Comment to USCG on NAIS policy

These are comments on policy for sharing information from the
U.S. Coast Goard’s NAIS, a network of 140 AIS receivers covering
U.S. coastal waters, as solicited by Federal docket USCG-2009-0701.

(The docket request for comments is here.)

First, I declare my interest. I am a lead of the GPSD project, a set of open-source software tools for collecting and processing GPS and AIS data. GPSD is extremely widely deployed on Linux and Unix systems including navigational suites, SBC telemetry packages, and cellphones. In connection with this project, I am also the editor of the most complete publicly available description of AIS/AIVDM decoding.

One goal of my work is to make access to high-quality GIS information generally available for purposes of research, day-to-day navigation, and public safety.

NAIS data is collected with tax funds. Thus, policy formulation should begin from a presumption that, absent a showing of specific unmanageable risks, the general public is entitled to free access to the data. The burden of demonstration should fall on those advocating restrictions rather than openness.

The policy I advocate is complete openness: that all data collected by NAIS should be made available in real time as an AIVDM stream via a Internet public feed at a stable and documented address.

1. How might providing real-time, near real-time, or historical NAIS information to the public impact maritime commerce?

Accurate and timely information is the life-blood of commerce. In the past, increases in the information richess of the environment in which market actors make decisions have shown a strong tendency to promote economic activity of all kinds, help markets clear more rapidly, and increase average wealth levels.

I see no reason for general publication of NAIS information to be an exception. I cannot predict what specific business strategies or tactics it will enable, but I think the precedent of GPS provides strong reason for optimism.

2. What would be the impact of providing this information, if any,
on the following?
a. Safety of ships and passengers or crew,
b. Security of ships and their cargo,

I do not anticipate a safety or security impact.

The substantial safety benefits of AIS in navigational and collision avoidance systems are readily collected by LOS (line-of-sight) use through local receivers. Ship velocities are low enough that non-LOS information over an Internet feed is not generally relevant in real time.

In the past, there has been some concern that Internet publication of AIS data might enable commerce-raiding, piracy or terrorism via non-LOS monitoring of ship locations. But in no conditions short of major war would commerce-raiding or piracy be an issue for the NAIS coverage area (U.S. coastal and inland waterways). And terrorists, unlike national governments, do not generally have the ability to throw cruise missiles over the horizon. Thus I conclude that the risk from publishing real-time ship locations in the NAIS coverage area is effectively nil in peacetime conditions.

c. Economic advantage or disadvantage to commercial stakeholders,

I see no disadvantage to anyone in a policy of open publication. Because Internet access is unreliable and expensive at sea and AIS receivers are cheap, substitution of real-time NAIS data for local receivers seems unlikely.

UPDATE: Commercial AIS providers object to open publication of NAIS data, but it is no part of the USCG’s mission to protect their business models at taxpayer expense. If they can offer a substantial value-add over information collected with taxpayer dollars, let them survive; if they cannot, let them fail.

A policy of unrestricted public access ensures that any commercial advantages will be symmetrically distributed without favor. Conversely, restrictions on the data would advantage large players with the resources to jump through bureaucratic hoops and/or good political connections — not a good outcome.

d. Environmental impact on extractable resources or coastal activities.

Difficult to call. On the one hand, open publication of NAIS data would probably increase general activity levels slightly, with concomitant slightly increased environmental risk. On the other hand, AIS is already being used for risk mitigation, e.g. by broadcasting whale pod locations. More general availability of such data might head off specific and serious environmental harms.

3. Is information collected by the NAIS considered sensitive?

I do not believe NAIS information should be considered sensitive in peacetime conditions.

a. Is real-time or near real-time information collected by the NAIS
viewed differently than historical NAIS information, and if so, how?

Historical NAIS information presents not even the minimal (wartime) risks of real-time informstion.

b. Does the sharing of information collected by the NAIS generate
concern about unfair commercial advantage? If so, for which segments of
the industry is this a concern?

As previously noted, open access would make asymmetrical commercial advantage impossible.

c. Is there a timeframe within which real-time or historical
information collected by the NAIS is considered sensitive or is no
longer considered sensitive?

See above.

d. Given that ships last for decades and that their capabilities
and capacities are relatively stable, is there a concern that
historical NAIS information might be analyzed to derive a competitive
advantage?

See above.

4. What controls on sharing real-time, near real-time, or
historical information collected by the NAIS with the public are
suitable?

General publication with no controls whatsoever would be the simplest, fairest, and least expensive policy.

a. Who should receive each type of NAIS information?

In the absence of wartime threats to U.S. littoral waters, all NAIS information should be made generally available in real time on a stable public Internet feed.

b. What are appropriate uses of information collected by the NAIS?

Research. Maritime traffic analysis. Robustness testing of AIS decoders.

c. Do message types matter?

I see no reason to complicate policy or implementation by distinguishing among message types. Publish them all and let the applications sort it out.

d. Should addressed messages be handled differently from broadcast
messages? Do addressed messages contain information significant to
understanding maritime activity? Should addressed messages be shared
with the public?

I see no reason to restrict access to addressed messages. Though addressed, the technological substrate of AIS is such that they are public broadcasts with no expectation of privacy. Privacy concerns are properly addressed via message encryption, which AIS readily supports.

UPDATE: I have added, in my resubmission of 16 Feb, a paragraph on why the USCG should not shield commercial AIS providers

27 comments

  1. BTW, I especially like this paragraph (emphasis mine):

    NAIS data is collected with tax funds. Thus, policy formulation should
    begin from a presumption that, absent a showing of specific
    unmanageable risks, the general public is entitled to free access to
    the data. The burden of demonstration should fall on those advocating
    restrictions rather than openness.

  2. I’m trying to figure out what similarities libgif circa 1990 has in common with GPSD today. Or maybe GPSD is only a cool hobby for you, and nothing more.

  3. I see no disadvantage to anyone in a policy of open publication. Because Internet access is unreliable and expensive at sea and AIS receivers are cheap, substitution of real-time NAIS data for local receivers seems unlikely.

    Given Moore’s Law and other measures of technological advancement, do you suppose this will always be the case? Do you foresee a future time in which the use of commercial satellites or terrestrial repeaters could make Internet access at sea, at least closer to the coast, to be more reliable and cheap? I can recall a time when Internet access to the average home was very expensive and unreliable (and I’m sure you can, too.) Now high-quality high-speed connections are nearly ubiquitous in urban areas around the country. I don’t see why this couldn’t be the case at sea within the next 10 years or so.

  4. >Given Moore’s Law and other measures of technological advancement, do you suppose this will always be the case?

    I think it’s likely, yes — because the cost of a local AIS receiver will be dropping along the same curve for the same reasons.

  5. “I’m trying to figure out what similarities libgif circa 1990 has in common with GPSD today. Or maybe GPSD is only a cool hobby for you, and nothing more.”

    It got me thinking too. ESR tends to say that the primary motivator behind OS development is that the programmer wants to scratch an itch of his own. What we can deduce from the bits about his lifestyle published on this blog is that it’s unlikely to have any GPS-related itches any standard, out-of-the-box SatNav-device wouldn’t scratch. As for programming for fun and challenge, that’s also unlikely – if I get it right, the purpose of the project is firewall mud and shit away so that others don’t have to deal with it. This is exactly the kind of challenge that’s not very enjoyable because enjoyable challenges are ones that involve exercises in complicated logic and not ones dealing with completely a unlogical set of disconnected “casuistry”. So the only explanation left is exactly what you mentioned – that there could be something, how to put it?… strategic about it.

    1. >So the only explanation left is exactly what you mentioned – that there could be something, how to put it?… strategic about it.

      You’re overthinking this. No, I don’t consider GPSD “strategic” — though I could be mistaken; I didn’t think GIFLIB was either, and wow was I wrong about that. The one thing they’ve got in common is that they’re the kind of infrastructure that’ll be on everyone’s cellphone someday, if it’s not already.

      No, I think the actual reason I’ve found GPSD rewarding is that it’s got a bit of every kind of challenge in it. Building the packet sniffer was an exercise in compiler technology; unsnarling the protocols is reverse-engineering; the error modeling requires numerical analysis; the daemon framework code has the form of a classic exercise in socket programming; there’s a custom application protocol at the center of the design. And so forth. A huge variety of different kinds of problems addressed in a codebase that’s still small enough that it’s comprehensible as a whole — approximately the complexity of a small operating system, and for many of the same reasons.

  6. ESR, the link to the AIVDM/AIVDO protocol is broken.

    Other than that, right on. Tax dollar generated information should always be open unless there’s a damned good reason not to.

    ESR says: link fixed, thanks.

  7. > No, I don’t consider GPSD “strategic” — though I could be
    > mistaken; I didn’t think GIFLIB was either, and wow was I
    > wrong about that.

    Cool hobby, then. Nothing wrong with that. I liked seeing the position feeds from ships streaming in in real time. Of course, its reassuring to imagine that someday my cell phone will have GPSD built-in, helping me find my way.

    Your USCG comment, on the other hand, has obvious strategic value, if only in the opening up of one data feed of one little corner of the vast machine that is the government.

  8. The “strategic” value in opening this information is mostly a matter of principle. We taxpayers paid for the data, so the presumption should be that we have a right to access it. We don’t know what people will do with the data once they get it — but that’s what’s great about a free marketplace. Someone will come up with a new way of doing things. BY DEFINITION, that means that the central authorities haven’t thought of it yet, and no one will know which box to check on the license form.

    Asking us to explain how the data will be used, to justify opening it up, is completely backwards. Instead, as Eric pointed out, those who think data should be hidden need to explain why, and any secrecy provisions need to be narrowly-tailored to address those legitimate concerns, not a blanket “that which is not expressly permitted is forbidden”.

  9. Private companies (like ours) are spending far less than the US government to develop our network of AIS receivers. We have more stations up than the US Coast Guard. What this sharing policy means is that our company will no longer be able to compete with the US Government giving away the data.

    The purpose of AIS wasn’t to give data to foreign governments and others such as private companies and end-users.

    I don’t think the public would like the idea that the US Government intends to provide the historical and real-time positions of their vessels to foreign governments and others. In fact, they would be horrified to hear big brother is watching at all.

    If this idea plays out, information about fishermen’s positions would be available to all including competitors. Where is my right as an AIS equipped vessel owner to use this tool for navigation while not having my exact whereabouts broadcast around the world?

  10. >What this sharing policy means is that our company will no longer be able to compete with the US Government giving away the data.

    Boo-hoo. See me cry tears of pity. Next time, try a business model that’s not utterly a creature of government fiat, fool.

    >Where is my right as an AIS equipped vessel owner to use this tool for navigation while not having my exact whereabouts broadcast around the world?

    You have no such right, because you’re broadcasting your position in clear for anyone to listen to. If you desire such privacy, the proper target of your ire is the regulations requiring you to so broadcast, not the disposition of the broadcast data by USGS or anyone else.

  11. Gee, Eric, I would think you would sympathesize with someone who is about to be steamrollered by an arbitrary change to a stupid set of government regulations, especially since the arbitrary change does not fix the stupidity. Particularly since, in this case, the data – ship positions – should be private. I don’t want to broadcast my personal position, or the position of my personal vehicle to the world at large and can understand why vessels at sea might have similar qualms.

    OTOH, I don’t want to broadcast my personal position, or the position of my personal vehicle to M Calouro’s company either.

    In other words, perhaps the best comments about the proposed rule change would propose that ships not be required to broadcast in the clear.

    Anyway my ignorance on the subject is profound. Maybe there is an absolutely rock solid reason why these positions must be broadcast in the clear – to avoid collisions. for example. Doesn’t seem to make sense, though. How did ships avoid collisions before? My idle speculation surely illuminates only my ignorance.

    M Calouro, however, seems to understand the situation, as do you. If you shout past each other for a while the resulting heat and light might illuminate things for the rest of us and be entertaining. Free ice icream!

    Yours,
    Tom

  12. >Maybe there is an absolutely rock solid reason why these positions must be broadcast in the clear – to avoid collisions. for example.

    Got it in one. One principal goal of AIS was to be able to build automated anti-collision systems that would operate reliably in conditions of zero or low visibility.

    >I don’t want to broadcast my personal position, or the position of my personal vehicle to M Calouro’s company either.

    Presently the regulations require that all passenger vessels and all cargo vessels of 300 tons or more displacement must have an AIS transponder operating at all times. The correct response, then, would be to seek a regulatory change allowing the installation of a privacy switch.

    There’s an obvious tradeoff here: the privacy switch would create a negative externality for the other vessels for which you would become an increased navigation hazard. I’m actually not sure which side of this I’d come down on.

    But the rights and wrongs of such a change are orthogonal to the real issue here, which is whether a company like M. Calouro’s can, by selling pooled AIS info, acquire some sort of entitlement to have information collected with tax dollars locked away from the public. To that I say “Hell, no!”.

  13. >But the rights and wrongs of such a change are orthogonal to the real issue here, which is whether a company like M. Calouro’s can, by selling pooled AIS info, acquire some sort of entitlement to have information collected with tax dollars locked away from the public. To that I say “Hell, no!”.

    Well I have to agree with you, except that, perhaps everyone, including the government, should be prohibited from collecting such information and selling it. I am pleased not to collide, but not pleased to have my position given either as a matter of public record, or sold to the highest bidder. Such a law would be difficult to enforce, however.

    Yours,
    Tom

    1. >Such a law would be difficult to enforce, however.

      And create some very odd contradictions and edge cases in the law, because there is strong precedent and doctrine protecting anyone’s right to receive and use over-the-air navigation signals for any purpose, including commercial purposes.

  14. >And create some very odd contradictions and edge cases in the law, because there is strong precedent and doctrine protecting anyone’s right to receive and use over-the-air navigation signals for any purpose, including commercial purposes.

    Well, my ignorance of the subject shows up again. This would make good cover stories or small search vessels a must when treasure hunting.

    Yours,
    Tom

    1. >You got the carriage requirements almost completely wrong.

      Interesting. The International Maritime Organization’s (IMO) International Convention for the Safety of Life at Sea (SOLAS) only requires AIS to be fitted aboard international voyaging ships with gross tonnage (GT) of 300 or more tons, and all passenger ships regardless of size.

      I learn from your link that the U.S. requirements are more detailed and strict than this. I will amend my decoding document accordingly.

  15. >>You got the carriage requirements almost completely wrong.
    >
    > Interesting. The International Maritime Organization’s (IMO) International Convention for the Safety of Life at Sea (SOLAS) only requires AIS to be fitted aboard international voyaging ships with gross tonnage (GT) of 300 or more tons, and all passenger ships regardless of size.

    You’ve been reading/quoting wikipedia again, haven’t you?

    http://en.wikipedia.org/wiki/Automatic_Identification_System says:
    The International Maritime Organization’s (IMO) International Convention for the Safety of Life at Sea (SOLAS) requires AIS to be fitted aboard international voyaging ships with gross tonnage (GT) of 300 or more tons, and all passenger ships regardless of size.

    The actual requirement is:
    All ships of 300 gross tonnage and upwards engaged on international voyages and cargo ships of 500 gross tonnage and upwards not engaged on international voyages and passenger ships irrespective of size.

    http://www.navcen.uscg.gov/enav/ais/AIS_Regs_SOLAS_MTSA_FR.pdf

    1. >You’ve been reading/quoting wikipedia again, haven’t you?

      Yes. I’m pretty careful about using Wikipedia as a rule, but expected it to be reliable for this sort of fact. Disappointing…

  16. Tom: Well, my ignorance of the subject shows up again. This would make good cover stories or small search vessels a must when treasure hunting.

    M Calouro: If this idea plays out, information about fishermen’s positions would be available to all including competitors. Where is my right as an AIS equipped vessel owner to use this tool for navigation while not having my exact whereabouts broadcast around the world?

    I want to address these.

    In general, what is to stop me from simply following a fishing or treasure hunter to sea and maintain visual or radar contact the whole time? Legally nothing provided I obey proper navigation rules while doing so. The case of treasure search vessels being followed has actually occurred minus AIS (read “Ship of Gold in the Deep Blue Sea” by Gary Kinder for an account of treasure hunters being shadowed by competitors at sea).

    So concern for protection of commercial trade secrets such as fishing grounds kind of rings hollow on that basis. Furthermore, AIS is line-of-sight. So the USCG making AIS data available over the Internet is only going to matter when vessels are close inshore arriving or departing ports with receiver coverage. Once their over the horizon, it ceases to be of us. Since most fisherman tend to be well offshore, the value of AIS in determining their locations diminishes unless you’re in a boat following them at a distance with a receiver.

    Hell, watch the Discovery Channel for cases of fishermen following each other to play with their heads or mess with their catches. All AIS does is make it easier close in to be an asshole at sea. See “Whale Wars” for an example of that.

    The value of AIS outweighs its negatives. I find the real-time AIS feeds fascinating on land. On the water, it’s no longer academic to me but an issue of safety. I’m planning to fit an AIS receiver to my sailboat because of its value. Much cheaper than radar and on the Chesapeake Bay in marginal conditions it may be a lifesaver. The ability to know what the freighter 3-5 miles away is doing and allow me to plan accordingly is priceless.

    It is difficult on the water to determine the speed of a vessel approaching you head-on. Which is what big ships do to small boats on the Bay all the time since they are restricted to a fixed north-south shipping channel from Baltimore to Norfolk. They’re always approaching you head-on. AIS can tell me before I can figure out on my own (not that I should depend solely on AIS over good watchkeeping and other sensors but it helps widen my decision and safety loop) when I should maneuver to be well clear of a ship approaching me.

    At night that type of information can save lives. A sailboat is all but invisible to a large ship on the water. AIS can tell me its coming. In conditions like fog or a squall where visibility is diminished, it can mean the difference between getting home or getting run down.

    Matt

  17. This is topical again this month, since some news outlet blew the lid off AIS.

    Mail to esr@thrysus.com soft-bounced after 4 days, twice.

    Bruce Schneier has the story, as did Sans; can dig up links, if you can’t find them.

    1. >Bruce Schneier has the story, as did Sans; can dig up links, if you can’t find them.

      I can’t. Please post.

      I seem to be getting normal mail, so you might also try re-mailing them as a test.

Leave a Reply to esr Cancel reply

Your email address will not be published. Required fields are marked *