I think a valid part of the question of who invented something is "who built the first working device" - describing something in theory and building working device are not the same thing.
AG Bell wasn't the first one to conceptually invent the telephone, he was among the first (along with Elisha Gray) in making practical working telephone and later a practical working telephone system.
So, who invented the Satellite then? What about the steam engine? The helicopter?
Sometimes the inventors are so far ahead of their time that the materials science first has to catch up (in some cases only a few millenia) before they can realize their devices. Effectively it is then the first person after whoever did the materials science part to create the device that gets to claim the invention.
So we get Sikorski, and not Da Vinci.
We get Arthur C. Clarke who claims the 'communications satellite' even though the moon was there all along and the Sputnik was the first working very crude device (it was one way only, it said 'you lost the space race' in a single bit of message).
We get Newcomen, Jerónimo de Ayanz y Beaumont (I had to look that up, I can never remember the man's full name), and Hero of Alexandria competing for the steam engine title, with all of them holding some part of the credit.
Pointing at an inventor is hard, and 'who built the first working device' is one way of doing this but it assumes a singular effort whereas most things are team efforts and misses the bit that the idea itself can be an instrumental step in getting your 'true' inventor to make their claim, standing on the shoulders of the giants before them. In isolation, we all probably would invent the hammer in our lifetimes, if that.
Savery is referenced in the "Watt steam engine" Wikipedia article.
Also worth noting that even though Watt improved on Newcomen's design to improve efficiency, Watt's design still sucked: it took Wilkinson's boring technology to actually get them to work well.
Ah yes, and a better claim than some (there are quite a few more).
I always wondered what the effect of these in absolute time is. Consider: if someone had come up with a viable steam engine in the year 1000 or so, how would that have change the course of history?
I remember reading an analysis of that a while back but I can't remember where I saw it. The upshot was that the industrial revolution turned on more than just innovation, it also required just the right combination of natural resources (coal) and economic conditionsL deforestation in England driving the use of coal for heat, which drove mining, which drove the need for pumps, which made Savery's engine economically viable. Before that steam power would not have found an application.
Also, the piston and cylinder (which came later) were derived from canon-making technology, so that had to come first too.
I remembered the same article too, I'm sure I could have found it again knowing it was listed on HN and/or recalling the wording. Good article too, going in depth on steam engines - the first ones were so inefficient they had to be really close to a really cheap source of fuel and be useful.
> t also required just the right combination of natural resources (coal) and economic conditionsL deforestation
This argument is repeated often, but I don't think it's really true. Both Savery and Newcomen's engines were initially aimed at evacuating flooded metal mines and not coal mines.
"A few Savery pumps were tried in mines, an unsuccessful attempt being made to use one to clear water from a pool called Broad Waters in Wednesbury (then in Staffordshire) and nearby coal mines."
That is correct, but the article explicitly addresses this point and argues that the evidence points to Lilienfeld producing a working transistor.
"Later, some people claimed that Lilienfeld did not implement his ideas since "high-purity materials needed to make such devices work were decades away from being ready,"[CHLI] but the 1991 thesis by Bret Crawford offered evidence that "these claims are incorrect."[CRA91] Lilienfeld was an accomplished experimenter, and in 1995, Joel Ross[ROS95] "replicated the prescriptions of the same Lilienfeld patent. He was able to produce devices that remained stable for months."[ARN98] Also, in 1981, semiconductor physicist H. E. Stockman confirmed that "Lilienfeld demonstrated his remarkable tubeless radio receiver on many occasions".[EMM13]"
For many things (computers, rocketry, aerospace, etc.) and different reasons, Germany in the years around the second world war, was a pretty bad place to get international credit for your accomplishments.
"It was able to squeak, but not to speak. Experts and professors wrestled with it in vain. It refused to transmit one intelligible sentence." [0]
"A translation of Legat's article on Reis' invention was obtained by Thomas Edison prior to his filing his patent application on a telephone in 1877. In correspondence of 1885, Edison credits Reis as having invented "the first telephone", with the limitation that it was "only musical not articulating"." [1]
Fascinating stuff nonetheless, these inventors and their ideas... See also previous experimenters [2]
Yeah - though Bell's first apparatus wasn't much better - the invention of the carbon microphone is what really what set the telephone on to being a practical device. The rest of it was trying to build a network to connect people - and that was really hard (and capital intensive).
What blows my mind that we absolutely take for granted today is insulated wires. The technology and supply chain to mine or to find into metal and also to farm cotton and wool and formed that into protective tubing before the advent of plastic insulation. The amount of technology that goes into making a "simple" USB-cable beggars belief if you stop to think about it. Even a simple #2 wooden pencil with an eraser on top is beyond the knowledge of one person to produce, nevermind a USB-c cable!
In Victorian London, electricity was distributed around the home using bare wires with an air gap. It was 32v though.
Later it was superseded with lead wrapped in paper, until the Knob and Tube system. This comprised of single-insulated copper conductors installed within walls and ceilings, this wiring was encased in porcelain insulating tubes with cloth-lined sleeves.
One knob for Live and one knob for Neutral. The wires were held in place by porcelain knobs nailed to the house frame. Where wires passed through wood framing, they were threaded through porcelain tubes to prevent them from contacting the wood.
My house once had knob and tube wiring. Over the past 15 years I have replaced most of what I assume is the second-generation cloth-covered wiring (which dates back to the 40s, 50s, and 60s). Every once in a while, I come across the insulators from an older electrical system, but most of the wire that went with those insulators was pulled out long ago. The only remains are short bits of wire that were wrapped around the insulators.
I have a cable here that is interesting. When I first saw it I thought how strange, a cable that is multi-stranded bare wire for this application (connecting a camera). Then I looked at it under a microscope and realized not only is it insulated wire, it is shielded wire. Mindblowing. I pity the people that have to handle that stuff.
Small addendum: at least in Germany, early telephone wires (up until the 1950s?) were wrapped in paper, drenched in oil. The bundles were then enclosed in a lead-copper alloy to protect them from moisture.
Reminds me of the (not entirely accurate) story of someone trying to make a cheeseburger from scratch and realizing it required most of modern civilization.
To some degree, this is a consequence of the nature of the field you're working in:
* if the physics is so completely understood that you can confidently predict something will work from your sofa, and give an error-free recipe to build it, you indeed can invent from theory... but how deep can this invention be if the problems of the field are completely solved?
* if you are working in a field at the edge of human understanding, you cannot have the confidence in your ideas without having tested them experimentally; a theoretician makes at most a minor contribution to the actual inventions being realized, because he's producing - most likely somewhat wrong - hypotheses.
This latter kind of "theoretical" inventions are heavily subject to survivorship bias. Fifteen competent theoreticians make different predictions - all according to best, though incomplete, model of the world; a successful experiment validates exactly one of them, and we end up exalting the lucky winner as the "inventor".
In practice, any unexplored corner of the field will contain surprises; these will require extra theoretical development to cover.
Usually things like imperfect understanding of materials get in the way. Pretty much the reason you need both theory and experiment to make progress in every single area of matter-based technology (i.e. not software).
I had that thought too, describing that something might be physically possible isn't really inventing it, you have to build (and arguably sell) the device too. Re-organizing someone else's equations and saying it's technically possible is maybe enough to publish a paper but certainly doesn't rise to the standard of inventing in my mind
There's something to be said that mass production is another distinct stage of invention. Karl Benz may have invented the first internal combustion engine car, and plenty more built cars by hand for the rich, but Henry Ford made cars anyone could have for cheap.
I agree with that. It is a bit like the idea that having an idea for a startup is less important than the ability to get it to a level where investors might be interested.
A while ago there was an artical posted here about all the world changing inventions that came out of Bell Labs. It was easy to show for most of them that they weren't the actual inventor, and in many cases not even the first producer. They were the first to make it practical for mass production, however.
>Where are the physical limits? According to Bremermann (1982), a computer of 1 kg of mass and 1 liter of volume can execute at most 1051 operations per second on at most 1032 bits. The trend above will hit the Bremermann limit roughly 25 decades after Z3, circa 2200. However, since there are only 2 x 1030 kg of mass in the solar system, the trend is bound to break within a few centuries, since the speed of light will greatly limit the acquisition of additional mass
They shift from talking about the transistor density to somehow considering a supermassive construct. Reminds me of LLM mashups.
It's a natural extension of the ideas being discussed - the limit in computation per gram of mass has energetic bounds, as well, with configurations nearing the upper limit that start looking more like nuclear explosions than anything we'd regard as structured computation. The extremes are amazing to consider - things that look and act like stars, but are fantastically precise Turing machines, and so on.
It's a theme that sci-fi authors have explored deeply. Accelerando is a particularly fun and worthwhile read if you haven't already!
> The naive extrapolation of this exponential trend predicts that the 21st century will see cheap computers with a thousand times the raw computational power of all human brains combined
i.e. putting an upper bound on the exponential with solar system mass
Is there any evidence that Shockley was aware of Lilienfeld’s FET? I mean, given Shockley’s general disposition, I wouldn’t be surprised if he’d stolen things, but we have his early prototypes and such to back up his claim of invention. The next step is to prove a through line to the modern era.
The Sumerians knew geometry, trigonometry, and other advanced mathematics. That knowledge perished, and then the Greeks rediscovered things. The Greeks then get credit because their work is what paved the way for all of us today. The same would apply to Shockley. If Lilienfeld’s work didn’t make it to the USA in time to have had an effect on development in the valley, then it doesn’t really matter for the US industry.
The entire debate is tiring. It would be better if these reviews would put the actual device physics of the different concepts into context.
Is there any report of a reproduction of the device proposed by Lilienfeld in his patents? If he managed to make functional devices back then, it should be possible today? (Note: Cu2S is not a very well controllable semiconductor...)
From the second document:
"The primary obstacle for $Cu_2S$ TFTs is degeneracy. Spontaneous copper vacancies form with negligible energy cost in the sulfur lattice. As a result, stoichiometric $Cu_2S$ is thermodynamically unstable in air, rapidly oxidizing or losing copper to form substoichiometric phases ($Cu_{2-x}S$) with hole concentrations exceeding $10^{20}-10^{21} \text{ cm}^{-3}$."
This explains why there are zero reproductions of Lilienfelds devices. It should be noted that Lilienfeld is one of the inventors of electrolytic capacitors and did therefore know where well how to create extremely thin insulating layers as needed for TFTs. It is not impossible to assume that he could have used other semiconductors (e.g. CdS) with his concept. However, the patents seems to specifically mention Cu2S, which does not yield functional TFTs.
Though I'm unsure of the author's motivations, there's certainly a tendency with Shockley similar to what one sees with Hitler, where people will occasionally question or cast doubt on Hitler's WWI service record--which was by all accounts exemplary. (I.e., Bad Guy can't also possibly have done Good Thing.)
The term "transistor" is much too vague, so the question "who invented the transistor" is ill posed.
There are different kinds of transistors, which use different principles for controlling electrical currents, and which have been invented by different people at different times.
Lilienfeld has invented 2 kinds of transistors: MESFETs (metal-semiconductor field-effect transistors) and depletion-mode MOSFETs (metal-insulator-semiconductor field-effect transistors) (US patents 1,745,175 and 1,900,018). Of these 2 kinds of transistors, the second is rarely used today, while the first is used mostly in special applications for discrete power devices.
The parent article is very wrong in claiming that the transistors invented by Lilienfeld are those that are most frequently used today, which are the enhancement-mode MOSFETs, which have a very different structure and for which much less of the semiconductor materials are suitable than for the simpler transistors invented by Lilienfeld.
The enhancement-mode MOSFET has been invented in 1960 by Martin M. Atalla and Dawon Kahng from Bell Laboratories (US patents 3,206,670 and 3,102,230).
Bardeen and Brattain have discovered the point-contact transistor, which had been used only for a few years before becoming completely obsolete, but which was the first kind of transistor offered as a commercial product.
William Shockley has invented 2 kinds of transistors, which have been very important and which remain the best in some special applications, the bipolar junction transistor (BJT) and the junction field-effect transistor (JFET) (US patents 2,569,347 and 2,744,970). William Shockley has also the huge merit of developing a theory of the physics of semiconductor materials which allowed everybody to design transistors and many other kinds of semiconductor devices.
Therefore the 3 Nobel winners have invented together 3 kinds of transistors and they are rightly called inventors of some of the kinds of transistors, i.e. point-contact, BJT and JFET.
Lilienfeld was the first who conceived some possible structures for a triode that uses a semiconductor instead of vacuum or gas. This was a great advance, but this still resulted logically from the prior knowledge that one can make diodes using either vacuum or gas or a semiconductor, and one can make triodes using either vacuum or gas. Thus he attempted to fill the missing combination.
The Bell Laboratories are truly guilty of trying to hide the fact that their post-war research on making a semiconductor triode had indeed started with the purpose of making alternatives to the devices invented by Lilienfeld, of which they were well aware. However, they eventually invented 4 different kinds of transistors, all of which had structures and principles of operation very different from the 2 Lilienfeld transistors, even if all 6 kinds can be considered as variants of semiconductor triodes.
The first 3 kinds of transistors that have been invented, the 2 Lilienfeld transistors and the point-contact transistor of Bardeen and Brattain, can be made using pieces of a homogeneous semiconductor material. This is why they have been discovered first.
The next 3 kinds of transistors, invented by Shockley and by Atalla with Kahng, contain P-N junctions, so they were invented only after William Shockley had developed the theory of the P-N junctions.
When you come in the UK as a continental european (french for that matter), you notice you're inundated with tons of trivia of how great british inventors are, and how many things they invented/discovered. They even make TV shows about it.
Nothing like this exists in France and people have no clue about who invented stuff, nor do they care.
Now with internet, where truth is supposedly coming from english world, there are plenty of opportunities to rewrite history 1984 style.
It's really damning now with AI auto translating every english to your local language.
I wish Jürgen Schmidhuber would switch back to actually doing AI research instead of having become completely obsessed with "who invented what" because he feels like he has somehow been academically "robbed" at some point in his career.
He's now officially become a full-blown pariah in the AI world, most relevant people in the space running away at the first sight of his goatee at conferences, knowing exactly the kind of complete and utter crank he's become.
In fact I am generally ignorant on the topic of who invented the transistor, nor do I in general particularly care about who invented what.
The quest for academic fame is something I've always utterly failed to understand.
And, if it the author was anyone but JS I'd not have said anything.
What honks me off about this guy though is to see a someone who did in fact do early impactful work on recurrent neural networks believe that:
a) that automatically gives him some sort of special status wrt the rest of humanity
b) because he didn't get the recognition he believes he is due, has completely stopped doing anything useful in the field, turning instead into an absolute crank that every one in the AI field makes fun of, and with a holy mission to rewrite history to assign credit where credit is due everywhere he believes there was an injustice.
c) every time I see someone with an exceptionally well-working brain waste their time because of ego or sheer stubbornness on shite like this instead of using it to do more interesting work, it makes me very sad.
Schmidhuber is a textbook example of this, and the other perfect example of this is Chomsky, a very smart man, who basically - because of his oversized ego and profound stubbornness - ended up wasting his entire life energy working on a linguistic dead-end AND a political philosophy dead-end.
I have a real hard time understanding how the brain of that kind of folks operates, being so bright on certain axes and totally and utterly dumb on others, especially the total lack of self-awareness.
Wow that is super vitriolic, more if from a colleague. Sure the hehe debate is about if he is to be considered more or less accomplished, but going from non-success to pariah, what is the need?
> but going from non-success to pariah, what is the need?
Not sure "need" is the appropriate word here.
Grab anyone in who has worked on AI in the last 30 years, and pronounce the word "Schmidhuber" and watch the face of you interlocutor: you'll either get an eyroll or a smirk, but rarely a lively discussion on what he's "invented".
I'll just leave this here https://scholar.google.com/citations?user=gLnCTgIAAAAJ&hl=en so maybe you realize that's a bit of a tall claim from a random about one of the top researchers in AI, no matter what their opinions are. Perhaps you should look up what a "crank" actually is before labeling researchers, just because they don't match your religion.
Nope. The contrary as a matter of fact. But the facts are:
1) nothing worth talking about since the LSTMs
2) most of the reasons why he's been visible in the last 20 years is because of shit like this (JS harassing Ian Goodfellow about attribution in the middle of a technical presentation. Watch the face of Ian, pretty symptomatic of when AI folks have to interact with JS these days).
He is usually insufferable (and misleading) when he talks about AI innovation priorities. This article is different, though. It really does seem like Lilienfeld invented the transistor first and should be given credit for that. It also really does look like the official inventors knew that and were somewhat dishonest.
Funny, btw, that nobody here has mentioned that Lilienfeld also invented the electrolytic capacitor.
How would you distinguish the article from an honest write up about transistors? That is, you know about his crusade in ML, but if you didn't, how would you decide that this article is written in bad faith or not?
I agree that context matters, and I had the same thought as you. But does that mean that anything he writes on the topic of "who was first" is inherently tainted?
I personally detest the way we sanctify some sole individuals while forgetting the bulk of the community. I don't care who published the first patent for the transistor. He or She certainly cannot be credited for all the work that has been put into it so that I can today use a hand held device to post this comment.
I see where you're coming from, but while that's the case with most stuff ("normal science") , it often isn't the case for truly revolutionary stuff. Many breakthroughs happen not because of the bulk of the community, but against it, often at the highest cost for the individuals.
Surprisingly also not true. Yes, people go against the grain and it is required to actually make paradigm shifts but they're never alone nor did they build from scratch. It may be few against many but it is almost never one against all. That one only prevails due to support from others. Those names don't shine but it doesn't mean they weren't critical to the advancement of a field
Strong claims - maybe good time to do some homework instead of arguing without evidence?
Galileo was sentenced to house arrest for heresy.
Boltzmann died by suicide after lack of acceptance by the scientific community.
It's a very long list and something that's been studied, actually.
You do realize Galileo and Kepler were around at the same time, right? Galileo is the best example of what I claimed.
You criticize me for "not doing my homework" yet you haven't heard of Castelli[0]? He's even in the fucking Wiki article on Galileo lol[1]
> Boltzmann died by suicide after lack of acceptance by the scientific community.
Your knowledge of Boltzmann seems to be as deep as your knowledge of Galileo.
> It's a very long list and something that's been studied, actually.
You're absolutely right. You should read some of it. Or if you don't like reading I do highly recommend An Opinionated History of Mathematics[2]. Blåsjö even has a whole season dedicated to Galileo.
Sorry, but having two people instead of one pushing for similar/related ideas is not supporting the main claim ("we sanctify some sole individuals while forgetting the bulk of the community"). Nor is stealing ideas and having the wrong name on results. It's rather common to have multiple names in theorems because of this. That's still beside the point, when claiming it's the "community" or hive mind.
It's not that I don't see where you're coming from. What's been gesticulating at is called "Stigler's law of eponymy" (https://en.wikipedia.org/wiki/Stigler%27s_law_of_eponymy). Pythagorean theorem was known 1000 years before Pythagoras and a ton of results are wrongly attributed to. But that doesn't make revolutions a collective effort.
Was Galois', Gödel's or Grothendieck's work obviously the common sense of the community?
Or wasn't it the "community" that tried to "solve" the parallel postulate for 2000 years before Bolyai did?
Or maybe everyone in the Copenhagen crowd cheered Bohm, Everett and Bell because it was all obvious in the community?
Thanks for the link, but when I said "do homework" I didn't mean podcasts :), but rather either general or concrete work in the history and philosophy of science, like Thomas Kuhn's "Structure of scientific revolutions".
It seems like you agree with me yet simply do not easy to. I can't do anything with that. You seek evidence that is incredibly difficult to present. History is recorded differently now than in the past yet things are the same. You're ignoring the biases of how it's recorded. Any example I give you'll say it's not enough and point out another. It's a fruitless game
> but when I said "do homework" I didn't mean podcasts
Maybe you should judge the information on its merit rather than its medium
Possibly. I don't think either extreme opinion is correct— i.e., both can be partially correct, hence we may agree on that. Merit is often arbitrarily attributed (and sometimes, perhaps often enough, wrongly). I don't think that's enough to say that that's "always" or "most often" the case. That was my main point. You argued that no one does something strictly alone or from scratch. I think the "from scratch" part is irrelevant and think that a single person, a very small group, or a few independently (even if at different points in time) still warrants the same individual recognition.
Neither really fits what Jürgen Schmidhuber argues for. He clearly argues against intentional non-attribution and/or ignorance. Both are evidently the case deeply in current AI research (probably also outside, but this is what OP is about). At the same time, claiming that his points are worthless because he's just pointing to other individuals that suffer from the same misattribution problem is generally a wrong take. Work concentration and results are simply non-linear and generally clustered around individuals and groups. Failing to attribute, either from ignorance or malice, is equally wrong.
“It takes a thousand men to invent a telegraph, or a steam engine, or a phonograph, or a photograph, or a telephone or any other important thing—and the last man gets the credit and we forget the others. He added his little mite — that is all he did. These object lessons should teach us that ninety-nine parts of all things that proceed from the intellect are plagiarisms, pure and simple; and the lesson ought to make us modest. But nothing can do that.”
This is understandable, but maybe you cannot offer an alternative to just give credit nobody. Schmidhuber is playing the game by its rules being self-consciously contrarian enjoying himself, while offering an interesting historic pinch of retrospective justice.
@book{Kuhn1962,
author = {Kuhn, Thomas S.},
title = {The Structure of Scientific Revolutions},
publisher = {University of Chicago Press},
year = {1962},
address = {Chicago},
note = {Often cited with various editions, e.g., 50th ed. 2012},
keywords = {paradigm shift, normal science, scientific revolutions}
}
If we’re talking about invention rather than discovery, who created the inflection point between potential and actual use seems relevant if not dispositive.
I typed this into Gemini, "who invented the transistor?" and it correctly cites "John Bardeen, Walter Brattain, and William Shockley at Bell Telephone Laboratories in December 1947".
> One of the first electric gates was invented by a Canadian Fred W. Watson in 1881. It was designed to be used for railway systems ... “A catch connected with an electro-magnet keeps a gate closed,” reported The National Tribune on October 9, 1884. [3]
I think a valid part of the question of who invented something is "who built the first working device" - describing something in theory and building working device are not the same thing.
AG Bell wasn't the first one to conceptually invent the telephone, he was among the first (along with Elisha Gray) in making practical working telephone and later a practical working telephone system.
So, who invented the Satellite then? What about the steam engine? The helicopter?
Sometimes the inventors are so far ahead of their time that the materials science first has to catch up (in some cases only a few millenia) before they can realize their devices. Effectively it is then the first person after whoever did the materials science part to create the device that gets to claim the invention.
So we get Sikorski, and not Da Vinci.
We get Arthur C. Clarke who claims the 'communications satellite' even though the moon was there all along and the Sputnik was the first working very crude device (it was one way only, it said 'you lost the space race' in a single bit of message).
We get Newcomen, Jerónimo de Ayanz y Beaumont (I had to look that up, I can never remember the man's full name), and Hero of Alexandria competing for the steam engine title, with all of them holding some part of the credit.
Pointing at an inventor is hard, and 'who built the first working device' is one way of doing this but it assumes a singular effort whereas most things are team efforts and misses the bit that the idea itself can be an instrumental step in getting your 'true' inventor to make their claim, standing on the shoulders of the giants before them. In isolation, we all probably would invent the hammer in our lifetimes, if that.
> Newcomen, Jerónimo de Ayanz y Beaumont ... and Hero of Alexandria
Thomas Savery too: https://en.wikipedia.org/wiki/Thomas_Savery
>> Newcomen, Jerónimo de Ayanz y Beaumont ... and Hero of Alexandria
> Thomas Savery too: https://en.wikipedia.org/wiki/Thomas_Savery
Savery is referenced in the "Watt steam engine" Wikipedia article.
Also worth noting that even though Watt improved on Newcomen's design to improve efficiency, Watt's design still sucked: it took Wilkinson's boring technology to actually get them to work well.
* https://en.wikipedia.org/wiki/John_Wilkinson_(industrialist)...
People underestimate how much improving precision over the years/decades/centuries has allowed technology to progress:
* https://www.simonwinchester.com/precision-praise
* https://www.goodreads.com/book/show/35068671-the-perfectioni...
Ah yes, and a better claim than some (there are quite a few more).
I always wondered what the effect of these in absolute time is. Consider: if someone had come up with a viable steam engine in the year 1000 or so, how would that have change the course of history?
I remember reading an analysis of that a while back but I can't remember where I saw it. The upshot was that the industrial revolution turned on more than just innovation, it also required just the right combination of natural resources (coal) and economic conditionsL deforestation in England driving the use of coal for heat, which drove mining, which drove the need for pumps, which made Savery's engine economically viable. Before that steam power would not have found an application.
Also, the piston and cylinder (which came later) were derived from canon-making technology, so that had to come first too.
Probably https://acoup.blog/2022/08/26/collections-why-no-roman-indus... HN discussion https://news.ycombinator.com/item?id=32607187 (509 points | Aug 2022 | 519 comments)
Yes, that was it! Thank you! How did you know?
Usually "Roman"+"Good post" or "Medieval European"+"Good" can be replaced by "acoup"+"I'm feeling lucky".
PS 1: If you can remember how to spell "acoup", it's the harder part.
PS 2: IIRC he only (mostly?) post about those topics, not about random old stuff, because it's his expertise area.
I remembered the same article too, I'm sure I could have found it again knowing it was listed on HN and/or recalling the wording. Good article too, going in depth on steam engines - the first ones were so inefficient they had to be really close to a really cheap source of fuel and be useful.
> t also required just the right combination of natural resources (coal) and economic conditionsL deforestation
This argument is repeated often, but I don't think it's really true. Both Savery and Newcomen's engines were initially aimed at evacuating flooded metal mines and not coal mines.
Source? Because Wikipedia says:
"A few Savery pumps were tried in mines, an unsuccessful attempt being made to use one to clear water from a pool called Broad Waters in Wednesbury (then in Staffordshire) and nearby coal mines."
That is correct, but the article explicitly addresses this point and argues that the evidence points to Lilienfeld producing a working transistor.
"Later, some people claimed that Lilienfeld did not implement his ideas since "high-purity materials needed to make such devices work were decades away from being ready,"[CHLI] but the 1991 thesis by Bret Crawford offered evidence that "these claims are incorrect."[CRA91] Lilienfeld was an accomplished experimenter, and in 1995, Joel Ross[ROS95] "replicated the prescriptions of the same Lilienfeld patent. He was able to produce devices that remained stable for months."[ARN98] Also, in 1981, semiconductor physicist H. E. Stockman confirmed that "Lilienfeld demonstrated his remarkable tubeless radio receiver on many occasions".[EMM13]"
For many things (computers, rocketry, aerospace, etc.) and different reasons, Germany in the years around the second world war, was a pretty bad place to get international credit for your accomplishments.
"It was able to squeak, but not to speak. Experts and professors wrestled with it in vain. It refused to transmit one intelligible sentence." [0]
"A translation of Legat's article on Reis' invention was obtained by Thomas Edison prior to his filing his patent application on a telephone in 1877. In correspondence of 1885, Edison credits Reis as having invented "the first telephone", with the limitation that it was "only musical not articulating"." [1]
Fascinating stuff nonetheless, these inventors and their ideas... See also previous experimenters [2]
[0] https://en.wikipedia.org/wiki/Johann_Philipp_Reis
[1] https://en.wikipedia.org/wiki/Reis_telephone
[2] https://en.wikipedia.org/wiki/Johann_Philipp_Reis#Previous_e...
Yeah - though Bell's first apparatus wasn't much better - the invention of the carbon microphone is what really what set the telephone on to being a practical device. The rest of it was trying to build a network to connect people - and that was really hard (and capital intensive).
What blows my mind that we absolutely take for granted today is insulated wires. The technology and supply chain to mine or to find into metal and also to farm cotton and wool and formed that into protective tubing before the advent of plastic insulation. The amount of technology that goes into making a "simple" USB-cable beggars belief if you stop to think about it. Even a simple #2 wooden pencil with an eraser on top is beyond the knowledge of one person to produce, nevermind a USB-c cable!
In Victorian London, electricity was distributed around the home using bare wires with an air gap. It was 32v though.
Later it was superseded with lead wrapped in paper, until the Knob and Tube system. This comprised of single-insulated copper conductors installed within walls and ceilings, this wiring was encased in porcelain insulating tubes with cloth-lined sleeves.
One knob for Live and one knob for Neutral. The wires were held in place by porcelain knobs nailed to the house frame. Where wires passed through wood framing, they were threaded through porcelain tubes to prevent them from contacting the wood.
My house once had knob and tube wiring. Over the past 15 years I have replaced most of what I assume is the second-generation cloth-covered wiring (which dates back to the 40s, 50s, and 60s). Every once in a while, I come across the insulators from an older electrical system, but most of the wire that went with those insulators was pulled out long ago. The only remains are short bits of wire that were wrapped around the insulators.
https://en.wikipedia.org/wiki/Knob-and-tube_wiring
I found some of that in a very old farmhouse. It looked like a very good way of setting fire to a structure.
I have a cable here that is interesting. When I first saw it I thought how strange, a cable that is multi-stranded bare wire for this application (connecting a camera). Then I looked at it under a microscope and realized not only is it insulated wire, it is shielded wire. Mindblowing. I pity the people that have to handle that stuff.
https://down-ph.img.susercontent.com/file/sg-11134201-7repx-...
Indeed.
Small addendum: at least in Germany, early telephone wires (up until the 1950s?) were wrapped in paper, drenched in oil. The bundles were then enclosed in a lead-copper alloy to protect them from moisture.
https://commons.wikimedia.org/wiki/File:KabelbaumFMAachen_49...
Well, and the first really 50 years of telephony - you were probably as likely to have an uninsulated line as an insulated one.
Remember that the first telephones were 1-wire with a ground return, and that configuration was quite common until the early 40's.
Reminds me of the (not entirely accurate) story of someone trying to make a cheeseburger from scratch and realizing it required most of modern civilization.
To some degree, this is a consequence of the nature of the field you're working in:
* if the physics is so completely understood that you can confidently predict something will work from your sofa, and give an error-free recipe to build it, you indeed can invent from theory... but how deep can this invention be if the problems of the field are completely solved?
* if you are working in a field at the edge of human understanding, you cannot have the confidence in your ideas without having tested them experimentally; a theoretician makes at most a minor contribution to the actual inventions being realized, because he's producing - most likely somewhat wrong - hypotheses.
This latter kind of "theoretical" inventions are heavily subject to survivorship bias. Fifteen competent theoreticians make different predictions - all according to best, though incomplete, model of the world; a successful experiment validates exactly one of them, and we end up exalting the lucky winner as the "inventor".
> how deep can this invention be if the problems of the field are completely solved?
You're confusing depth with originality.
A field may be very well understood, but also very deep.
In theory, that's true.
In practice, any unexplored corner of the field will contain surprises; these will require extra theoretical development to cover.
Usually things like imperfect understanding of materials get in the way. Pretty much the reason you need both theory and experiment to make progress in every single area of matter-based technology (i.e. not software).
I had that thought too, describing that something might be physically possible isn't really inventing it, you have to build (and arguably sell) the device too. Re-organizing someone else's equations and saying it's technically possible is maybe enough to publish a paper but certainly doesn't rise to the standard of inventing in my mind
There's something to be said that mass production is another distinct stage of invention. Karl Benz may have invented the first internal combustion engine car, and plenty more built cars by hand for the rich, but Henry Ford made cars anyone could have for cheap.
I agree with that. It is a bit like the idea that having an idea for a startup is less important than the ability to get it to a level where investors might be interested.
A while ago there was an artical posted here about all the world changing inventions that came out of Bell Labs. It was easy to show for most of them that they weren't the actual inventor, and in many cases not even the first producer. They were the first to make it practical for mass production, however.
You may like this book called The Origins of Efficiency: https://press.stripe.com/origins-of-efficiency
Or the author's newsletter: https://www.construction-physics.com/
As Elon Musk said, the machine to make the machine is much harder to make.
If the patent office would require a working prototype it would fix patents. Working device is a good standard.
The only point in asking in the first place is pride and/or greed.
Theodore Maiman and the laser.
What’s up with the ending that makes no sense?
>Where are the physical limits? According to Bremermann (1982), a computer of 1 kg of mass and 1 liter of volume can execute at most 1051 operations per second on at most 1032 bits. The trend above will hit the Bremermann limit roughly 25 decades after Z3, circa 2200. However, since there are only 2 x 1030 kg of mass in the solar system, the trend is bound to break within a few centuries, since the speed of light will greatly limit the acquisition of additional mass
They shift from talking about the transistor density to somehow considering a supermassive construct. Reminds me of LLM mashups.
It's a natural extension of the ideas being discussed - the limit in computation per gram of mass has energetic bounds, as well, with configurations nearing the upper limit that start looking more like nuclear explosions than anything we'd regard as structured computation. The extremes are amazing to consider - things that look and act like stars, but are fantastically precise Turing machines, and so on.
It's a theme that sci-fi authors have explored deeply. Accelerando is a particularly fun and worthwhile read if you haven't already!
The Bremermann limit is about computational density. It makes sense to talk about computers that are ever more powerful but not necessarily larger.
So we talk about the supercomputers we call cell phones that are orders of magnitude more powerful than the desktop computers I used years ago.
It doesn’t make sense to talk about making super large computers before reaching the density limit, that’s a confusion of concepts.
Making computers faster has involved making them smaller because the speed of signal propagation.
> It's a theme that sci-fi authors have explored deeply.
Having ingested every science fiction book that has been digitized, a fun evening to be had is writing your own book with ChatGPT.
Computronium! Kurzweil goes into this in the Singularity is Near.
I've always wondered if Warlock from the New Mutants was made of it.
It seems to refer to the previous paragraph:
> The naive extrapolation of this exponential trend predicts that the 21st century will see cheap computers with a thousand times the raw computational power of all human brains combined
i.e. putting an upper bound on the exponential with solar system mass
Is there any evidence that Shockley was aware of Lilienfeld’s FET? I mean, given Shockley’s general disposition, I wouldn’t be surprised if he’d stolen things, but we have his early prototypes and such to back up his claim of invention. The next step is to prove a through line to the modern era.
The Sumerians knew geometry, trigonometry, and other advanced mathematics. That knowledge perished, and then the Greeks rediscovered things. The Greeks then get credit because their work is what paved the way for all of us today. The same would apply to Shockley. If Lilienfeld’s work didn’t make it to the USA in time to have had an effect on development in the valley, then it doesn’t really matter for the US industry.
This lists many transistor patents from oldest to newest.
https://patents.google.com/?q=(H03F3%2f16)&sort=old
The Matare/Welker Patent is missing though
https://patents.google.com/patent/US2673948A/en.541
The entire debate is tiring. It would be better if these reviews would put the actual device physics of the different concepts into context.
Is there any report of a reproduction of the device proposed by Lilienfeld in his patents? If he managed to make functional devices back then, it should be possible today? (Note: Cu2S is not a very well controllable semiconductor...)
Edit:
Gemini Deep Research summary here, its quite informative: https://docs.google.com/document/d/1jE0wQVeWP9Eiybh_C6zMKeZ5...
Also specifically on Cu based TFT: https://docs.google.com/document/d/1_B2x2gBPKgGFVgJyQ0qzPdI4...
From the second document: "The primary obstacle for $Cu_2S$ TFTs is degeneracy. Spontaneous copper vacancies form with negligible energy cost in the sulfur lattice. As a result, stoichiometric $Cu_2S$ is thermodynamically unstable in air, rapidly oxidizing or losing copper to form substoichiometric phases ($Cu_{2-x}S$) with hole concentrations exceeding $10^{20}-10^{21} \text{ cm}^{-3}$."
This explains why there are zero reproductions of Lilienfelds devices. It should be noted that Lilienfeld is one of the inventors of electrolytic capacitors and did therefore know where well how to create extremely thin insulating layers as needed for TFTs. It is not impossible to assume that he could have used other semiconductors (e.g. CdS) with his concept. However, the patents seems to specifically mention Cu2S, which does not yield functional TFTs.
In his notes from November 1, 1913 Thomas Henry Moray described what later would be called the Moray valve. It used germanium.
Shockley developed the first working bipolar junction transistor, which was the first transistor to ever enter mass production, which is why he's credited for it: https://en.wikipedia.org/wiki/Bipolar_junction_transistor
Though I'm unsure of the author's motivations, there's certainly a tendency with Shockley similar to what one sees with Hitler, where people will occasionally question or cast doubt on Hitler's WWI service record--which was by all accounts exemplary. (I.e., Bad Guy can't also possibly have done Good Thing.)
The term "transistor" is much too vague, so the question "who invented the transistor" is ill posed.
There are different kinds of transistors, which use different principles for controlling electrical currents, and which have been invented by different people at different times.
Lilienfeld has invented 2 kinds of transistors: MESFETs (metal-semiconductor field-effect transistors) and depletion-mode MOSFETs (metal-insulator-semiconductor field-effect transistors) (US patents 1,745,175 and 1,900,018). Of these 2 kinds of transistors, the second is rarely used today, while the first is used mostly in special applications for discrete power devices.
The parent article is very wrong in claiming that the transistors invented by Lilienfeld are those that are most frequently used today, which are the enhancement-mode MOSFETs, which have a very different structure and for which much less of the semiconductor materials are suitable than for the simpler transistors invented by Lilienfeld.
The enhancement-mode MOSFET has been invented in 1960 by Martin M. Atalla and Dawon Kahng from Bell Laboratories (US patents 3,206,670 and 3,102,230).
Bardeen and Brattain have discovered the point-contact transistor, which had been used only for a few years before becoming completely obsolete, but which was the first kind of transistor offered as a commercial product.
William Shockley has invented 2 kinds of transistors, which have been very important and which remain the best in some special applications, the bipolar junction transistor (BJT) and the junction field-effect transistor (JFET) (US patents 2,569,347 and 2,744,970). William Shockley has also the huge merit of developing a theory of the physics of semiconductor materials which allowed everybody to design transistors and many other kinds of semiconductor devices.
Therefore the 3 Nobel winners have invented together 3 kinds of transistors and they are rightly called inventors of some of the kinds of transistors, i.e. point-contact, BJT and JFET.
Lilienfeld was the first who conceived some possible structures for a triode that uses a semiconductor instead of vacuum or gas. This was a great advance, but this still resulted logically from the prior knowledge that one can make diodes using either vacuum or gas or a semiconductor, and one can make triodes using either vacuum or gas. Thus he attempted to fill the missing combination.
The Bell Laboratories are truly guilty of trying to hide the fact that their post-war research on making a semiconductor triode had indeed started with the purpose of making alternatives to the devices invented by Lilienfeld, of which they were well aware. However, they eventually invented 4 different kinds of transistors, all of which had structures and principles of operation very different from the 2 Lilienfeld transistors, even if all 6 kinds can be considered as variants of semiconductor triodes.
The first 3 kinds of transistors that have been invented, the 2 Lilienfeld transistors and the point-contact transistor of Bardeen and Brattain, can be made using pieces of a homogeneous semiconductor material. This is why they have been discovered first.
The next 3 kinds of transistors, invented by Shockley and by Atalla with Kahng, contain P-N junctions, so they were invented only after William Shockley had developed the theory of the P-N junctions.
When you come in the UK as a continental european (french for that matter), you notice you're inundated with tons of trivia of how great british inventors are, and how many things they invented/discovered. They even make TV shows about it. Nothing like this exists in France and people have no clue about who invented stuff, nor do they care.
Now with internet, where truth is supposedly coming from english world, there are plenty of opportunities to rewrite history 1984 style. It's really damning now with AI auto translating every english to your local language.
It's going to be Schmidhuber, isn't it?! /s
I wish Jürgen Schmidhuber would switch back to actually doing AI research instead of having become completely obsessed with "who invented what" because he feels like he has somehow been academically "robbed" at some point in his career.
He's now officially become a full-blown pariah in the AI world, most relevant people in the space running away at the first sight of his goatee at conferences, knowing exactly the kind of complete and utter crank he's become.
Was anything he claimed in the article incorrect? Personally, I enjoy these types of historical stories.
I'm not criticizing the article at all.
In fact I am generally ignorant on the topic of who invented the transistor, nor do I in general particularly care about who invented what.
The quest for academic fame is something I've always utterly failed to understand.
And, if it the author was anyone but JS I'd not have said anything.
What honks me off about this guy though is to see a someone who did in fact do early impactful work on recurrent neural networks believe that:
a) that automatically gives him some sort of special status wrt the rest of humanity
b) because he didn't get the recognition he believes he is due, has completely stopped doing anything useful in the field, turning instead into an absolute crank that every one in the AI field makes fun of, and with a holy mission to rewrite history to assign credit where credit is due everywhere he believes there was an injustice.
c) every time I see someone with an exceptionally well-working brain waste their time because of ego or sheer stubbornness on shite like this instead of using it to do more interesting work, it makes me very sad.
Schmidhuber is a textbook example of this, and the other perfect example of this is Chomsky, a very smart man, who basically - because of his oversized ego and profound stubbornness - ended up wasting his entire life energy working on a linguistic dead-end AND a political philosophy dead-end.
I have a real hard time understanding how the brain of that kind of folks operates, being so bright on certain axes and totally and utterly dumb on others, especially the total lack of self-awareness.
Wow that is super vitriolic, more if from a colleague. Sure the hehe debate is about if he is to be considered more or less accomplished, but going from non-success to pariah, what is the need?
> but going from non-success to pariah, what is the need?
Not sure "need" is the appropriate word here.
Grab anyone in who has worked on AI in the last 30 years, and pronounce the word "Schmidhuber" and watch the face of you interlocutor: you'll either get an eyroll or a smirk, but rarely a lively discussion on what he's "invented".
Nothing vitriolic about describing reality.
I'll just leave this here https://scholar.google.com/citations?user=gLnCTgIAAAAJ&hl=en so maybe you realize that's a bit of a tall claim from a random about one of the top researchers in AI, no matter what their opinions are. Perhaps you should look up what a "crank" actually is before labeling researchers, just because they don't match your religion.
Your link did not work for me. "We're sorry…but your computer or network may be sending automated queries."
It's his Google Scholar profile; you can search for it.
It's Juergen Schmidhuber's Google Scholar page
Did I claim anywhere that his early work wad bad?
Nope. The contrary as a matter of fact. But the facts are:
1) nothing worth talking about since the LSTMs
2) most of the reasons why he's been visible in the last 20 years is because of shit like this (JS harassing Ian Goodfellow about attribution in the middle of a technical presentation. Watch the face of Ian, pretty symptomatic of when AI folks have to interact with JS these days).
https://www.youtube.com/watch?v=AJVyzd0rqdc&t=3778s
He is usually insufferable (and misleading) when he talks about AI innovation priorities. This article is different, though. It really does seem like Lilienfeld invented the transistor first and should be given credit for that. It also really does look like the official inventors knew that and were somewhat dishonest.
Funny, btw, that nobody here has mentioned that Lilienfeld also invented the electrolytic capacitor.
But: it's obvious why he wrote the article. It is to bolster his own claims, not to give Lilienfeld his due.
How would you distinguish the article from an honest write up about transistors? That is, you know about his crusade in ML, but if you didn't, how would you decide that this article is written in bad faith or not?
I agree that context matters, and I had the same thought as you. But does that mean that anything he writes on the topic of "who was first" is inherently tainted?
Because I read the article and checked the citations. It's a dead giveaway.
I personally detest the way we sanctify some sole individuals while forgetting the bulk of the community. I don't care who published the first patent for the transistor. He or She certainly cannot be credited for all the work that has been put into it so that I can today use a hand held device to post this comment.
I see where you're coming from, but while that's the case with most stuff ("normal science") , it often isn't the case for truly revolutionary stuff. Many breakthroughs happen not because of the bulk of the community, but against it, often at the highest cost for the individuals.
Surprisingly also not true. Yes, people go against the grain and it is required to actually make paradigm shifts but they're never alone nor did they build from scratch. It may be few against many but it is almost never one against all. That one only prevails due to support from others. Those names don't shine but it doesn't mean they weren't critical to the advancement of a field
Strong claims - maybe good time to do some homework instead of arguing without evidence?
Galileo was sentenced to house arrest for heresy. Boltzmann died by suicide after lack of acceptance by the scientific community. It's a very long list and something that's been studied, actually.
You do realize Galileo and Kepler were around at the same time, right? Galileo is the best example of what I claimed.
You criticize me for "not doing my homework" yet you haven't heard of Castelli[0]? He's even in the fucking Wiki article on Galileo lol[1]
Your knowledge of Boltzmann seems to be as deep as your knowledge of Galileo. You're absolutely right. You should read some of it. Or if you don't like reading I do highly recommend An Opinionated History of Mathematics[2]. Blåsjö even has a whole season dedicated to Galileo.[0] https://en.wikipedia.org/wiki/Benedetto_Castelli
[1] https://en.wikipedia.org/wiki/Galileo_Galilei#Controversy_ov...
[2] https://www.podbean.com/podcast-detail/85etf-7ecf8/Opinionat...
Sorry, but having two people instead of one pushing for similar/related ideas is not supporting the main claim ("we sanctify some sole individuals while forgetting the bulk of the community"). Nor is stealing ideas and having the wrong name on results. It's rather common to have multiple names in theorems because of this. That's still beside the point, when claiming it's the "community" or hive mind.
It's not that I don't see where you're coming from. What's been gesticulating at is called "Stigler's law of eponymy" (https://en.wikipedia.org/wiki/Stigler%27s_law_of_eponymy). Pythagorean theorem was known 1000 years before Pythagoras and a ton of results are wrongly attributed to. But that doesn't make revolutions a collective effort.
Was Galois', Gödel's or Grothendieck's work obviously the common sense of the community? Or wasn't it the "community" that tried to "solve" the parallel postulate for 2000 years before Bolyai did?
Or maybe everyone in the Copenhagen crowd cheered Bohm, Everett and Bell because it was all obvious in the community?
Thanks for the link, but when I said "do homework" I didn't mean podcasts :), but rather either general or concrete work in the history and philosophy of science, like Thomas Kuhn's "Structure of scientific revolutions".
It seems like you agree with me yet simply do not easy to. I can't do anything with that. You seek evidence that is incredibly difficult to present. History is recorded differently now than in the past yet things are the same. You're ignoring the biases of how it's recorded. Any example I give you'll say it's not enough and point out another. It's a fruitless game
Maybe you should judge the information on its merit rather than its mediumPossibly. I don't think either extreme opinion is correct— i.e., both can be partially correct, hence we may agree on that. Merit is often arbitrarily attributed (and sometimes, perhaps often enough, wrongly). I don't think that's enough to say that that's "always" or "most often" the case. That was my main point. You argued that no one does something strictly alone or from scratch. I think the "from scratch" part is irrelevant and think that a single person, a very small group, or a few independently (even if at different points in time) still warrants the same individual recognition.
Neither really fits what Jürgen Schmidhuber argues for. He clearly argues against intentional non-attribution and/or ignorance. Both are evidently the case deeply in current AI research (probably also outside, but this is what OP is about). At the same time, claiming that his points are worthless because he's just pointing to other individuals that suffer from the same misattribution problem is generally a wrong take. Work concentration and results are simply non-linear and generally clustered around individuals and groups. Failing to attribute, either from ignorance or malice, is equally wrong.
“It takes a thousand men to invent a telegraph, or a steam engine, or a phonograph, or a photograph, or a telephone or any other important thing—and the last man gets the credit and we forget the others. He added his little mite — that is all he did. These object lessons should teach us that ninety-nine parts of all things that proceed from the intellect are plagiarisms, pure and simple; and the lesson ought to make us modest. But nothing can do that.”
― Mark Twain
This is understandable, but maybe you cannot offer an alternative to just give credit nobody. Schmidhuber is playing the game by its rules being self-consciously contrarian enjoying himself, while offering an interesting historic pinch of retrospective justice.
We all stand on the shoulders of giants. Giants who are just a bunch of people in a trench coat
Oskar Heil is a curious name, especially considering where and when he was from
Check translate.google.com, for example: the morpheme is overloaded semantically so it's (topically) coincidental.
This is a good piece of writing that nicely illustrates how what we perceive as "who invented something" is mostly a function of money and politics.
If we’re talking about invention rather than discovery, who created the inflection point between potential and actual use seems relevant if not dispositive.
Yeah, it's all about the inflection point. Otherwise how are you even dispositive?
I typed this into Gemini, "who invented the transistor?" and it correctly cites "John Bardeen, Walter Brattain, and William Shockley at Bell Telephone Laboratories in December 1947".
If Lillienfeld had working examples, how is Gemini correct that it was the Bell people?
When I ask Gemini and look at the sources it links this article...
Who invented the electric fence gate?
How does the electric fence gate lead to transistors?
Relay > History: https://en.wikipedia.org/wiki/Relay :
> [1809, 1835, 1837, 1840 (Samuel Morse; Morse telegraph) ]
Electric gate > Electric Gate History: https://en.wikipedia.org/wiki/Electric_gate :
> One of the first electric gates was invented by a Canadian Fred W. Watson in 1881. It was designed to be used for railway systems ... “A catch connected with an electro-magnet keeps a gate closed,” reported The National Tribune on October 9, 1884. [3]
Flip-flop (electronics) > History: https://en.wikipedia.org/wiki/Flip-flop_(electronics)