Hacker News Reader

Comment by Animats | Hacker News Reader

Now the high-frequency trading people will have to build accelerators and detectors so they can transmit in a straight line through the earth. They could knock off an entire millisecond or two between London and New York and beat everybody else.

View on HN

It gets proposed fairly often. The tricky part is detecting enough neutrinos in a coincidence window to make a meaningful trading signal.

Such an experiment is also very expensive.

Expensive for HFTs?

Expensive enough to not be profitable

Thankfully!

A $300 million cable is being laid across the Atlantic just to cut off 5ms for high-frequency trading.[1] Bandwidth will cost 50x normal on that cable.

[1] http://www.telegraph.co.uk/technology/news/8753784/The-300m-...

Spread Networks built a $300m line between Chicago and NJ just to save 3 milliseconds. That project involved building a tunnel.

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

http://www.nytimes.com/2014/04/14/opinion/krugman-three-expe...

Neutrino experiments cost of order a billion dollars. The event rate for anthropogenic neutrinos sources at the MINOS detector (from the article) is a few hundred events over 1 year.

https://arxiv.org/pdf/hep-ex/0607088v2.pdf

So the bandwidth for this state-of-the-art billionish-dollar detector is measured in μbps :)

So you're saying there's a chance? /s

Flipped over, if you can invent a reliable neutrino detector and influencer that costs say $150M you have a very risk tolerant market to sell into.

Don't give them any ideas! We've seen how the last 50 yrs worked out! /s

Reply on news.ycombinator.com