Comment by rayiner | Hacker News Reader

Comment by rayiner | Hacker News Reader

"That's fine, that's capitalism - and these incremental improvements lead to slow productivity gains that at least quicken the pulse of economists. But maybe let's drop the pretense that we're curing cancer unless, you know, we're curing cancer."

Reasonable and astute point.

"An Economist article this year exploring the state of innovation noted that centuries of scientific progress means it takes ever longer for people to reach to the frontier of any field, much less push past it. It also stressed that much of the technological low-hanging fruit is long gone."

Also true, and oft-overlooked.

"Thiel is at least putting his money where his mouth is at Founders Fund, his venture capital firm making investments into robotics, artificial intelligence and biotechnology (alongside a lineup of dot-coms)."

Ironically, AI was jump-started by DOD dollars,[1] and biotechnology by NIH dollars. Robotics had substantial manufacturing implications and got money from that area, but has also benefited heavily from defense money (think UAV's, etc).

[1] http://en.wikipedia.org/wiki/AI_winter#DARPA.27s_funding_cut...

It would seem that capitalism has done a great job solving problems where value-creation yield a sustainable level of value-capture: farming, manufacturing, etc.

However, there's a lot things where there's a ton of value-creation without enough capture. This is where collective action problems must be solved. This could be a consortia like an IEEE standard, gov't funding like the NSF, or something like a city. Many options, but none purely capitalist in the traditional sense of the word.

The article makes that exact point:

> Only governments had the long view and available cash to fund things like supercolliders, deep space programs and - oh yeah - the development of the Internet.

Thats not true. There was no long view government plan to create what we now call the Internet.

The whole point of blue-sky R&D funding is that you can't plan exactly what you're going to get out of it. Today DARPA is a bit more focused on practical application than it used to be, but it's ethos is still: "let's throw $10-30 million at a bunch of smart people and see what happens." The "plan" is to fund fundamental R&D and hope something really useful comes out of it. In that sense, the development of the internet was a fulfillment of that plan.

Exactly! Nearly any concerted focus on a difficult engineering problem (ex: NASA) will yield positive externalities in the form of engineering improvement (manufacturing methods, chip design, solar, etc.).

If your engineering challenge is putting filters on pictures, you're probably going to have less positive spillovers - not that you wouldn't be solving a real problem, but it's a different kind of thing.

I haven't done any kind of scientific survey, but it seems the best government projects are those without any kind of expectation of anyone profiting or making the money back.

Not that I don't like profit motives. I love them. But if there is a profit motive, you have to really really question why private industry isn't doing it instead.

DARPA is amazing, at least for their history and what has come from them.

Dies any other country have a DARPA equivalent? Or, more specifically, did any other country have a DARPA equivalent 40+ years ago?

I'm sure Russia would at least have something like it.

DARPA is a very interesting agency. It's got a $3.2 billion budget, but less than 250 employees and more than half are technical people. They're structured to maximize the amount of money that goes to the actual engineering teams and minimize internal inertia. The program managers get $20 million or so to throw around each year, and have tremendous authority over their projects and not a lot of oversight. They're on the younger side for people with that kind of budget authority and usually have PhD's themselves. They're limited to 4-5 year stints, so it's not a position that attracts people looking to make a career in bureaucracy.

ARPA itself was a direct response to the soviet's ability to allocate assets to state innovation like sputnik.

Maybe if government wasn't so massively wasteful, there'd be a few dollars to throw at blue-sky R&D funding these days, but there isn't.

"blue-sky R&D funding" is the definition of "massive waste" to many people. To you, it's probably feeding poor children or something else.

But there were government plans to create telecom networks that ultimately led to the internet.

While there was a ton of amazing stuff going on inside Bell Labs, it really seems like progress in telecom didn't accelerate until after the ATT breakup. If you think the cell phone situation now is bad, you used to have to rent your home phone, like a cable box, from the phone company.

dreamdu5t you are so wrong that you deserve to be onvoted, but I dont do that. The basic IP technology was all government funded and developed by BB&N. The original internet was ARPAnet, funded by the government to link the government, military, universities, and military contractors (.gov, .mil, .edu, and .com). The Internet itself was set up separate from DARPAnet by government legislation and funding. The whole domain registration mechanism was at one time under the ultimate control of the Commerce Department. The world wide web protocol was developed at CERN, funded by multiple governments. The browser that became Netscape was developed at the University of Illinois. It was only after the Internet was set by the government did we get the explosion of users and applications. The competitors to the the internet developed by private corporations, like MCI mail or Prodigy, are all dead now.

When you're spending billions/trillions of tax dollars, you're going to have successes here and there.

There's no question that the government has some successes to point to. I actually agree that government is decent at some basic research.

Then again, the private sector has had plenty of successes as well.

My main objection is that we don't really understand the opportunity cost of having the government spend the money vs letting the private economy do so itself - yet we let the government continue to make the decisions on what's fair.

Can you elaborate why that is ironic? Most of the basic science is funded by governments (throughout the world, not just the US). Isn't that supposed to be the role of government (funding research that has no immediate financial prospects)?

Because Thiel is famous for his libertarian views. Though having heard him discuss his thoughts on the roles of government and private enterprise in person, I was surprised to find that he's much less ideological than he is commonly portrayed as; he expressed support for a moderate amount of regulation but preferred that it be more personal and accountable than (ie based on the judgment of experienced civil servants) than legalistic and systematized.

I think he has a good point. Japan and Europe have more 'hands on' regulatory regimes and even though regulations are stereotypically more onerous in Europe, it's arguably more results-driven and less of a procedural minefield. Of course the downside of this is that there aren't as many avenues of appeal and if the judgement of a regulatory authority goes against you then you may have to just grin and bear it; that's problematic for startups which don't have a history of successful commercial operations behind them and may find compliance more difficult and expensive.

Reply on news.ycombinator.com