[api]

"fixing government, curing cancer, improving medicine, rebooting our space program, fixing education, etc."

I think a big part of it is less pay. Important things pay very little compared to frivolous metoo.io apps and social advertising and whatever.

Capitalism is a greedy optimizer. When it finds a local maximum it goes all in. It does not explore the fitness landscape.

[msandford]

I think another part of the problem is that there aren't really any important things in the public sector right now. The last public sector thing that engineers got excited about was the Apollo program.

Fixing government? Yeah you might be able to modernize a lot of the tech stack that runs various governmental entities but you're not going to actually fix the entities. You'll be implementing failed or broken policies in node.js instead of cobol. It's a step in the right direction but it doesn't actually fix the core of what's broken.

Curing cancer? There's no Federal Department of Stopping Cancer, either you work for a big pharmaceutical company or you get a PhD and scramble for grant money.

Rebooting our space program? That happens every administration and it always starts with some kind of dream to do audacious thing X that won't get any significant funding.

Fixing education? I'm not really sure where to start here. You can't fix from within since there's a constant struggle by the administrators to stay in power because teachers don't have any upwards mobility besides administration. And trying to fix it from the outside? That's even tougher, Kahn Academy has some traction with people but little with teachers who fear it'll put them out of work.

Find me something inspiring that's not going to be mired in politics and bullshit and I'm all ears. I don't think any of those are all that inspiring. I'd even deal with less pay for something really impressive.

[jmalicki]

>Curing cancer? There's no Federal Department of Stopping >Cancer, either you work for a big pharmaceutical company or >you get a PhD and scramble for grant money.

Actually, there is a Federal Department of Stopping Cancer - the National Cancer Institute.

http://www.cancer.gov/

[msandford]

Point taken.

But it's got a $5b a year budget which is pretty small in the biotech world. It's even worse when you consider that 42% of it is grant money. So my statement isn't quite right, but in my opinion, still close to right. Maybe 70% true.

http://www.cancer.gov/aboutnci/budget_planning_leg/plan-2013...

http://obf.cancer.gov/financial/factbook.htm

Inflation adjusted NASA's budget during the Apollo program was $30 billion a year or so. And that was in a realm that's better understood.

I'm not necessarily suggesting that we should start spending $50 billion or $100 billion a year on cancer research. But I would argue that it might take that kind of investment to really move the needle in a meaningful way and provide enough good paying opportunities for nerds to get them interested.

Thanks for the link though, even if it makes me look a little dumb.

[wtracy]

<<Capitalism is a greedy optimizer. When it finds a local maximum it goes all in. It does not explore the fitness landscape.>>

Upvoted specifically for this line. That strikes me as a key insight into the free market's strengths and weaknesses.

[api]

I've thought this for a long time. Ultimately I think it's that private enterprises must have a low appetite for risk. The only time you see private enterprises investing in fundamental R&D or otherwise really exploring, it's when they are either so huge that they have immense cash hoards (e.g. Google) or they are monopolies with guaranteed cash cow income streams (e.g. the old Bell Labs). The vast majority of private enterprises cannot risk investing in anything without a very high probability of return, and that means being conservative and doing things that have already been proven to work. (Or that are only a slight deviation from things that work, or that build on already proven tech, etc.)

Governments have immense disadvantages too. They are slow and inefficient and are absolutely horrible at taking things the "last mile" from lab prototype to product. Governments flat out suck at "shipping." Yet they have the ultimate guaranteed income stream -- taxes -- and so can afford to invest in "high risk, high payoff" R&D and basic science that might never produce direct payoff (but may produce large indirect payoffs later).

The optimum seems to be, as with many things in nature, a balance. There may also be third options that have yet to be discovered, such as innovations in finance that enable risk to be spread more intelligently.

[sjg007]

"fixing government, curing cancer, improving medicine, rebooting our space program, fixing education, etc."

-- There are a bunch of startups already in these areas...

[eli_gottlieb]

It's also an Unfriendly optimizer. Capitalism's local maxima often consist of things like Gilded Ages and Great Depressions in which return on investment is maximized to the hilt at the expense of nearly all other values. Yes, once we force it out of those local maxima, it goes back to generating wealth in a way that people actually want it to, but eventually, without deliberate management to keep it out of them, it always finds another damn local maximum to get stuck in.