yeah, a single M1 Abrams tank costs over $4 million, and we have 9000 of them.

DARPA R&D money always feels like money better spent than a lot of other military spending. We get future dividends, often civilian.

Absolutely this. With the culling of most corporate R&D groups over the last 20-30 years, DARPA pie-in-the-sky research projects seem to be one of the last real sources for investigating anything that can't be immediately and quickly monetized.

How much real research, as opposed to early-stage product development, has ever happened at corporations? We can easily name a few specific examples (Bell Labs, Xerox PARC) presumably because they weren't the norm. (And there were doubtless some others at companies that weren't as aligned with what's thought of as today's tech industry such as DuPont.)

You might enjoy the book "The Entrepreneurial State" by Mariana Mazzucato makes a similar argument that a lot of innovation is actually driven by the state, because they can afford to invest a lot in research without an immediate payoff. The author mentions DARPA in particular as a great example of the US stimulating innovation that then later benefits companies. For example, while the iPhone is a great piece of innovation, it built on innovations made possible by state funded research.

There's also a very good book by Bill Janeway (of Wargburg Pincus notoriety) on the same subject called 'Doing Capitalism in the Innovation Economy: Reconfiguring the Three-Player Game between Markets, Speculators and the State'.

He argues the same point about the states roll in RnD (also explicitly calls out DARPA) and also their role in booms and busts.

There’s some, obviously, especially in companies in chemicals and materials. I think the push in the 80s-90s to remove slack from corporations to reward managers and shareholders killed a lot of basic research they were doing for posterity.

Have a look at MSR.

What about them? They make nice gear. They've done some nice innovation around some of the gear they make. That seems different from pumping multi-millions of dollars into research that may never have direct commercial value--unless you're talking about a different MSR.

Even before then, Wikipedia claims government was, at least in part, responsible for the soda can[1].

https://en.wikipedia.org/wiki/Defense_Production_Act_of_1950...

[1] as far as I've seen, soviet vending machines used a shared glass. They must've been very confident both (a) that the glasses wouldn't grow legs, and (b) in their public health initiatives.

https://www.youtube.com/watch?v=F68bbzOOOdY&t=2644

Fusion researchers will read that and feel like chopped liver.

I dunno, when you consider the design intent of the Abrams (operating environment, crew protection, etc.) 4 million seems like a reasonable price tag to me.

Whether the US Army needs 9000 tanks is a separate discussion though.

US DoD had as a strategic goal the ability to fight and win a two theater global conflict against peer or near peer nations up until a few years ago, when it was changed to fight and win one theater, while not losing the other (holding ground until the other theater is won).

If you're looking at needing to fight China and Russia, at the same time, over an extended period of time (2+ years), having 9,000 tanks in your back pocket seems like a good idea.

Also, it's a jobs/pork barrel program. Congress keeps ordering tanks that the Army puts into storage because the M1 is a beast of a machine that keeps a lot of people employed. The US manufacturing base is pretty depleted, and ordering tanks you don't necessarily need is a good way to keep the skills and tool chains you could need in the future around.

>Also, it's a jobs/pork barrel program. Congress keeps ordering tanks that the Army puts into storage because the M1 is a beast of a machine that keeps a lot of people employed. The US manufacturing base is pretty depleted, and ordering tanks you don't necessarily need is a good way to keep the skills and tool chains you could need in the future around.

Sure, that's nothing new to anyone who's familiar with Cold War economics. The Soviet Union was able to keep its economy crawling along without sanctioning free enterprise by making enormous investments into collectively-owned (i.e. state-owned) products like nuclear submarines and combat aircraft. Turns out you can achieve much the same goal but also benefit the people if you put that effort into state-funded infrastructure. Whether the US body politic will support reducing M1 Abrams production in favour of fixing the USA's crumbling infrastructure is again a separate discussion.

A little numbers correction: it's more like $11 MM per tank. And not to build, to refurbish.

More precisely: for the 2020 fiscal year, the US Army did not procure any new M1 Abrams tank, and is not budgeting any for the 2021 year either. Instead it is modernizing existing tanks at a cost of $1.7 BN for 165 units in '20, and $1 BN for 89 units in '21. This comes at about $10.6 MM per tank.

[1] https://www.asafm.army.mil/Portals/72/Documents/BudgetMateri...

[2] https://www.asafm.army.mil/Portals/72/Documents/BudgetMateri...

That's the choice of your fellow citizens who'd rather spend money on weapons than feed people, and will come up with elaborate stories about why your uncle deserves to be in that position. :(

The other side of this is we spend a lot of money feeding people in the US. Unfortunately, it’s not an efficient system consisting of farm subsidies, food stamps, school lunch programs etc. If the goal was simply adequate nutrition for every American we could easily have that with current budgets. But, instead money must also end up in the correct people’s pockets while also feeding people.

," he said, on an ARPA-funded network.

"My uncle is starving while defence projects get lots of money" and "Some defence projects have led to good outcomes for society".

Without aiming to insult here¹, is it interesting to anyone that the same things we say when we're 14-15 years old are the same things we say decades later? I remember having had this argument when I was a kid (and very space-obsessed) and it was almost exactly like this. I remember my classroom, even, and it was between Physics and English and the teacher hadn't come in yet.

It's like some number of people have to be thrown into the war of the ideas periodically to preserve the battle lines exactly as they are. Stop and you lose ground.

¹ like not saying I'm better than you guys, I could just as well do what you're doing.

> It's like some number of people have to be thrown into the war of the ideas periodically to preserve the battle lines exactly as they are. Stop and you lose ground.

Well, yes. Overton window and all that. Similar discussions on USENET 20 years ago. Although I think in some ways they've moved on, and in some ways regressed - the openness of troll conservativism, for example.

I sometimes think of these ideological fault lines as behaving like the geological ones; pressure maintained for years in equilibrium, then there is a crisis and a huge slip all at once. With a lot of stuff getting demolished along the way.

I'd be impressed if you could find a single hop in the modern network we're using that's actually run by ARPA. The ISPs, the hosting, the interconnect - all private. Who owns the transatlantic link?

Calling the internet an ARPA system is like insisting on calling telephones the Bell system.

Yes, but it’s exactly what DARPA generally does. They fund people to propose and prototype. I remember my Dad trying to get grants in the SBIR program. In phase I, they’re funding projects that are merely proposals for something they could eventually fund. Things may graduate from ARPA and go through the normal procurement process or they might be shelved.

The government and military are very large consumers of internet hardware, software and services, all of which would look very different if they waited on the private sector. (Probably something specified by ITU).

I feel for you, but military spending is a massive federal jobs program. Each job created has a job multiplier effect, creating additional jobs in the surrounding area. Manufacturing jobs producing military equipment have a much bigger job multiplier effect. So if you were to wave a magic wand and half military spending overnight, the economic impact would be disastrous, and many more people would be in dire economic situations.

That doesn't seem to be an issue for other Western countries with much smaller militaries (per capita).

We have the US for that military thing.

I find it interesting that this argument is only ever being made when there is talk about cutting military spending. Somehow it is never the problem that there is job losses when other spending (in particular social spending) is being cut.

There was actually a great response by a member of the German Labour Party (SPD) in the discussion about the spending on the Eurofighter project (sometime in the 90s IIRC). His response to the jobs argument was: We could also build a pyramid in honour of Helmut Kohl and it would create a lot of jobs, that doesn't mean it's a good idea.

Yeah it’s a ridiculous argument. The money being spent on those jobs is taken by force (which has an unmeasurable chilling effect on the economy), and then allocated according to central planning, inevitably in a way far less efficient than the market would.

So what if we used military construction money on schools and education? Or maybe capturing carbon? Plenty of useful things to the world we could spend the money on. No need to just cut spending and bank the money.

My comment was in reply to the GP's statement that it was a waste of money. When the federal government gives out money, it has a ripple effect as that money is spent on goods and services. Creating a service-based job will have a larger ripple effect, and creating a manufacturing job generally has the largest ripple effect.

So if you have a magic wand or just happen to become dictator, please keep in mind the multiplier effects when deciding where to spend the money. ;-)

Couldn't we use that money to create jobs that have a more meaningful impact on people's day to day? Invest in alternative energy, fixing our infrastructure, improving public transit, etc... Put that investment in technology into technology that is more productive instead of the nth iteration of some piece of military technology that we'll never use.

I want that sweet high-speed rail, baby!

That would cool, but it would be a multi-year, probably multi-decade process. Not overnight.

Reread the comment you are replying to (especially the context of "than") and consider the scale of the resources invested in the military.

DARPA accounts for a pretty small part of the military budget and is probably not the reason for poverty.

>>DARPA R&D money always feels like money better spent than a lot of other military spending.

I think "better spent than a lot of other military spending" was the point of the post you responded to. The military may be wasting The People's Money, but DARPA seems to have low costs / high return comparatively, and often has high returns for civilians/society.

What return? Dead people? Pollution? Fear?

Some notable ones off the top of my head include the Internet, GPS, computer mice (and basically the foundation of UI that uses them), onion routing, and prosthetic limbs.

Compare DARPA's $3B budget to e.g. the DoD's $650-700B budget. You can argue that either or both should be spending some portion of their budget on "the people" in a myriad of ways, but that argument gets way too close to personal politics for an engagement here to be worthwhile, IMO. Don't forget to vote!

With the exception of prosthetic limbs, so far those technologies have proven to be net negatives for humanity, only accelerating our suicide:

GPS: Nuclear missile guidance system

Internet: distributed nuclear missile command and control system

Computer mice: windows

Onion routing: fbi honeypot, enabled multiple violent insurgencies in northern africa and southwest asia.

And the list of actual successful darpa projects is longer than the one you provided, and the omissions are not nearly as benign as the ones you have cherry-picked.

And that's just what made it out of the lab. How many full ride college scholarships have they left on the cutting room floor? How many insulin shots? How many pounds of fresh food? C'mon man. DARPA makes war, not life.

I don't think those are very good examples. My internet must be defective. I haven't been able to launch a single nuclear missile from it.

Same for GPS-- compare the number of deaths from GPS guided nuclear missiles (0) to the number of lives saved by GPS (through more efficient farming, turn by turn directions, E911, etc)

DARPA has driven a lot of research in self-driving cars, space exploration, internet infrastructure (TOR for example), neural implants and so on. It's budget is also lower than some weapon systems that have never even gone into production.

I dislike the american military-industrial complex, but if I had to keep one thing from it it might be DARPA.

Lasers, fusion research, GPS, the internet...

Talk to the American people who will not vote on a candidate that insists on cutting funding to the military industrial complex (of which DARPA gets a very small percentage anyway) and the American people who hate the "welfare state" and "socialism," both of which would make it possible for your uncle to get assistance from the government via taxes.

Data: DARPA's 2019 funding was $3.556 Billion dollars. DoD budget for 2019 was $616.2 Billion dollars.

Lastly, a lot of DARPA's research actually gets used for projects that benefit the public. Other people already listed a ton of projects like GPS, so I wont do that here.

If your uncle is dumpster diving for food I doubt he pays any taxes after refunds.

Good sarcasm needs an element of absurdity or irony. Pathos and non-sequitur alone aren't usually sufficient.

$5m by 1985, with another $3.5m earmarked [1]. Accounting generously for inflation, that's $25m today [2]. It would be interesting to know what the price would look like if they tried the same thing today.

[1] https://archive.macleans.ca/article/1985/11/4/the-pentagon-f...

[2] https://www.inflationtool.com/us-dollar/1981-to-present-valu...

Mantis by winchymatt perhaps?

A two ton diesel powered hexapod.

https://youtu.be/1h9Mw-s9mzI

It's amazing how much you can achieve with a relatively small team of experts. I also believe people have gone "blind" to how much a million is and what it can afford. I mean a million can employ a decent software developer for a decade, and a solo developer can achieve a lot in a decade. I've probably produced about a million LOC in that period in half a dozen languages.

I was curious; the Manhattan Project cost $2 billion (in 1939) and employed 130,000 people (https://en.wikipedia.org/wiki/Manhattan_Project, first paragraph).

Colin Furze could probably do this for less than 50,000£, and it would even spit fire!

I love that guy, and you're right. So many of his projects is him trying find a different way to locomote a flame thrower.

Millions seems about right. If it took more than $999 million I’d be pretty surprised. Hydraulics are tricky but they’re not that tricky.

The kinematics are the tricky part. Trajectory planning is still quite difficult today even with our speedy modern CPUs

Trajectory planning what? Actually solving the inverse kinematics of 3dof manipulators is trivial. I did so 20 years ago using a Basic Stamp microcontroller for my 18 servo hexapod. I was 15.

Planning foot placement is a bit harder but I would say the compute heavy part today would be perception.

Trajectory planning for foot placement.

I think of “trajectory planning” as: you have a known target position and you have to figure out how to get there. That’s not hard with a 3dof manipulator.

But figuring out where to place the feet, which I consider a perception problem, is very hard. But I think that’s what you’re talking about and I just use different terminology. That’s why I was confused. Sorry if I came off wrong.

I don't think the technology behind it was particularly new or novel. It was simply arranging known technology to a new application.

Given the lack of research, but perhaps the high experimentation, millions make sense.

It was new in that it was computer controlled. General Electric had demonstrated a 4-legged walking truck in 1965.[1] Arms and legs were slaved to the operator's arms and legs. It was very hard to drive. Trying to do that with computer control in 1985 was worth trying.

Building the prototype at full scale, not so much. If they'd gone through a few iterations starting with a dog sized version, it might have worked out better. But there was a political problem. DARPA funded a pony-sized legged walker in the late 1960s[2], and it was publicly shamed as a waste of taxpayer money by Sen. Proxmire. So DARPA went for one big enough to carry a real load.

Early thinking about legged locomotion was all about gait and foot placement. It turns out that balance dominates foot placement, and slip control dominates balance once you get off the flat.

Today, most of the problems have been solved, and there's still no market. Which is disappointing.

[1] https://youtu.be/coNO9FpDb6E

[2] http://cyberneticzoo.com/walking-machines/1968-phony-pony-fr...