Science absolutely depends on people meeting and exchanging ideas. Whether that happens at a conference as they exist today is one thing, but if you don't spend money on getting people from around the world to meet and exchange ideas, you're going to lose a key aspect of the scientific process.
> If the OP's specific research was into The Changing Mating Habits of the Delta Smelt Due to Habit Destruction, then probably it was money that could far better spent paying tuition for, say, medical students or even just letting tax payers keep their money and spend it in a way that directly benefits their family, their community, and themselves.
The problem is it's very hard to know ahead of time which research directions will yield fruit. If we knew how to only fund good research, then science funding would be very easy. Unfortunately, that's not the case -- oftentimes things that are sure bets fail, and things that are rejected as "not promising" result in a breakthrough. So we have to fund a lot of stuff, some of which is not obviously going to yield a great ROI.
On the one hand, yes, funding science the way we do results in a lot of "wasted" funding. There are tons of inefficiencies. On the other hand, the way we fund science has been wildly successful in terms of the benefits we have reaped. Look around you, you can see them everywhere in every sector.
The danger is we pull back funding to things that are "sure bets" and they turn out to be duds while we miss out on other less sure opportunities. That would be a loss for everyone involved.
Please read my sibling reply about DARPA grand challenges. This knowledge was built using public dollars by people who publish papers, which are being read today by people building products. That's the great cycle of progress.
Notably DARPA felt the need to do this because they didn't trust private industry to do it on their own; with no money in driverless cars, the government figured industry would get there only if there was some catalyst, which they provided (successfully).
If you only ever go where the work is, then you're going to be left behind by societies that have a vision and leadership that will work to make it real.
That's an interesting counterfactual but it doesn't really mean much to reality. Fact is, what happened is that public research got the results where private industry did not, and everyone is better for it, including private industry. Progress doesn't wait for private industry to be certain of its profits; if it doesn't happen here, it'll happen elsewhere.
It also complete bullshit, pharmaceuticals are heavily dependent on the basic research that is done by colleges and anyone that thinks these private entities can do more research than the best universities in the planet is just insane.
For decades now the main difference between the US and the other economies is the amount of highly qualified labor across the board, this move to destroy academia and elevate the stupid and unqualified will be the end of America and the whole world will be poorer for it.
Yes, the entire DARPA "challenge" series has been about jumpstarting the US robotics industry. People who were involved in those went on to found driverless car companies, which then went on to create a market for driverless cars, and now America is a leader in the industry.
And it needed to happened because the state of American robotics was sad in 2004; the very first challenge was a disaster when all the cars ran off the road, with zero finishing the race. Top minds from MIT and Stanford got us that result. But they held the challenge again and again, and 20 years later we have consumers making trips in robo taxis.
e.g. Kyle Vogt, participated in the 2004 Grand Challenge while he was at MIT, went on to found Cruise using exactly the techniques that were developed at the competition.
So while Elon Musk is busy slashing whatever federal spending he can through DOGE, it's only because of federal spending that he can even fantasize about launching a robot taxi service.
America the leader in the driverless car industry? Not entirely sure it is still true. At least might not be true much longer. China is already building EVs en mass and some of them have, according to some people I met at least, better self-driving capability.
Well that is a goal post. They can do it doesn’t matter if they are allowed to or not. My friends tesla I’ve seen it in action. It goes on the highway. It goes on all the streets even hilly ones unlike waymo. It can even find a free parking spot in the lot and fully park the car. All while he does nothing at all with his hands on his lap. So yeah it totally can self drive.
If that's how you're using it, then you're doing it wrong, because you're supposed to be supervising the drive the entire time. It's not self-driving if you're supervising. And you're required to supervise because... they're not confident it can actually self-drive.
where did i say i wasn't supervising? Are you saying the self-summon is the unsupervised part? Cause i can flip that logic on it's head and say, if they weren't confident that it can summon to me, why would they implement the feature?
The charge is that Tesla can't do self-driving successfully. If they could, you wouldn't need to supervise, as is the case with e.g. a Waymo taxi. That they require you to supervise is an admission that their system is not sufficient for self driving, i.e. they're not doing it successfully.
Waymo taxi's are always in geofenced areas with full HD maps down to the centimeter. I can't take it to tahoe like i can with my Tesla. my point is... my tesla drives itself... all the time and never have to disengage. People can shout technicalities all the time, and regulations where tesla fails because it "Doesn't have an operator that can take over" but for all intents and purposes, my car drives itself from my door to tahoe and back without me having to take over a single time.
If that's not full self driving, but waymo's geofenced, $45 for a ride down the street is, we just disagree.
What you mean is for your intents and purposes. Others in the thread have pointed out specific intents and purposes for which Tesla's approach fails -- driving when you don't have to pay attention. Which is the core function of self-driving, so not being able to do it is kind of whole thing.
> If that's not full self driving, but waymo's geofenced, $45 for a ride down the street is, we just disagree.
Geofenced or not, self driving is about not having to pay attention while the vehicle is in motion. If you have to supervise it, that's a very different thing from a system that will all you to looking at your phone or sleep.
Tesla's approach of trying to drive everywhere instead of a geofenced area is part of why their system is failing to deliver self driving.
Trying to do one thing well before expanding the performance envelope is good systems engineering practice. But Tesla has been widely testing their systems on the greater public, which has tragically resulted in deaths. This is why at the end of the day Tesla requires you to supervise their system while you operate it.
Are the Teslas in the Vegas Boring Loop still driven by humans? If so, how is it that Tesla seems unwilling to assume liability for what has to be one of the simplest driving tasks?
If it crashes, it's my fault. At every point i'm supervising.
Except in self summon, and if it side swipes the car on the way out, it's obviously still my fault. That's just never happened to me.
Where in my sentence did i say I wasn't fully in control of the situation? I just say i very, very rarely even have to disengage in situations.
On the very rare occassion that i do disengage, it's not really that the car is going to put me in a life threatening situation, it kinda just stops... and tweeks out a bit. Mainly at some super wierd triangle intersection in some of the small towns along the california coast.
Honestly i've come to "feel" the car after using it. I'll disengage if i even have a shadow of a doubt it's not going to work, and in situations where i've seen it "fail" before. It might have accomplished it, but instead i just drive through the wierd intersection and reengage.
This has already turned into a rant, but one last point; Have you driven in the other cars in Austin? They do the same thing. When it tweaks, or thinks it might tweak, they patch over to a human who takes control of the car.
Why don't you add anywhere to that list? If the car drives itself, why is it geofenced into only places google has HD mapped down to the centimeter. My car can drive itself to Tahoe, can a waymo?
Waymo's strategy is to be extremely cautious and slowly improve the system and increase its scope over time with the goal of establishing self-driving cars as a long-term viable solution. They know they need to increase the trust of many people. Therefore, they geofence to locations where they have an understanding with the local politicians and government, near support facilities, and high quality data.
Tesla chose a different strategy. It's hard to collect enough data to know exactly how safe it is.
That is the difference between "safety first" and "my personal convenience first", which again boils down to insurance liability.
As long as you pay for the people you injure and kill on the way, you can let your Tesla drive you anywhere, even if it can't do it. You can let it try, and maybe it fails. That's totally fine. You will be held liable, but you get to enjoy your trip to Tahoe.
Self-driving companies that have to pay for the systematic faults of their systems will usually move different.
Because Waymo is not stupid enough to take liability for a situation which they aren't extremely confident about. Tesla is also not stupid; they just don't take liability period.
> Disengagements occur when the self-driving system is deactivated with control handed back to humans because of a system failure or a traffic, weather or road situation that required human intervention.
> Waymo, for example, drove 352,545 miles in the state during the period with only 63 disengagements. Cruise vehicles drove about a third less, at 127,516 miles, and had 105 disengagements.
> The third best performance came from Nissan Motor Co, which drove 5,007 miles and had 24 disengagements, meaning that its vehicles had disengagements on average every 208 miles.
Notice that Tesla isn't even included. That's because they don't actually have full self driving tests ongoing like this. Just the half-assed version they beta test with their customers on public roads.
Just curious, but if it really was up to full self driving, why don't you think Tesla would have it certified as such? Being first to market as a true self driving vehicle would be a huge business win.
I don't think it's fair to call a car self-driving if the self-driving disengages itself every time its about to get into a nasty wreck because of its own actions. It's facially "self-driving except for when it's not", the "not" times you, of course, cannot predict.
It wouldn't have survived without them though. State/Federal EV tax credits & carbon credits are government subsidies, and are not a natural product of the "free market".
It speaks volumes tho that one of his clauses was to be treated as a co-founder. He clearly has a chip on his shoulder, and his technical abilities are to be questioned. Ultimately, people choose to believe Elon is smart or a good guy (saving the environment etc.). I think he is a remarkable conman. He was able to convince remarkable engineers to work on these products under the pretext of saving the environment, while paying relatively low wages (a few years ago Tesla pay was not comparable to other car manufacturers despite its California location). Also, I find it really funny that he was illegal in the US for a few months, and currently opposes immigration the way he does. Regarldess of your views on these matters, any man that takes a piece of the pie (subsidies, immigration etc.) and denies it to the next person is either a conman or a hypocrite, or both. I think he is a conman, who knows the next person will resort to tactics similar to his, and wants to avoid it. He never cared about the environment, given he is doing everything in his power to kill other EVs and maximize his own profits. There are many skeletons in the Musk and Thiel closets, enough to fill a graveyard. Hypocrisy is truly seen as a virtue in DC
Uh. California began offering $5,000 rebates for electric vehicles in 2007, before Tesla sold its first car. It was literally a selling point touted by Tesla.
From Tesla's blog (2009): Tax incentives: Why the Roadster costs less than its sticker price
It's ironic that the much more significant ultimate success of deep learning happened despite a lack of government funding, if Hinton is to be believed. The 90s were a neural net winter, and success required faster computation, a private success.
I lose zero sleep at the prospect that there would be zero government robotics research funding. If the advantages are there, profit seekers will find a way. We must stop demonizing private accumulations of capital, "ending" billionaires and "monopolies" that are offering more things at lower cost. Small enterprises cannot afford a Bell Labs, a Watson Research, a Deep Mind, a Xerox PARC, etc.
Hinton and his students studied for years on US (and then Canadian) government grants. The year Alexnet came out, Nvidia was awarded tens of millions by DARPA for Project Osprey.
It's an odd historical revisionism where from Fairchild to the Internet to the web to AI, government grants and government spending are washed out of the picture. The government funded AI research for decades.
I think their point is the billions in private investment which preceded those millions.
I think this is a common issue in computer science, where credit is given to sexy "software applications" like AI when the real advances were in the hardware that enabled them, which everyone just views as an uninteresting commodity.
> I think their point is the billions in private investment which preceded those millions.
But the "billions" didn't precede the "millions". They're just completely incorrect, and anyone that knows even a tiny amount about the actual history can see it immediately. That's why these comment sections are so polarized. It's a bunch of people vibe commenting vs people that have spent even like an hour researching the industry.
The history of semiconductor enterprise in the US is just a bunch of private companies lobbying the government for contracts, grants, and legal/trade protections. All of them would've folded at several different points without military contracts or government research grants. Read Chip War.
You seem to be arguing that the second government touches anything then everything it does gets credited to the government funding column. Seems simplistic to me, but you can believe what you like. Go back far enough and there was only private industry, and no government funding until the space race basically.
Either way the fact remains that the billions spent developing GPU's preceded the millions spent to use those GPUs for AI. Not sure what it has to do with polarization of the comment section. I assume it's just people seeking an opportunity to heap abuse on anything close to a representative of the evil "other side".
> Go back far enough and there was only private industry, and no government funding until the space race basically.
How do you think the railroads were built in the US? The bonds of the Pacific Railroad Acts date back to the 1860s. Pretty easy to build a railway line when government foots the bill.
Government funding of research. We were talking about the NSF after all, not free markets versus central planning.
On that though, I read somewhere that the hierarchical committee-led operation of the funding agencies is the same way communist systems dole out money for everything else too. Not sure if they were being completely serious.
From 1901 up to FDR's election in 1932, 5 Americans won Nobel Prizes in the sciences. There was not much government funding back then, and not much was going on either.
So your argument is that nothing is communism? The fact that it's a single large organization allocating resources is rather key to the whole point. That the same organizational structure doing it is interesting to me anyway. I suspected this line of thinking is too triggering for some people though.
A corporation is not an economic system, just a tiny participant of one. And I'd rather describe their decision making as hierarchical yes, but by middle managers implementing the agendas of higher ups, not necessarily by committees. When they operate by committee they tend to be at their worst...
Many industries are uninvestable in their early days. How many get to the point where private funding makes sense without initial government funding for fundamental science and research? Where will we be in 15 years if the government starts pulling funding like the NSF? We might find the private money at that time is funding those future industries in other countries instead.
Seeing all the recent tariff fights and actually finding out what the story is behind some of the different industries, I am becoming much more of the opinion that other countries take over industries as the result of specific agendas targeting those industries and maintaining a large degree of monopoly over them. The US has not reacted much because each country only took one industry or so and it was a way to manipulate them or appease them or whatever, but it is turning into death by a thousand cuts. I definitely think the US government needs to be a lot more involved than they have been in a range of ways. That list of ridiculous-sounding cancelled NSF grants wasn't it though. If you're talking about the SBIR program, that is pretty tiny. I assume it will continue, it is legally set to be at 2% or whatever.
> You seem to be arguing that the second government touches anything then everything it does gets credited to the government funding column.
Absolutely not. This is an obvious bad faith interpretation of my comment.
> Either way the fact remains that the billions spent developing GPU's preceded the millions spent to use those GPUs for AI.
Again, you're just obviously completely factually wrong to anyone who has even a modicum of casual interest in the history of these technologies.
> Not sure what it has to do with polarization of the comment section. I assume it's just people seeking an opportunity to heap abuse on anything close to a representative of the evil "other side".
And one more time for the people in the back. Anyone with any amount of actual knowledge on the topic at hand can immediately dismiss your entire argument because it isn't based in anything resembling fact. It's just you wishing or hoping that it might be somewhere close to true. This is just that scene from Billy Madison: "Everyone in this room is now dumber for having listened to it. I award you no points, and may God have mercy on your soul."
I wonder if it deals more with the approachability of software applications. If I even begin to think I’d compete with NVIDIA delivering similar hardware, I’d very quickly realize I was an idiot. Meanwhile as a single individual, there is still a reasonable amount of commercial markets of software I really do have some chance at tackling or competing against. As software complexity rises it’s becoming far less tractable than it was in say the 90s but there are still areas individuals and small sums of capital can enter. I think that makes the sector alluring in general.
Hardware is just in general capital intensive, not even including all the intellectual capital needed. So it’s not that it’s uninteresting or even a commodity to me, it’s just a stone wall that whatever is there is there and that’s it in my mind.
That difference in difficulties is kind of the point. Imagine, as an extreme, a company makes a machine with certain functions performed based on which button combinations you press. A second company gets a patent for using the first company's machine for doing various tasks by pressing various button combinations, which are new uses of the machine no one had thought of yet. Now the second company has all the bargaining power in the market and so gets giant margins, despite doing a tiny fraction of the work it takes to make those tasks possible.
I wonder if our current system ended up this way because it is the most efficient in terms of specialization, or because the patent system drove things in this direction where the people last dealing with customers (i.e., those making the software layer) have the best info of what tasks the customers want to do with their computers, and hence patent the solutions first. Leaving hardware vendors no choice but to serve the software monopolies (one after another since the 80's).
You are suggesting unilateral disarmament. Allowing other nations, not all of them friendly, to take the lead in science and technology as they continue to fund their own research and poach our best and brightest.
Once something has a predictable ROI (can be productized and sold), profit seekers will find a way. The role of publicly funded research is to get ideas that are not immediately profitable to the stage that investors can take over. Publicly funded research also supports investor-funded R&D by educating their future work force.
The provided examples do not clearly support the idea that industry can compensate for a decrease in government-funded basic research. Bell Labs was the product of government action (antitrust enforcement), not a voluntary creation. The others are R&D (product development) organizations, not research organizations. Of those listed, Xerox PARC is the most significant, but from the profit-seeking perspective it's more of a cautionary tale since it primarily benefited Xerox's competitors. And Hinton seems to have received government support; his backpropagation paper at least credits ONR. As I understand it, the overall deep learning story is that basic research, including government-funded research, laid theoretical groundwork that capital investment was later able to scale commercially once video games drove development of the necessary hardware.
Bill Wadge ruined my life! I stumbled upon his book on his language Lucid in the library when I was in grad school, and it opened up my eyes about different programming models. It inspired me to try making my own programming language and I’ve been doing it ever since!
Going on 10 years now, no sign of losing interest. Don’t read his book if you have an interest in the thing you’re doing; it’ll turn you into a compiler dev and you’ll never do your interest again.
Lucid was a strong inspiration for my programming language Intonal, particularly the “fby” operator, which took over ~5 years of my life. Probably going to open source it soon
This is really cool! What do you mean by "the" compiler for our chip? Is this a compiler backed like LLVM which generates efficient machine code, that other languages can target? I'm an author of a dataflow programming language, what kind of resources do you have that I could read about targeting your hardware? I've been waiting for hardware to catch up with the language I'm building, so that's why I'm interested!
> What do you mean by "the" compiler for our chip?
We are designing a dataflow general-purpose processor that can directly executes the dataflow graph from the compiler.
> Is this a compiler backed like LLVM which generates efficient machine code, that other languages can target?
The compiler itself takes LLVM as the input, so if your framework can output LLVM, you should be able to target our compiler backend and hardware. We had a prototype working with Rust as well. The compiler does not however use LLVM to produce the machine code on our hardware -- custom mapper/assembler and linker is used to produce the machine code that runs on the hardware.
> what kind of resources do you have that I could read about targeting your hardware?
We have published a couple papers about our compiler and hardware design: https://www.efficient.computer/media. Unfortunately we don't have a concrete plan to open up the compiler yet. Please stay tuned!
It's bimodal. Most of the world uses one paradigm of programming (declarative programming via Excel and SQL), while developers use another paradigm (imperative programming via Python/C/C++/Javascript et al.).
How do you deal with writing code with multicursors when you have to type the same thing multiple times? With monospace I just ctrl+alt+down a couple times on aligned text and then type. With proportional fonts I don't suppose it's easy to align text exactly, so do you just not use multicursors or is there a solution you came up with that works?
I don't use multicursors much, not every year. I'm using emacs. I register a sequence of keys and apply it multiple times with control e, control e, control e etc.
I’ve been a professional programmer for 17 years and have never used multicursors. I don’t even fathom under what conditions you’d want to. I use Find and Replace.
Everyone programs a little differently. I often use it when e.g. using intrinsics and I want to change types. Find and replace isn't especially helpful when they're different names with substructure you need to modify locally.
reply