Big Screen for NYPD

Roland Piquepaille writes "With millions of emergency calls every year, the New York City Police Department (NYPD) decided to invest in a new command and visualization center in order to keep up with the ocean of data it has to deal with. According to this article from BusinessWeek Online, the display system consists of hundreds of Mitsubishi digital light-processing (DLP) monitors covering three walls. The NYPD thinks it will help it to also manage the hundreds of thousands of annual arrest records and to further reduce crime in the city. You'll find more details and references in this overview, which includes impressive pictures of former visualization centers built by Imtech, which will integrate the NYPD one."

Comments

Lower Crime?

[nuclear305]

How exactly does a new wall of monitors lower crime?
Are these monitors secured to the wall in some new way to prevent theft?

All joking aside, how does this lower crime? If a Bigger, Better display helps lower crime, doesn't that imply that they are currently allowing things to slip through the cracks because they can't manage their data?

Viewing the Wrong Way

[tintruder]

Perhaps the public would be better served if the screens were placed on the outside of the buildings, looking inward at what is going on in government.

The magnitude of the fraud, waste and abuse so rampant throughout government pursuing boondoggles like this is endless, and the excuse is always some "Sound Bite Focused" explanation that "It's for the Children", "It's for the fight against terrorism", or some other thinly veiled B.S. intended to take ever more tax money from citizens and waste it on needless government programs.

A smaller example of this was in Portland, OR where the police needed an extra quarter million dollars in order to be able to track "Racial Profiling" in traffic stops. Seems that none of the cops were able to record the vital statistics of who they stopped unless they were given Palm Pilots (and all sorts of other alleged I.T. expenses to support them).

Seems nobody even considered those little paper notebooks and a few boxes of pencils.

Amazing how the public seems entirely ignorant of the paramilitarization of the police and the resulting "Us against Them" rift that continues to widen.

The best thing that can be done in the U.S. (Short of Jeffersonian suggestions of periodic revolutions to toss out abusive and tyrannical politicians) would be to cut all government spending and staffing by 25% immediately, and 50% within 5 years.

Political showpieces and $$ for supporters

[Anonymous Coward]

I worked years for the IT division of a major city's police department and I can assure you that nothing is a bigger waste of money than such things as visualization systems, etc. These are, as in most politics, a way to pass huge amounts of money to political supporters and friends.

Any "trends" in criminal activity are located far faster by the cops on the beat than by a computerized system. These guys know their beat like the back of their hand: details that would never be stored in a computer are at their fingertips. They are _extremely_ observant. By the time any "visualization center" knows it has a problem the cops have been on it for hours at least. This is a form of "swarm intelligence": independent agents (police officers) cooperating, exchanging information and coordinating activity. If you impose a hierarchical command structure, the flow of information can be imparied.

As for "trends": what are you looking for? A 5% increase in convenience store robberies? Day to day police work deals not with statistics but with individual incidents. "Trends" are important, but mostly to politicians and bureaucrats who must fund police work long-term. The police are concerned that someone robbed two Stop N'Go's in the west borough in the last 3 hours, killing 3 people. That's not trend analysis, that's a f'ing problem to be solved quickly.

There are good uses for statistics and trend analysis in police work, but they don't require a huge realtime display of information - they require only a CRT that can produce a graph or a map and some quiet time for the captains to think about how they will allocate their beat's manpower next month or how they can justify a request for additional manpower for a particular precinct where crime levels are rising year-to-year. This is traditional spreadsheet and database work.

Re:Political showpieces and $$ for supporters

[ratsnapple tea]

Since you seem to be so skeptical about the usefulness of computerized crime tracking technology, I think you might find this article from the Economist an interesting read. I couldn't tell whether it's subscriber-only, so I'll reproduce part of it here:

CRIME maps, which record the locations of incidents in order to help predict where criminals are going to strike next, are used by police throughout the world. But the past is not always a helpful guide to the future, and a team of criminologists from University College, London, led by Kate Bowers, think they can do better. A test of their new model, unveiled in this month's British Journal of Criminology, suggests it is 30% better at predicting crime than traditional methods.

It is a cliché to say that crime spreads like a disease, but previous work by Dr Bowers and her colleagues found that this is exactly how crime does spread. Using statistical techniques developed to study the transmission of infections, they found that burglaries cluster in space and time in predictable ways. For example, properties within 400 metres of a burgled home, particularly those on the same side of the road, are at an increased risk of being broken into for up to two months after the initial incident.

Using these and other findings, the team created algorithms that predict where criminals will strike next, and then used those algorithms to generate "prospective hot-spot maps". These divide an area into 50-metre squares--a level of resolution chosen because 50 metres is a typical line-of-sight for a police officer in an urban area--and give a crime forecast for each square.

In their paper, Dr Bowers and her colleagues reveal the results of a study of burglaries in Merseyside, in northern England. Using historical data, they pitted their predictive modelling method against two traditional crime-mapping systems. They found that their method successfully "hindcasted" 62-80% of burglaries. The traditional techniques, by contrast, hindcasted only 46% of those incidents.

Computerized crime tracking technology like COMPSTAT is already helping to make police departments more efficient, focused and accountable in the real world. No, it won't alert you to a Stop'n'Go shooting spree in the last 3 hours, but it does help you clarify the big picture, about where carjackings are becoming more common, which neighborhoods are becoming more robbery-prone, that sort of thing. And that information can be immensely useful to an overworked precinct with limited resources (overtime, etc.) to do their jobs.

I'm not defending this expensive realtime display covering three walls of a command center, but I don't think the facts justify your skepticism about the use of trend-finding in police work.

Ah, that's the thing.

[Eevee]

In the images shown in the links, and from the limited experience I've seen of them in person, they aren't looking at the same thing. There's five or six different things being displayed (and a monitor showing CNN because the boss thinks it's neat) that have nothing to do with each other. If the people only use a small portion of what's being displayed (or don't really use it at all), it's a pretty expensive toy.

Now if the display was one 'thing'--for example, a wide area network status with some of the monitors devoted to a map showing the links, while others showed statistics--then I see the value. And I'm sure there are places using them in just this way; is it just that people showing off the multi-monitor displays feel a need to be flashy with ten different things being displayed, then go back to a boring yet practical application?