From: Mike Steven (mikesteven@ndirect.co.uk)
Date: Sat Apr 01 2000 - 13:37:24 MST
Michael S. Lorrey <retroman@turbont.net> wrote:
> > >
> > > Lets look at the number of cameras in a total ubiquitous open
> > > surveillance system:
> > >
> > > Range of average camera(at useful resolution): 50 m radius
> > > Field of view: 90 degrees
> > > Area under surveillance by 1 camera: ~7500 square meters
> > > Area of dry surface of Earth: 148,300,000 square kilometers or,
> > > 14,830,000,000,000,000 square meters
> > > Number of cameras necessary to cover every square meter of the dry
land
> > > on the planet:
> > > 20,000,000,000,000, or 20 trillion cameras,
> >
> > But the vast majority of those cameras would be pointing at a place
where
> > there aren't any humans, vehicles or robots and so would be worthless. A
> > fairly universal surveillance system (obviously I agree no system will
be
> > perfect) could get by with a tiny fraction of that number of cameras,
> > particularly if the cameras are mobile. As for cost, already today you
can
> > get an (admittedly rubbish quality) web-cam for less than £50, and at
that
> > price it wouldn't be exhorbitantly expensive to have every room in your
> > house under surveillance. (You may be relieved to hear that I don't have
any
> > plans to start webcasting Meonthetoilet.com any time soon :)
>
> Good to hear. However, the fact that a camera isn't pointing at anything
> doesn't mean it won't be on the network, and won't be getting data
> downloaded from it by automated filter/analysis systems, and won't be
> taking up bandwidth.
My point was that the majority of your hypothesised 20 trillion cameras
would be unnecessary (and therefore wouldn't exist). There would be little
point in putting a camera in the middle of an uninhabited forest (unless you
want to check if a falling tree makes a noise when there's no one around to
hear it). Also, even very crude filtering technology (cameras don't record
unless the image changes, let's say) could drastically reduce the need for
bandwidth and memory.
> Moreover, a current day web-cam is of pathetic
> quality as an instrument of surveillance.
Agreed, but significantly better cameras do exist and I don't think it would
be very controversial to claim that their cost is likely to plummet. Also,
note that even a cheap and nasty, present day web-cam would still be of some
use for surveillance. (It might not be able to ID someone by their
fingerprints from across the room but it would be able to detect the
presence of an intruder). See also Eugene's comments wrt gait analysis etc.
> Additionally you brought up a
> factor that would INCREASE the number of cameras on the network:
> enclosed areas. For example, my apartment is less than 100 square
> meters, yet I would need 5 cameras to cover it all on the interior, so
> there would be many billions of areas of suboptimal coverage where you
> would need overlap. My original calculation was that there was optimal
> area coverage and camera utilization.
Even with a great deal of redundancy, there is no way you would need to have
20 trillion cameras to keep track of 6 billion people (or perhaps ten
billion in the near-ish future). Particularly if some of the cameras are
mobile.
> > If that level of surveillance is possible today, I don't think it's a
huge
> > stretch to postulate a future system which could cover not just your
house,
> > but also shops, businesses, public areas etc and at significantly higher
> > quality than the best available today. Sure, if you decide to go hiking
in
> > the middle of nowhere there won't be any cameras, apart from those you
bring
> > with you (or ones which follow you if we soup up the tech a bit).
>
> However, this system would be totally dedicated toward security, nothing
> else. Most people wouldn't care to spend more than 1-5% of their annual
> income on personal security.
But by the time you add up spending on defense, the police & insurance you
already spend significantly more than 5% of your income on security - plus
there's the cost of maintaining all those guns. :) Surely in a future which
contains considerably nastier threats than today you'd be prepared to spend
even more.
> You are talking about a system that would
> cost several times the entire GDP of the planet to build and maintain on
> an annual basis. Cameras wear out, cameras get destroyed by weather, and
> by people who don't want to be seen.
Nope, I'm talking about a system which a fairly well-off individual could
implement a primitive personal version of TODAY. If I so desired, I could
have every room in my house (and it's exterior) covered by web-cams. I could
also strap a series of small video cameras and microphones about my person,
all of which could upload in real time everything they record to multiple
remote servers. Meanwhile speech-to-text software could be transcribing
every word I say (or at least most of them) and those of people near to me
or on my property. Likewise voice analysis software could match the voices
to existing profiles to get their names or flag a voice it hadn't heard
before. Even some kind of basic threat detection system is fairly easy to
do - My house might send me a message as I approach that one of the external
cameras has "spotted" someone hiding behind my dustbin, perhaps waiting to
attack me. I bring up the image being recorded by that camera on my palm
top (with mobile data link), confirm that the lurker looks suspicious and
alert the police.
Note that whilst all the above might severely eat into my beer money, it's
all possible with technology available now. And this technology is going to
get (much) cheaper and (much) more powerful.
<snip>
I wrote:
> > To carry on the poisoner example, Let's say the poisoner is president of
the
> > Magic Circle (and therefore adept at slight of hand, used to concealing
> > things from cameras and a master of disguise). He slips something in my
food
> > as the waiter brings it from the restaurant's kitchen. Despite all
this,
> > the fact that he walked right past my plate just before I was served
(and
> > then keeled over dead after my first bite) would be registered by the
Crime
> > Analysis Computer. The CAC could then rewind the tape, tracking his
> > movements backwards (perhaps over several cameras, or via his own
personal
> > camera(s) until the point when he's putting on his fiendishly clever
> > disguise, identify him and send round the cops. (Or if it's looking at
> > historical footage, fast forward whilst tracking him to his current
> > location). Obviously it's all a bit late for me personally, but the very
> > high probability of retribution is likely to be an effective deterant to
the
> > committing of the dastardly deed in the first place.
Michael S. Lorrey <retroman@turbont.net> replied:
> And you are assuming that he would be inserting the poison in your food
> in an area that is covered by what you've admitted will be a suboptimal
> system (not likely).
If I'm particularly paranoid about the prospect of being poisoned I could
avoid food which has been prepared in a non-monitored area. However, I
would think that any reputable restaurant might have such monitoring to
reassure their clientel (not just with regard to poison but also health &
safety regs etc).
> You are also assuming that the act will be
> performed by the individual themselves, rather than by an autonomous
> agent.
Obviously autonomous agents would complicate things, but by the time the
technology arrives for autonomous assasin-bots, presumably similar defensive
counter-measures will also exist (or be rapidly developed). Eventually
you're going to need active shields and all the rest.
I wrote:
> > As for EMP and most other camera disabling technologies (all those where
the
> > network knows the camera has been disabled), presumably this would
trigger
> > immediate alarms, which would probably cause me to, at the very least,
stop
> > eating and be on my guard (and probably take further defensive action if
I
> > was the sort of person to have a sworn enemies list :). In any case I
would
> > presume that the act of disabling a camera in a public place (or on
someone
> > else's property) would itself be a crime and footage just prior to the
zap
> > would probably be enough to identify the zapper (or at least place them
on
> > the scene).
> >
> > Also, if the technology was good enough to have a personal mobile camera
> > covering everyone then anyone who disabled their own camera (prior to a
> > crime) would be inviting suspicion and might automatically be tracked by
> > other cameras as they pass. Yeah, we're talking a lot of computing power
and
> > memory (but hands up anyone who doesn't think future capabilities in
these
> > areas won't dwarf today's).
Michael S. Lorrey <retroman@turbont.net> responded:
> Dwarfing todays total global capacity for ONE application? I don't think
> so.
But given full Drextech today's "total global capacity" could fit in
something the size of a sugarcube - and think what you could do with a
computer of similar density which was the size of a room. Plus you wouldn't
need anything as powerful as nanotech to run a system which can cover most
of the
people, most of the time. (Terms which have cropped up on in this thread
like "total", "ubiquitous" and "universal" are overstating the case - no
system could cover everybody from every angle all the time).
I wrote:
> > Incidentally, I'm not saying that this kind of surveillance capability
> > doesn't make me uncomfortable, but if such a system manages to prevent
some
> > religious nut from wiping out humanity before I can grow a diamond
> > space-ship and head for my nano-engineered asteroid utopia then the
> > (admittedly big) bad points might be worth it.
To which Michael S. Lorrey <retroman@turbont.net> replied:
> THis is the worst weasley argument to make. The paradox of this argument
> is that:
>
> a) your perceptions of crime and our trust in humanity today are based
> upon the lowest common denominator
> b) the lowest common denominator is a low paid, low intellect, low
> educated lowlife.
Ignoring victimless crime, the majority of criminals do fit this
description - for every 1 Professor Moriarty, there are 500 thugs.
> c) at the same time, you claim to hold the transhumanist ethic that
> technology will lift up and improve the lives of all people, increasing
> their intelligence and capacity to learn, and access to knowledge and
> information.
Yep.
> d) that your perceptions of crime are based on the lowest quartile of
> the population, while your expectations of the future population reflect
> the highest quartile.
If most victimless crime now is committed by the lowest quartile (conjures
up images of Snow White and the seven deadly dwarves) then
this seems reasonable. I don't see any paradox in my expectation that
future technology will enable these people to exceed the capability of the
"highest quartile" today.
> e) you are assuming that the transhumanist trends will make human beings
> better people, but when it comes to crime you assume that they will
> become the worst possible human beings.
I would like to think that transhumanist trends will make human beings
better people, but what if they only make them more capable people? (I can't
see you, or myself for that matter, supporting government controlled brain
editing, surgically enforcing some norm of niceness). Given that, some
people are still likely to turn to crime (if they can get away wiith it) and
as technology improves, the possibility that an individual could take out a
significant proportion of humanity will also rise. Not a cheering prospect.
> This is the perceptual dissonance in the arguments for total
> surveillance that makes it not only distasteful for me, but IMHO exposes
> the hypocrisy of the people who beleive in it.
Given that surveillance tech is already powerful and will become vastly more
so, I see only two future possibilities:
1) Highly intrusive surveillance is developed but made illegal for everyone
but governments (which use it to spy on their powerless citizens).
2) Highly intrusive surveillance is developed but the citizens have access
to it themselves and can use it to monitor governments (and each other).
You may not like either option, but given a choice between the two, I'd
rather have the second.
<snip>
I wrote:
> > Current network security / anti-virus stuff may not be great (far from
it,
> > in fact), but it is improving and even at today's standard I doubt any
> > individual or organisation would be able to totally take out the
internet
> > short of launching a lot of nukes (and even then there's likely to be
some
> > survivors in a bunker somewhere who'll still be able to play Quake). If
this
> > is true today, I imagine future systems are likely to be less rather
than
> > more vulnerable to such attacks.
Michael S. Lorrey <retroman@turbont.net> wrote:
> As chips drop their operating voltage to get greater processing for less
> power consumption and heat generation, they actually become easier to
> EMP.
Fair enough.
Taking the babyblanketnet out completely is not necessary to make
> it useless. An extended campaign to simply corrupt the reliability of
> the information it put out, and produce areas of spotty coverage, will
> cause people to lose faith in the system and refuse to pay for its
> upkeep.
>
Again I think you're overestimating the cost of such a network, and
underestimating its capability to help catch criminals (including those who
attacked the network itself). Also, if the network is in the hands of the
citizens and it more or less eliminates crime, (and reduces government
corruption, war-mongering, international tension etc) then people might be
prepared to go to great lengths to maintain it.
This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 15:27:48 MST