komaromy

For ETH, that's not really true: you can use it to pay for computation on the Ethereum blockchain. The value you assign to that computation may not be very significant, but it's hard to justify the notion that it's zero.

What % of all ETH is currently held in ICO accounts?

The best way to do this is probably to force the customer to specify a refund address at time of purchase, which is what ShapeShift does.

> 2/3's of the total amount was owned by Ethdevs

Source?

It could be that post-series-B companies are less attractive acquisition targets. I don't have any data on this.

He'd also be the true inventor of Git then. It's a widely useful data structure.

> but perhaps the most anti-leftist aspect of them all is how the show is utterly grim and hopeless and nothing ever changes. all radical initiatives are shut down and ruin lives, and it works more as a cautionary tale than an inspiration.

I think the Hamsterdam concept would be pretty easily implemented at the federal level, which was where it was ultimately shut down from in the show.

It's unfortunate that our federal drug policy is now moving in the exact opposite direction.

vs simulation, it lets you get trajectories from the way people actually move without having to make insanely complicated models. vs. precollected training data, maybe the recording apparatus is less expensive/complex?

Not just strong; it hasn't failed to place at least second in any TCEC season (the de facto computer chess world championship) since 2013 [0].

[0] https://en.wikipedia.org/wiki/Top_Chess_Engine_Championship#...

Good. I hope they're looking into a pipeline that reaches these students before they're junior- or senior-aged as well.

Yes, probability is helpful to me in everyday life.

I'm ~2200 and I wouldn't say it helps me much in everyday life.

Use of Weapons - Iain M. Banks

Looks really cool.

I recently hit a roadblock when trying to implement the original DeepMind Atari algorithm [0] with TensorFlow. They don't mention this in the paper, but the network wasn't trained to convergence at each training step (maybe this would be obvious to people more well-versed in deep learning, but it wasn't to me coming from a classical RL background).

As it turns out, TensorFlow's optimizers don't have a way to manually terminate training before convergence. That meant I was getting through several orders of magnitude fewer training steps than the DeepMind team did, even when accounting for my inferior hardware. This might not be a problem in some learning cases, where training more on certain examples lets you extract more information from them, but in games with sparse rewards it's bad.

Of course, TensorFlow does let you do the gradient calculations and updates by hand, but I wasn't prepared to go that far at the time. Maybe in the next few weeks I'll dive back into it.

[0] https://arxiv.org/pdf/1312.5602.pdf

I'm not sure about the quality for standard, but the overall chat community is more like modern-day Reddit than 00s ICC. That place was a garden.

> And there are very few establishments that are designed exclusively for the wealthy.

This is an excellent description. Some people still flaunt their wealth, which is true everywhere, but I feel a sense of the city belonging to everyone that lives in it.

Of course, that's my view as a fairly well-off white male who's never experienced anything close to homelessness. People from different circumstances may have different opinions.

Slavery never struck me as something that one believes in.

The NRA's past is grounded in the interests of gun manufacturers. There's no equivalent industry with an existential interest in selling as much free speech as possible. Media orgs would only care if an appreciable portion of what they did was actual journalism.

I'm not too familiar with Go, but in chess we have millions of text-format games and very little real-time footage.

> Almost every good place pays their CS PhD students enough to squeak by without getting into debt. I wish this was more widely known.

Right. The debt problems that people have after PhDs are more often due to their undergrad loans sitting around for 4-7 years while they were earning enough to subsist and not more.

ICC, unfortunately, was built around a desktop CLI, BlitzIn, that really suffers from the transition to mobile. BlitzIn can do both of those things but I have no idea how you would accomplish them through the app.

I'll always remember BlitzIn fondly (it was the first CLI I used, before I even knew what that was) but ICC's glory days were the late 90s and early 00s.

> The fact that you can't run Python 2 and Python 3 at the same time

I'm confused by this statement. I can start a python2 interpreter, background it, then start a python3 interpreter. Is there some other area that they inhibit each other in?

I thought about it over a few hours last night and decided against it.

In the worst scenarios, it won't matter what country I'm in as long as it's still on this planet.

In the merely really-bad scenarios, people that look like me won't be the ones in trouble and I feel obliged to stay and help those that are.

My mostly-uneducated o/u on the time to beat the best human players is 8 years.

If you're considering Linode, you should read up on their history of security incidents and make sure it's a risk you're willing to take.

> almost-free VPS service

I guess, if most VPS services cost >$5/minute: https://www.reddit.com/r/ethtrader/comments/53xt58/daily_dis...

There are two widely-used implementations, and more that don't have as much adoption: https://github.com/ethereum/go-ethereum https://github.com/ethcore/parity

That's why the proposed solution is 'Sure, we'll do the time-wasting thing, but it's going to cost you more to pay us to do it'.

Fix is now merged into the development go-ethereum branch: https://github.com/ethereum/go-ethereum/pull/3006#pullreques...

Excellent turnaround time, especially considering that most of the principal developers were asleep in Shanghai when the disruption started.

> They are basically saying it's useless to them so why not let somebody else give a try.

Their standard reason is that most products use tons of internal API calls/private modules and it would take a lot of hours to replace those with code they're willing to release.