pi by LongearedBat · · Score: 1

      Interesting variant/alternative to steam punk.

    6. Re:raspberry pi by Anonymous Coward · · Score: 0

      "code braking"

      Yes, we must stop code, and brakes seem like a good way to do it.

    7. Re:raspberry pi by Agripa · · Score: 1

      I don't know about the prospects for the Internet. Mechanical computers and later electro-mechanical computers are likely as far as anyone would have got. Its really the transistor that gives you computation that is fast enough to run complex packet switch protocols and gives you signaling properties that are conducive to long ranger "high speed" communications.

      The speed is not so much a problem as the integration and reliability.

      The Analytical Engine could have replaced literal calculators, people doing rote calculations usually for military applications up to World War 2. It would have been too large and unreliable to replace mechanical analog computers used in things like fire control.

    8. Re:raspberry pi by RockDoctor · · Score: 1

      Closer to a full century than a half century. Babbage was trying to get this working in the 1840s, not the 1890s (though the quality of mass-production wasn't up to the task then).

      --
      Birds are not dinosaur descendants;birds are dinosaurs, for all useful meanings of "birds", "are" and "dinosaurs"
    9. Re:raspberry pi by RockDoctor · · Score: 1
      They had scanners for digitising images and transmitting them by mechanised Morse code in ... the 1870s or 1880s, IIRC.

      Underestimating people has a very long history of getting things wrong. We might not be able to "re-run the experiment", but we do have a good number of examples of the ease of underestimating strangers.

      --
      Birds are not dinosaur descendants;birds are dinosaurs, for all useful meanings of "birds", "are" and "dinosaurs"
    10. Re:raspberry pi by Tablizer · · Score: 1

      we saw how instrumental even "slow" electro-mechanical machines were [used for specific applications].

      The "general purpose" thing didn't offer enough advantages to overcome hardware issues. Hardware was still such a strong factor that purpose-built machines were either more economical than a general purpose computer, or the difference was too small to make R&D on general purpose computers worth it.

      Purpose-built machines are almost always more efficient at their intended task than general purpose computers. The advantages of general-purpose were slight compared to hardware issues.

      As a thought experiment, suppose many at the time did understand the advantages of general-purpose computing. Would they have considered the advantages worthy enough to start building hundreds of mechanical GP computers?

      I doubt so. Building something like the Internet with the hardware of the time would likely be a practical failure. It would be like Comcast snags x 1000.

      Babbage, Grace Hopper (high-level programming languages), and the Xerox "labbers" were dreamers. Their visions were ahead of the hardware of the time.

      Actually, the reason high-level programming language research initially got funding is that the military contractors couldn't transfer programs across brands and models of computers. The human-readability issue was not the main motivator. Grace initially mostly focused on the readability issue, not vendor-independence.

      --
      Table-ized A.I.
  5. The world's first programmer... by K.+S.+Kyosuke · · Score: 1

    ...surely didn't appear until computers appeared.

    --
    Ezekiel 23:20
    1. Re:The world's first programmer... by AthanasiusKircher · · Score: 4, Insightful

      So a man discovers computers, a women figured out how to do a diddly on it, and we celebrate her as the first programmer without calling him the first programmable machine inventor.

      Typical Feminism, and these days, Typical Slashdot.

      Umm... huh? From TFS:

      Between 1822 and 1847, Charles Babbage worked on a number of designs for general-purpose programmable computing engines, some parts of which were built during his lifetime and after.

      That sounds like we start TFS by recognizing Babbage's contributions for designing these "programmable computing engines." And, just in case that's not enough for you, there's a link to the Wikipedia article on Babbage right in that sentence in TFS which acknowledges in its opening paragraph that Babbage: "...is best remembered for originating the concept of a programmable computer."

      So, I'm really not sure how you get that Slashdot is somehow recognizing Lovelace without acknowledging Babbage's contributions.

      Can we get past a week here without our allotted dosage of this bullshit for fucking once?

      Ada Lovelace did some interesting stuff. She was an interesting person. And she's mentioned in the summary because Slashdot had a story on her bicentennial last year. Babbage's bicentennial was in 1991... but should we celebrate him this year too even though it's just his bicenquasquigenary (a term 99.9% of people haven't even heard of, because we generally don't celebrate 225 year anniversaries)? Would that make you happier?

    2. Re:The world's first programmer... by Anonymous Coward · · Score: 1

      Oh, you're so full of shit. Babbage has been called the "father" of the computer for ages.

    3. Re:The world's first programmer... by MightyMartian · · Score: 4, Insightful

      Lovelace was a woman, and to the Libertarian Neanderthals that frequent Slashdot, that means mentioning her is a violation of some sort of sense of maleness. There are just a lot of very angry men out there, and any time anything is attributed to a woman, they literally start foaming at the mouth, shouting "SJW". They're a rather pathetic lot who, I suspect, don't spend very much time around women, or possibly other humans at all. They can be safely ignored however, they are a shrinking demographic.

      --
      The world's burning. Moped Jesus spotted on I50. Details at 11.
    4. Re:The world's first programmer... by Plus1Entropy · · Score: 4, Funny

      they are a shrinking demographic

      Probably a direct effect of not spending much time around women.

      --
      Only crack the nuts that crack. You don't put the ones that don't crack in the sack.
    5. Re:The world's first programmer... by Darinbob · · Score: 1

      Babbage is widely acknowledged, you just have some sort of blinders on if you missed that.

      Ada Lovelace is known for two things. Being the daughter of Lord Byron, and popularizing the work of Charles Babbage (making his work more readable to English speakers). She'd not be remembered if it weren't for Babbage, and Babbage would not be remembered if it were not for Lovelace. Don't know why you feel the need to pick sides between the two.

    6. Re:The world's first programmer... by Anonymous Coward · · Score: 0

      I should comment that I have seen feminists claim that the achievements of women are downplayed compared to those of men, and I have seen anti-feminists claim that the achievements of men are downplayed compared to those of women - but the only ones I have seen *advocate* for discrimination are feminists.

    7. Re:The world's first programmer... by Anonymous Coward · · Score: 0

      lol. I love masterbating myself when sexist men get angry.

    8. Re:The world's first programmer... by RockDoctor · · Score: 1

      Typical Feminism, and these days, Typical Slashdot.

      Oi, dumbfuck!

      Submitter here. I spent the thick end of an hour composing the original summary (somewhat edited now), checking, selecting and linking to relevant sources. I even considered dropping a mail to the manager of Plan-28 to send him a heads-up in the event of his servers getting Slashdottted (but decided against it as I didn't know if the submission would get to the front page).

      And then you don't have the plain common decency to actually read the fucking thing.

      You are a fuckwit. No wonder you haven't got the metaphorical balls to put your name to your idiocy.

      --
      Birds are not dinosaur descendants;birds are dinosaurs, for all useful meanings of "birds", "are" and "dinosaurs"
    9. Re:The world's first programmer... by RockDoctor · · Score: 1

      You could make a very good case that turning a designer's drawing of a pattern for woven fabric into a set of Jacquard Loom cards was an act of programming. The Jacquard loom had no computational capability, and in particular had no capability for a logical branch in it's actions.

      --
      Birds are not dinosaur descendants;birds are dinosaurs, for all useful meanings of "birds", "are" and "dinosaurs"
  6. Re: Question for Slashdot by UnknownSoldier · · Score: 1

    [[citation]] ?

  7. Ada's cousin by Anonymous Coward · · Score: 0

    The Byron family was prolific, including Mary Shelley (Frankenstein), Lord Byron, Lady Ada, and others. These are my predecessors, so including being a cousin of Frankenstein, I guess I am also a cousin of the world's first computer! And then, I am a software and electrical engineer... :-)

  8. Re: Question for Slashdot by Anonymous Coward · · Score: 5, Informative

    Original AC here. You're being trolled. That said, after looking into the claims in this thread, it looks like Timothy got canned. And I do have a source for that: https://mobile.twitter.com/timothylord/status/715960545271132160. That sucks.

  9. Re: Question for Slashdot by Anonymous Coward · · Score: 0

    (Who Gives a Fuck, 2016)

  10. Difference Engine by cyberpunkrocker · · Score: 5, Funny

    Charles Babbage's "Difference Engine" created much controversy in its time, but his equally ingenious "Indifference Engine" was received with... "meh".

    1. Re:Difference Engine by dpilot · · Score: 3, Interesting

      My wife and I recently went to England with friends, and one of our stops was at the London Science Museum to see the Babbage Difference Engine #2. (with built-in printer) I wasn't aware that there were enough drawings generated to even attempt the Analytical Engine.

      --
      The living have better things to do than to continue hating the dead.
    2. Re:Difference Engine by K.+S.+Kyosuke · · Score: 2

      Babbage probably made the crucial mistake of not using relays. I'm mildly convinced that a relay version would have been feasible in his time.

      --
      Ezekiel 23:20
    3. Re:Difference Engine by Anonymous Coward · · Score: 1

      Actually, there are far more than mere drawings. Babbage was a mathematician, and developed an entire symbolic language devoted to describing mechanical devices, and it was that language about which he was most proud; the Analytical Engine was, in a way, merely the means by which to develop that language in practice.

      Unfortunately, his language never became widely adopted.

    4. Re:Difference Engine by Anonymous Coward · · Score: 0

      Relays didn't work 100 years later for Mark 1 or for zuse.

    5. Re:Difference Engine by K.+S.+Kyosuke · · Score: 1

      Relays actually worked pretty well. Anyway, they definitely worked better than early 19th century cogs. You might of course wish for some ingenious, part-count limiting design - some of the early computer designs were quite awful, which admittedly was largely due to their designers' inexperience, but I still believe that Babbage was more likely to come up with such a design than his machinists with superior methods for precise mechanical engineering in the 1830-1840s, the lack of which killed many of Babbage's valiant efforts. Not to mention the inherent flexibility of wiring (pun not intended) instead of axles should you decide to change your design at some point.

      The real issue I see is rewritable memory technology. A stack-based MISC-like design could minimize the part count for the CPU unit (not just for reasons of cost but especially for reliability), but there would still be storage requirements. Either densifying the code through microcode (a grid of wires, really - quite feasible) or using Harvard architecture for programming and putting most of the code in ROM (again quite feasible due to the Forth-like nature of stack machines, which makes binary programming by means of a grid of wires by hand not an act of insanity) would minimize program storage costs, but there's still the issue of where to put the data. You might wish for several hundred bytes even for the most basic calculations such as printing better function tables.

      --
      Ezekiel 23:20
    6. Re:Difference Engine by Anonymous Coward · · Score: 0

      rip off of leibniz' work

    7. Re:Difference Engine by Agripa · · Score: 1

      The real issue I see is rewritable memory technology.

      Core memory would be out without some way to build sense amplifiers. Latching relays could be used for RAM. Paper tape could be used for mass storage.

    8. Re:Difference Engine by RockDoctor · · Score: 1
      ... and that is why I wrote up the message I received about progress on the project : to inform people who didn't know.

      An hour of my life, NOT wasted.

      --
      Birds are not dinosaur descendants;birds are dinosaurs, for all useful meanings of "birds", "are" and "dinosaurs"
    9. Re:Difference Engine by RockDoctor · · Score: 1

      I've disembowelled 1940s-era mechanical calculating machines that did storage with a 10x10 array of spring-loaded pins. Given the necessary precision of mass-production (which was Babbage's main problem), then the availability of appropriate memory becomes a question of money. And strong-enough floor beams.

      --
      Birds are not dinosaur descendants;birds are dinosaurs, for all useful meanings of "birds", "are" and "dinosaurs"
    10. Re:Difference Engine by K.+S.+Kyosuke · · Score: 1

      Core memory would be out without some way to build sense amplifiers. Latching relays could be used for RAM.

      These were exactly my thoughts. Had future physics and chemistry knowledge been available at Babbage's time, I would have also considered the use of selenium and neon lamps, possibly to create flip-flops and other active elements. (Do you also find alternative digital logic building blocks as fascinating as I do?)

      Ah, I take it you're familiar with the Jacquard loom. :)

      --
      Ezekiel 23:20
    11. Re:Difference Engine by Agripa · · Score: 1

      Gas discharge lamps can be used to make memory and many logic elements (and they were for a short time) however I excluded them because they essentially use vacuum tube technology. If you can produce them, then you can produce vacuum tubes.

      https://en.wikipedia.org/wiki/...
      https://en.wikipedia.org/wiki/...

      Another alternative logic technology they could have used involves fluidics. I have never seen fluidic memory but if you can make logic gates, then you can make memory.

      The Jacquard loom was what I was thinking of for paper memory.

      I wonder if it is possible to implement a core memory sense amplifier using magnetic amplification. Magnetic amplification was still World War 2 and later technology.

  11. Re: Question for Slashdot by MightyMartian · · Score: 2

    I guess this all explains why the nice little checkbox not to show ads doesn't survive a page refresh. Meet the new boss, samer than the old boss...

    --
    The world's burning. Moped Jesus spotted on I50. Details at 11.
  12. Re: Question for Slashdot by Anonymous Coward · · Score: 0

    Thank the fuck Christ. Best news ever

  13. Re: Question for Slashdot by Anonymous Coward · · Score: 0

    He died AFTER he got canned you insensitive clod!!!

  14. Re: Question for Slashdot by 110010001000 · · Score: 1

    Wow did you read his Twitter feed? Incomprehensible. This guy was an editor?

  15. Re: Question for Slashdot by Anonymous Coward · · Score: 0

    He seems to have turned into an anti-IRS extremist since he was let go, or maybe he's decided not to hide it any longer. I wonder why? It's not like he's got loads of income to tax.

  16. Re: Question for Slashdot by Darinbob · · Score: 1

    He was always interested in stories with that bit of conspiracy to them, so it's no surprise.

  17. No, they're "constructing" it by russotto · · Score: 1

    It was never built in the first place, so it can't be "reconstructed".

    1. Re:No, they're "constructing" it by RockDoctor · · Score: 1

      Parts of it were built.

      --
      Birds are not dinosaur descendants;birds are dinosaurs, for all useful meanings of "birds", "are" and "dinosaurs"
  18. Computers are LBA-complete by tepples · · Score: 1

    The closest traditional mathematical model to a physical computer is a linear bounded automaton (LBA), which is a Turing machine unable to move the head outside an area proportional to input size. It recognizes context-sensitive languages.

  19. Re: Question for Slashdot by FatdogHaiku · · Score: 1

    [[citation]] ?

    {Fiat}

    --
    You have the right to remain sentient. If you give up the right to remain sentient, you will be elected to public office
  20. Just Imagine by Anonymous Coward · · Score: 0

    Can you imagine though, if the Analytical Engine had been built? Would the British government have understood it enough to make effective use of it? How many would get built? Would other countries (Germany perhaps) recognize a competitive threat and build their own? Or steal plans for the original?

    One imagines the earliest engineers, programmers, even Babbage and Ada Lovelace, realizing they need to increase the cycle time of the device. They could have connected early steam engines to the input crank to achieve that. Then the AE breaks because they stress it too much, so the AE undergoes a round of upgrades to make it mechanically more durable.

    Eventually they get some decent programs going and grow to depend upon the device. Version 2 follows but they have huge problems upgrading and all their programs are incompatible. Everything we experience today, just 100 years prior.

    I imagine it would be much like the early days of the mainframe. They would be inventing and using tech as they went. Finally electronics are invented and a whole cadre of early programmers resent the new technology and point out all the things mechanical computers can do that electronic ones cannot do!

    1. Re:Just Imagine by Anonymous Coward · · Score: 0

      "One imagines the earliest engineers, programmers, even Babbage and Ada Lovelace, realizing they need to increase the cycle time of the device. "

      One imagines they needed to decrease it.

  21. Re:Question for Slashdot by Hognoxious · · Score: 1

    They were rebranded as manishs and EditorDavid, not necessarily in that order.

    --
    Confucius say, "Find worm in apple - bad. Find half a worm - worse."
  22. Re: Question for Slashdot by freeze128 · · Score: 1

    The date on the tweet is April 1st, so I don't know if I can believe it.

  23. it's means it is by Anonymous Coward · · Score: 0

    Hey idiots, think that concept will ever sink into your rock-like skulls?

  24. Re: Question for Slashdot by Anonymous Coward · · Score: 0

    Tweeting is like texting, it's kind of a different language. My mother corrected my grammar and spelling immediately most of my life. I cringe when getting texts from her now. What I'm trying to say is - don't judge someone's intelligence by their tweets.

  25. Deliberate Errors by Fnord666 · · Score: 1

    One of the more interesting things I remember from a video about the construction of the difference engine was the introduction of deliberate errors. Apparently the engineering drawings included deliberate errors in key pieces so that if they were fabricated as drawn, it would jam the machine up badly. This was in case someone stole or copied the plans but plays hell with constructing one today.

    --
    'The tyrant will always find pretext for his tyranny.' - Aesop's Fables
  26. Re:She was a nobody, a myth. by RockDoctor · · Score: 1
    Thanks for linking to your sources, AC. No wonder you're so proud of your work as to put your name to it.

    So, here's are some relevant links to material published about Ada Lovelace during her bicentenary.

    she still doesnâ(TM)t quite fit the mould of a traditional science heroine.
    Intelligent she might have been, but she was also manipulative and aggressive, a drug addict, a gambler and an adulteress.
    Alongside the character flaws,

    I don't see those as character flaws. Makes her more interesting ; maybe challenging. "Flaws?" Only if she let them be flaws.

    There was an Oxford symposium in 2015.

    --
    Birds are not dinosaur descendants;birds are dinosaurs, for all useful meanings of "birds", "are" and "dinosaurs"