There's already a bounty out for precisely that www.youtube.com/watch?v=DaWp...

It is. The error is roughly 7E-16 or 7*10^(-16). The negative power shows that the number is really small. 10^(-x) is the same as 1/(10^x), so in this case it's like dividing by 10^16. 7E-16 = 0.0000000000000007

Found it: www.sciencedirect.com/science/arti...

In the UK the wholesale price of electricity is set by the most expensive method of production - not the actual cost of production. Renewable energy *is* cheaper to produce - but we're charged as if it were produced by burning gas.

See: commonslibrary.parliament.uk/why-is-cheap...

It's not really going to be possible to keep bots out long-term if it's possible for humans to sign up without verification of their humanity - and even if there was verification that the account was being created by a human, what's to say the human isn't creating the account on behalf on an AI? 6/6

But eventually CAPTCHAs will become so hard that many people will give up, locking legitimate users out. Bots could also exploit humans to get access - I read in a paper about a bot which reasoned that it could pretend to be a visually-impaired person on fiverr to get a person to solve CAPTCHAS. 5/n

at which point a bot masquerading as a person can copy it. AI-powered bots are only going to become more and more able to act like humans as the technology develops, so technologies like CAPTCHAs will have to keep getting harder to be effective at keeping bots out. 4/n

The content could be sent over https to prevent interception, and requests could be tied to an individual account via a session cookie.

Fundamentally I don't think it's possible to completely bot-proof sites like this because at some point the content has to be visible to legitimate users, 3/n

They'd only need to download each piece of content once since once it's decrypted on their end they can make copies and send them to a central repository free of any restrictions forever. Also I'm not quite sure what the purpose of encrypting the data with the requesting user's public key is? 2/n

It's a nice idea but wouldn't deter a sufficiently-motivated company from scraping the data. They could just spin up thousands of bots with different IPs, randomised user agents, randomised delays between requests etc., make accounts for them and have each scrape a modest amount of data per day. 1/n

Sure you can! en.wikipedia.org/wiki/5D_Ches...

Definitely fake. Here's the original: www.gettyimages.co.uk/detail/news-...

I've been learning a bit of CSS to create personalised dark modes on websites that I frequent

There would have been people well over 100m above ground level thousands of years ago while they were topping Khufu's pyramid.

I think the graph counts people standing on fixed structures as being on the ground. We've had multi-storey buildings for thousands of years so clearly there would always be people >10m above ground level for all of this graph.

The enmity graph is bipartite. [Exercise for the reader: prove that the graph is a complete bipartite graph.]