[extropy-chat] Many eyes

[Dirk Bruere]

Dan Clemmensen wrote:

> Dirk Bruere wrote:
>
>> Dan Clemmensen wrote:
>>
>>>
>>>
>>> To increase pre-explosion coverage, the public should be encouraged 
>>> to make random pictures in public places, more or less continuously. 
>>> If nothing  interesting happens, most of these digital pictures will 
>>> never even be stored. If something bad happens, the pictures from 
>>> prior to the event would become available for analysis.
>>>
>>> Privacy? Sorry, These are pictures taken by individuals, in public 
>>> places. There is no right to privacy in this venue.
>>>
>>> I live in the Washington DC area. I thought of this  concept during 
>>> the ugly "sniper attack" situation last year.
>>
>>
>>
>> An average Londoner will be picked up on hundreds of cameras on a 
>> normal day.
>> The security services have had the City cameras connected to computer 
>> syatems that tracks every car entering and leaving that area for 
>> years, due to IRA attacks. I also recall that they were trialling a 
>> system in the 90s where the face of every person caught on camera was 
>> compared to a database of wanted people.
>>
>> The bombers will undoubtedly have been caught on camera and even now 
>> extremely powerful computersystems will be running pattern matching 
>> and people-tracing algorithms on all captured video for hours before 
>> and hours after the attack. I suspect that by now they may well have 
>> pictures of some of those responsible.
>>
>> It's how the last London bomber was caught.
>>
> Yes, Dick. The authorities do have a lot of video coverage, garnered 
> at great expense, and they are working hard at analyzing it. The 
> authorities have (in the last six hours or so) broadcast a request to 
> the public for any video or other pictures that the public may have.
>
> This request conveys two facts:
> 1) the "official" record is incomplete.
> 2) the "official analysts have not yet found the perpetrators.
>
> I do not contend that "official"cameras and analysts are inadequate. 
> Rather, I contend that society need not depend on the official 
> infrastructure. As individuals, we can instead fine the perpetrators 
> ourselves. Ten thousand amateurs are at least as effective as a 
> hundred professionals. This is the "cathedral" versus the "bazaar" 
> with a vengence (using the term advisedly.)
>
I'm not sure how useful that would be since most of the effort is 
looking for correlations between camera shots, in time sequence. That 
is, picking which faces out of a crowd occurs in what sequence on what 
cameras in order to determine movements pre and post event. It needs all 
the data to be run through the same algorithm, probably on a distributed 
computing system eg grid computing. I don't see the open source guys 
getting into the surveillance state thing and to me it sounds difficult 
since face recognition is only a subset of the problem.

-- 
Dirk

The Consensus:-
The political party for the new millenium
http://www.theconsensus.org



-- 
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.323 / Virus Database: 267.8.11/45 - Release Date: 09/07/2005