> the high velocities of steam locomotives might kill you
This obviously seems silly in hindsight. Warnings about radium watches or asbestos sound less silly, or even wise. But neither had any solid scientific studies showing clear hazard and risk. Just people being good Bayesian agents, trying to ride the middle of the exploration vs. exploitation curve.
Maybe it makes sense to spend some percentage of AI development resources on trying to understand how they work, and how they can fail.
> Warnings about radium watches or asbestos sound less silly, or even wise. But neither had any solid scientific studies showing clear hazard and risk.
In the case of asbestos, this is incorrect. Many people knew it was deadly, but the corporations selling it hid it for decades, killing thousands of people. There are quite a few other examples besides asbestos, like leaded fuel or cigarettes.
People thought that the speed itself was dangerous, that the wind and vibration and landscape screaming by at 25mph would cause physical and mental harm.
Yes, people grow old, but I don’t think there’s a rational fear people will ever live forever. However, there are diseases and conditions that are age-related that if solved would improve people’s quality of life, especially the last decade or two of their lives.
I believe it would be a far greater problem if those people didn't die. Aging populations are a huge problem around the globe and unless we'd improve the quality of life to such a margin that octo- and nonagenarians are able to care for each other, I think we're all better off with people dying of old age.
That is assuming the accelerating AI stays under human control.
We're racing up a hill at an ever-increasing speed, and we don't know what's on the other side. Maybe 80% chance that it's either nothing or "simply" a technological revolution.
The youngest was born in 2019, the same year he was incarcerated (April 2019). Pregnancy lasts 9 months, so even if the child were born in early 2020, there would be no reason to assume infidelity.
I wasn't assuming infedlity, I could have been more clear there. I really was just curious on timing how none of his children could have met him before he was imprisoned.
I was at Google when that happened, and had some inside sources (ie gossip from the guys who interviewed him, who said they didn't even ask any tree questions, and memegen).
I tried to find some outside sources, but nothing about Howell admitting his 'artistic license' outright. So take my fallible memory with a large pinch of salt.
> He is critical of string theory on the grounds that it lacks testable predictions and is promoted with public money despite its failures so far,[1] and has authored both scientific papers and popular polemics on this topic. His writings claim that excessive media attention and funding of this one particular mainstream endeavour, which he considers speculative, risks undermining public faith in the freedom of scientific research. His moderated weblog on string theory and other topics is titled "Not Even Wrong", a derogatory term for scientifically useless arguments coined by Wolfgang Pauli.
I'm sure he is expressing a valid principle. By funding one area excessively, you can soon build up reams of scientific literature and interest, but this area needs have nothing or little to do with genuine scientific interest. Funding determines the interest. It's all about money unfortunately, even science.
I had the same feeling doing a phd two years ago. I think there is an unpopular opinion in science about the limits of what we can look at and what is required to achieve that. My conclusion was that the requirements are an economy of infinite growth and a society based on consumption. I did a parallel in my mind between the idea that "knowledge will always grow" and "economy will always grow".
The limits are heavily related to social stability/agreement, and the tools we can have given the money(=social and environmental)/physical constraints.
I have the opinion that the FCC is the example of such bias: we don't really know what to look for, but we (the scientific community) have to survive so we'll build a political argument to keep getting funds.
I think the proposal was during a severe heatwave, and I also though "where is the social goal in that science? What does it will bring to society? Do we really need to know that far those things?". I think it's at this moment that I started loosing motivation too.
What is he in favor of? He seems to be criticizing both further experiments in string theory due to being not even wrong, and further experiments with the standard model, since it's either expensive or not getting anywhere. (I'm not a physicist and understand very little.)
It's not necessary to be in favor of anything, in order to criticize what's already out there. You can point out fatal flaws in an idea without needing to have an alternative in mind.
Woit himself is working (pretty casually to my eyes) on some mathematical reformulations of the Standard Model. He's not advocating aggressively for these ideas, and I'm pretty sure if you asked him, he'd say there's no evidence that any one idea should dominate the research landscape the way string theory did. None of what exists out there is particularly compelling. (This is one of the reasons people worked so much on strings. It's the best of a bad lot.)
- "...and is promoted with public money despite its failures..."
The amusing thing for me* is this is just coffee budgets for pen-and-paper theorists. What they say about academic disputes: the lower the stakes, the more intense the politics.
It’s not just coffee and blackboards and hoarded Japanese chalk: the goalpost slalom around supersymmetry drives discussions about what colliders to build and how to operate them [1]. Before scalar field excitation at 125 GeV it was predicted by many that the power and luminosity of that run would show weak bosonic superpartners in the first run. With Higgs at 125 GeV it gets really tortured as an argument.
This is also the subtext with the really aggressive public branding of “dark matter”, when it should really be called something like “large scale apparent gravitational anomaly” or some dry thing like that, it’s not an MCU franchise: positing a bunch of mass that has none of the other properties of matter is a perfectly fine line of inquiry, but the verbal capitalization of Matter is because weakly-interacting massive particles are another way to argue that maybe, just maybe maybe, this is indirect evidence for supersymmetry.
But most of all the damage is in attacking the definition of science: if you envelope-math metastable vacua consistent with compactified Calabai-Yau dimensions at (last I checked) order of 10^250 what you’re left with is “it’s strong anthropic, there’s no explanation”, which is exactly where Susskind and that lot have ended up.
There's also a hidden component in these budget calculations: it is hard for people doing more "traditional" physics to find tenured positions, since many faculty hires went to string theory. So the salary that went to string theorists at the expense of other subareas of physics is quite a large hidden component. Crushed academic ambition is as real-world as it gets, since it involves years of extremely hard toil, wasted.
Totally wild assertion. Most physics departments in the world have only one or two theorists and most of them are not in string theory. And most young physicists don't have the inclination to pursue string theory (even if they had the capability, which many do not).
Taking criticism of string theory hogging all the budget as "it's literally taking over the entirety of physics budgets" instead of "in the field of fundamental high-energy physics, there's no budget left over for alternative ideas to be developed to a similarly detailed level" is strawmanning.
This is just false, string theory only competes in the theoretical physics sub-area; condensed matter, astrophysics, lasers, all that stuff have their own pots of money.
That research budgets are split by subjects many times before arriving to a node where "string theory" is a possible leaf is not a controversial statement, it is reflected in basically all budget documents you will find.
For example, string theory funded by the NSF that "steals" money from laser research is plausibly only found in the "elementary particle physics - theory" program, which is part of the Physics division, which is part of the Mathematical and Physical Sciences Directorate.
sadly the page is pretty shit so the filter selection is probably reset, and they don't label by division program so you'll have to mouse over each one and categorise yourself. As a guide to how much work it is, there was 350 awards in the physics division and about 30 of those in the theory program.
> The amusing thing for me* is this is just coffee budgets for pen-and-paper theorists
Unfortunately not. From his website [0], the extent of the grant funding involved is much more than just coffee budget:
> The Black Hole Initiative that features this on its website: $16 million from the Templeton Foundation, $3.6 million from the Moore Foundation.
> The Simons Collaboration on Celestial Holography: $8 million from the Simons Foundation.
> NSF Grant: $400,000 from the NSF.
> DOE Grant: $3.5 million from the DOE.
This kind of money could fund a whole lot of other theory. Hell, it even could fund a lot of experiments (albeit not in high energy physics).
In my view, those are very trivial amounts, for an entire subfield of research. I can think of individual federal grants no one on HN has heard of or cares about that are are larger than all of those put together, which are outright fraudulent—those are dime-a-dozen.
Also, note that your first three examples aren't public money, rather private philanthropy. No one can speak against where Jim Simons gifts his billions (and in point of fact Simons is an expert in quantum field theory himself—no one's scammed him, if he's decided string theorists are worth donating to. He reads and understands the papers they write).
It's a fair anchoring point, isn't it? It's the theory and experiment side of the same field. We're just spending 0.01% of the experimental budget on some (possibly wrong and possibly dead-end) theory ideas, and the coffee that produced them.
In context, those are not shockingly high sums. The real problem seems to be what Woit summarizes in this 2004 (!!) comment on his blog:
> It takes a non-trivial amount of time and effort to absorb new mathematical ideas and by so dominating the mathematical end of particle theory for twenty years, string theory has monopolized the time of the mathematically sophisticated members of the community. It has also quite literally driven out of the field a lot of people who were interested in other sorts of ideas about how to apply mathematics to questions in particle theory.
Well, its not true that people only worked on string theory during these last 20 years: notably Woit himself didn't. That there's loads of people in Brazil, Russia (or frankly any place except Princeton and IAS) trying weird approaches sums up to nothing in his telling.
Honestly, I get the impression that what Woit is really upset about is that people like his idol Witten didn't switch to work on his ideas, because only the genius of "towering intellects" like Witten's could solve this very hard problem. 0
The sad thing about that funding is how small it is. To billionaires and governments this is pocket change. Simons (personally, not the foundation) could spend ten times that to fund research into alternative models and not even notice.
A global anything-goes-if-you're-qualified frontiers research program would cost a few hundred million dollars. The odds of it finding some game changers are likely pretty good.
Instead we're getting a $17bn revamp of the LHC to turn it into a "Higgs factory".
I know the word literally doesn't mean anything anymore, but Jenson and Lisa Su are first cousins once removed. Jensen's grandfather and Lisa's great-grandfather are the same person.
I know what the word "literally" means, there's a great Sorkin bit on it [1] that's eerily prescient given that show is like, 15 years old. Via your own citation it's apparently Lisa Su who doesn't know what "literally" means, as she asserts that they are "second cousins", which is literally false according if your second citation says "first cousins, once-removed", which Quora [2] says means they're closely blood-related.
I don't know how their family works, but in mine and most people from my neighborhood, a first-cousin once-removed is fucking family, they're blood. Not being a securities lawyer myself I'm not sure which definition, statue, or regulation would apply here, [3] seems close (and has a creepy rush-job feel about it that smells vaguely like Kushner shit of one kind or another, Feb 2020 on an accelerated basis?).
But whether this squeaks above the line of regulations and laws and whatnot getting midnight "lgtm" stamps in an election year is, I'd argue, substantially missing the point.
When I recently said:
"Now did Lisa Su decide to "concentrate on the supercomputing market with the MI300XYZ" and Jensen decided to "concentrate on AI with Hopper" independently to a degree where the market is perfectly partitioned? Who knows, I certainly don't have proof one way or the other. But if someone made a call being like "I'm thinking of focusing on X but don't really see our differentiation in Y. How's Cathy?", it wouldn't be the fucking first time." [4]
I thought at the time I was kinda pushing it with how flip that sounded, but lo and behold, I was insufficiently cynical.
So when I say that I literally don't understand why anyone is defending this trivially dubious cartel behavior complete with a 55.58% Net Profit Margin in an ostensibly competitive market both directly and indirectly subsidized by the taxpayer (TSMC isn't going to fight off the PLA with their next process node) [5], I think Leona Lansing knows that the public will burn the building down with this shit in it before they let this shit get much ickier.
This obviously seems silly in hindsight. Warnings about radium watches or asbestos sound less silly, or even wise. But neither had any solid scientific studies showing clear hazard and risk. Just people being good Bayesian agents, trying to ride the middle of the exploration vs. exploitation curve.
Maybe it makes sense to spend some percentage of AI development resources on trying to understand how they work, and how they can fail.