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.
"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).
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".
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.
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.
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
>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
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 we’re talking about invention rather than discovery, who created the inflection point between potential and actual use seems relevant if not dispositive.
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.
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.
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.
@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}
}
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.
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".
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.
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.
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 of 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).
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.
"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).
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".
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.
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.
Theodore Maiman and the laser.
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
The only point in asking in the first place is pride and/or greed.
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!
Computronium! Kurzweil goes into this in the Singularity is Near.
I've always wondered if Warlock from the New Mutants was made of it.
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 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
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?
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.
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.
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.
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...
We all stand on the shoulders of giants. Giants who are just a bunch of people in a trench coat
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.
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.
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.
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.
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 of 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
It's going to be Schmidhuber, isn't it?! /s