Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror
×
Earth Networking The Internet Science Technology

Internet's Deep Infrastructure Could Double As a Sensor Network For Earthquakes and More 37

citadrianne writes with an article at Motherboard that exposes an interesting under-use of the worldwide physical network that carries Internet traffic. Even though there are many thousands of miles of undersea cable (containing many times that length if you add up the various lengths of fiber), the physical body of the internet is remarkably un-useful when it comes to detecting things like seismic shifts. From the article: "Right now the current system of cables on the seafloor is deaf, dumb, and blind," said Rhett Butler, the director of the Hawaii Institute of Geophysics and Planetology at the University of Hawaii. "Although they carry trillions of bits of information and basically run the global economy at this point, they don't know anything about the environment they're in. They don't measure anything at all and that seems crazy."

According to Butler, AT&T and other telecom companies have paid lip service to the idea of integrating sensors into the cables, but he has watched proposal after proposal for smarter cables fall through for a variety of reasons.... "[In] a certain sense mankind has given the nod to lay cables across the open sea floor without any restrictions, so it seems to me to be a little reasonable [for the telecom companies to have] a little obligation on their part to help people out."
This discussion has been archived. No new comments can be posted.

Internet's Deep Infrastructure Could Double As a Sensor Network For Earthquakes and More

Comments Filter:
  • by Anonymous Coward

    "it seems to me to be a little reasonable [for the telecom companies to have] a little obligation on their part to help people out."

    The only obligation for companies are to make money in a lawful way... and lobby the law so it allows them to make more money.

    On one hand you mock (rightfully) those that see companies as physical persons but on the other request them some kind of moral ethos. WFT!?

    • by Adriax ( 746043 ) on Tuesday August 11, 2015 @10:30AM (#50293707)

      There's a good reason not to make them smarter too.
      Sensors could be used to detect military movements and possibly find small/stealth unit movements as well. So a sufficiently paranoid nation or group could decide to destroy those sensors.

      No company would want to give north korea or isis a reason to actively target undersea cables.

      • Not to mention the simple technical challenges of operating an actual sensor network at those depths and distances. If it were easy, they wouldn't be waiting on the Telecoms to foot the bill
      • > No company would want to give north korea or isis a reason to actively target undersea cables.

          It might give them MORE reasons but, given my understanding of their agenda, those two (at least) already have reasons
          to want a capability of targeting undersea cables.

          A bit more in the sensor department might at least give warning that something is about to be damaged
          and info about the thing doing the damaging.

    • the "missing bit" detector is in the architecture. if Area X stops moving traffic, for instance, there is stuff happening there that is overwhelming. might be the power's out. but it might be a flood, hurricane, alien landing, etc. if you can't ping anything in the area, start checking the ham bands and the visible satellites...

    • The only obligation for companies are to make money in a lawful way... and lobby the law so it allows them to make more money

      You, also, must be new here (Planet Earth, 21st Century, that is). We're talking corporations in a largely capitalist world, here; if they can fuck everyone, and screw the planet for gigantic profits, and get away with it, they'll do it in a heartbeat. Not spending more than they have to for their infrastructure, regardless of any 'benefit to mankind', is just SOP.

  • "that seems crazy." (Score:3, Informative)

    by Nutria ( 679911 ) on Tuesday August 11, 2015 @10:30AM (#50293703)

    Only to people who have been in the Ivory Tower waaaaaay too long.

    • by Anonymous Coward

      Power cables, ethernet cables, HDMI cables...

      None of them measure anything.

      You know what's crazy? The idea of cables with sensors.

      • by Falos ( 2905315 )
        Even the internet of things will be internet of things.

        [_] No Retard
        [_] Some Retard
        [X] Full Retard
    • by msauve ( 701917 )
      "Rhett Butler, the director of the Hawaii Institute of Geophysics and Planetology at the University of Hawaii"

      Scarlett: Are you tryin' to tell me you don't believe in the cause?
      Rhett Butler: I believe in Rhett Butler, he's the only cause I know.
  • What a fascist (Score:2, Insightful)

    by Anonymous Coward

    You let them lay cables, then demand that they "help people out" and "do their part", AFTER the cables are laid?

    This is why people fail at capitalism. A deal is a deal. Want a better one, make a different deal. It's not "unfair" that later on you realize that you don't have something you only just thought of.

  • by QuietLagoon ( 813062 ) on Tuesday August 11, 2015 @10:38AM (#50293781)
    Hmmmm... how to inject 100bps sensor data into a fiber optic-based data stream running multiple hundreds of Gb/s.

    .
    The acceleration is gonna tear those photons apart.

  • by Overzeetop ( 214511 ) on Tuesday August 11, 2015 @10:42AM (#50293827) Journal

    At his last presentation to AT&T the CEO - after hearing his pitch - was heard to say, "Frankly, dear sir, I don't give a damn."

  • Do one thing well (Score:5, Insightful)

    by petes_PoV ( 912422 ) on Tuesday August 11, 2015 @10:48AM (#50293859)
    If you want an array of seismic sensors, build one. If you want an inter-continental data network, build that. Don't try to hack the one to do the job of the other.
    • by David_Hart ( 1184661 ) on Tuesday August 11, 2015 @11:10AM (#50294071)

      If you want an array of seismic sensors, build one. If you want an inter-continental data network, build that. Don't try to hack the one to do the job of the other.

      Exactly. The complication and the added failure modes would go up drastically if sensors were incorporated into communications lines. What happens if all of the sensors fail? Do you tear up the communications lines just to fix them?

      What you do is build a separate system, build relationships with the communications companies, and make a business deal to have it laid at the same time as undersea cables. I agree that there are economic benefits to sharing the same deployment resources, but they should be separate infrastructures.

  • by pz ( 113803 ) on Tuesday August 11, 2015 @11:00AM (#50293965) Journal

    Surely there's some secondary or tertiary affects that can be used to measure cable movement like microphonics, and thus deduce seismic activity. From the title, I had thought that's what the posting was about. If you can influence the error rate of a disk drive by yelling at it (https://www.youtube.com/watch?v=tDacjrSCeq4), then can't you measure earthquakes with a long optical fiber?

    • If you can influence the error rate of a disk drive by yelling at it (https://www.youtube.com/watch?v=tDacjrSCeq4), then can't you measure earthquakes with a long optical fiber

      No, because you don't know where in the cable the event occurs. One possibility would be to dedicate a fiber to these measurements, using OTDR (optical time domain reflectometry) to detect an event. But this would only work on cables without repeaters. Given that these microphonic effects are most likely increadibly weak, and the attenuation incredibly high it probably wouldn't work.

      Long story short: these cables were meant for no other purpose than transmitting data -a task daunting enough in itself- and

      • by pz ( 113803 )

        And disk drives are made to deliver data reliably as well, but out-of-band signals can be used to detect other influences (latency reporting vibration, for disk drives). For optical cables, seismic activity might (for example) increase the number of packets that require data correction, and the error rate would be your out-of-band signal.

        Knowing that there is seismic activity somewhere along a long cable, even if you don't know where, is better than knowing nothing. With information from enough such cable

        • That is a very interesting proposal. If the amount or errors can give any additional information, and this could be used to locate seismic activity, why not exploit this indeed. Looking at the cable map, I have the impression most of them are roughly in the same region, so I have my doubts about the accuracy, but then again, I am no expert. It may tell more than nothing, as you remarked. Obviously, this is very different from the suggestions of the article, to install active sensors.
  • Not peanuts (Score:4, Insightful)

    by elgatozorbas ( 783538 ) on Tuesday August 11, 2015 @11:13AM (#50294115)
    From TFA:

    In terms of rote economics, implementing these sensors into the cable system would be “peanuts” compared to what telecom consortiums are already paying to lay cables across oceans. According to the October ITU report, adding these sensors would add an additional 5-10 percent to the cost of laying a new cable, which generally cost hundreds of millions of dollars.

    5-10 percent of hundreds of millions of dollars is not peanuts to me. While this may be an interesting idea, I can imagine that telecom operators are not enthusiastic to implement something this costly, which adds complexity to their installation with limited benefit for themselves.

    • Exactly. I suspect that the companies laying fiber would be willing to split the common costs with him; but they're not interested in paying for his sensor network for him.

      • Exactly. I suspect that the companies laying fiber would be willing to split the common costs with him; but they're not interested in paying for his sensor network for him.

        Only if they get paid for it.

        If someone wants it they have to pay for it. That's largely the way our economic system works at present. It will change as we get smarter AIs involved, hopefully without revolution, but in the meantime that's what we're stuck with.

Keep up the good work! But please don't ask me to help.

Working...