Hacker News Reader

Comment by perihelions | Hacker News Reader

Fascinating that there's a neutrino beam cutting straight through the earth, in a chord hundreds of miles long, and they expect to be able to measure dozens of events per second (edit: at the closer of the two detectors—not the 800 mile one). If they were to modulate this beam, they could tweet-by-neutrino in near real-time.

- "DUNE is split between two locations hundreds of miles apart: A beam of neutrinos originating at Fermilab, close to Chicago, will pass through a particle detector located on the Fermilab site, then travel 800 miles through the ground to several huge detectors at the Sanford Underground Research Facility (SURF) in South Dakota."

- "The DUNE detector at Fermilab will analyze the neutrino beam close to its origin, where the beam is extremely intense. Collaborators expect this near detector to record about 50 interactions per pulse, which will come every second..."

View on HN

> they could tweet-by-neutrino in near real-time

We demonstrated in 2012 a neutrino-based link which “achieved a decoded data rate of 0.1 bits/sec with a bit error rate of 1% over a distance of 1.035 km, including 240 m of earth” [1].

To put that into perspective, a through-the-core beam would reach its antipode about 24 milliseconds before its circumnavigating partner [2].

[1] https://arxiv.org/abs/1203.2847

[2] assuming a spherical earth with diameter 12 756 km [a] and circumference 40,075 km [b], and thus distance difference of 7 282 km, and a speed of light of 300mm m/s

[a] https://imagine.gsfc.nasa.gov/features/cosmic/earth_info.htm...

[b] https://en.wikipedia.org/wiki/Earth%27s_circumference

The relevant RFC is from (April 1st) 1991 - Memo from the Consortium for Slow Commotion Research (CSCR) https://www.rfc-editor.org/rfc/rfc1217.html

Section 4 deals with Jam-Resistant Underwater Communication

    The ULS system proposed in (2) above has the weakness that it is
    readily jammed by simple depth charge explosions or other sources of
    acoustic noise (e.g., Analog Equipment Corporation DUCK-TALK voice
    synthesizers linked with 3,000 AMP amplifiers).  An alternative is to
    make use of the ultimate in jam resistance: neutrino transmission.
    For all practical purposes, almost nothing (including several light-
    years of lead) will stop a neutrino.  There is, however, a slight
    cross-section which can be exploited provided that a cubic mile of
    sea water is available for observing occasional neutrino-chlorine
    interactions which produce a detectable photon burst.  Thus, we have
    the basis for a highly effective, extremely low speed communication
    system for communicating with submarines.

> make use of the ultimate in jam resistance: neutrino transmission. For all practical purposes, almost nothing (including several light-years of lead) will stop a neutrino

Signals aren't jammed by being blocked but overpowered. If you want to disrupt someone's neutrino comms, you don't start building lead walls. You flood their volume with neutrino noise. (I don't know how feasible that is to continuously do over a large volume.)

To jam, presumably you need to know precisely where the receivers are.

> To jam, presumably you need to know precisely where the receivers are

You need to know where you don't want functioning receivers.

IANAP (I'm not a physicist)...

Can a neutrino detector estimate the direction from which a neutrino arrived?

I'm wondering if that could help discriminate signal vs. jamming.

Some neutrino detectors can estimate direction, see

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

which should be able to spot a supernova in our galaxy easily.

Got me curious if that's enough margin for HFT firms to invest...

24 ms is massive. 0.1 bps makes it tricky to make use of.

It's going to be a lot less than 24ms though because that's the antipode which is worst case for a around-the-globe signal. However, the antipode for probably every single stock market is likely somewhere uninhabited. It's still going to be a huge improvement of course but I'm sure there's encode/decode times to convert to/from an electrical signal. And presumably the 0.1 bps probably means it also takes a long time to receive a single packet (error correction & collecting enough signal in the first place). So we're probably a long ways away from it making sense for that use case.

The cord connecting Europe to Australia is almost exactly through the centre of the planet.

Also, the latency difference is a lot higher than 24ms because the current best route is via a circuitous fibre link. Light is slowed down in the glass of the fibre significantly (about 30 to 40%).

With any luck we might work out how to slow it down even more:

https://sf-encyclopedia.com/entry/slow_glass (1966)

Selling picture view windows that sat out in the scenery of Yellowstone for a few years and nano-glass-dust total surveillance are just a few of the exciting possibilities.

Aren't satellite-satellite laser links the new standard for latency-sensitive communication since it's point to point and avoids delays from refraction?

I'm not sure if they have true global connectivity yet, but. . . I'm sure SpaceX is working on it.

> cord connecting Europe to Australia

Also Puerto Rico to just off Australia.

Melbourne, the neutrino city!

1 ms would still be a big deal in HFT. The low bitrate pretty much kills it though (and yes, probably whatever lengthy processing is required to get a signal out of the particle soup).

To be clear: 24 ms is a massive improvement you mean?

I assume yes considering much of HFT is measured in ns.

Transmit uncertainty by circumference ahead of time, and low bandwidth info that resolves uncertainty through the core?

Maybe they could use a large cluster of beams in parallel? :)

Most of those neutrinos will keep going though.

Imagine aliens come some day and ask us "What's with all the neutrinos?" :-)

They're either going to be very confused or understand it perfectly depending on how their society(ies) work.

Not to mention we'll have to listen to people complaining that they have a headache because of neutrinos..

—I’ve become magnetized by the neutrinos! Look, my keys stick to my arm!

—Ma’m, that’s just because you’re sweaty and sweat makes your skin sticky.

How soon until some high speed trader figures out this would be a faster way to send some data than sending it around the earth?

Kind of makes me want to hook a multi-mode decoding radio to the output of a high efficiency (lol) neutrino detector to see if any of those extrasolar neutrinos are just interstellar packet radio...

I'd like to imagine this as an early step towards omnidirectional point-to-point links blasting through the planet at each other to replace the need for switching/routing.

High frequency traders pay attention please.

> Fascinating that there's a neutrino beam cutting straight through the earth, in a chord hundreds of miles long

This is not the first time! Search about neutrinos from CERN to Gran Sasso.

OPERA collaboration!

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

The K2K experiment also sent a beam from KEK to SuperK in Japan. I think there might have been one or two other exeriments like this. If you spend the time and money to create a neutrino detector, doing an experiment like this is a reasonable add on.

>a neutrino beam cutting straight through the earth, in a chord hundreds of miles long

In my copy of this mental image, I wondered if Earths' gravity, and indeed our movement through the cosmos, bends that chord, and what it would sound like ..

Earth's gravity certainly does (though not by much); our movement through the cosmos shouldn't appreciably, since the Earth is in freefall. (But tides will affect it.)

The concept of near and far detector for a neutrinos experiment is not new. We already have NOvA which DUNE will be its successor and probably will be able to solve the neutrinos mass hierarchy by then.

Redhat might need to rename their orchestration product

And while we are at that. Unix communities would make a great contribution by changing "root" as the super user account convention /s. This will help physicists searching for ROOT on the web.

Reply on news.ycombinator.com