![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
An issue of security.
Nov. 21st, 2006 11:33 amSomeone got into the (secure?) carpark and smashed out the windows in Paul's car, all to get at a CD case, some cheap sunglasses and the change in the ashtray.
Nasty and upsetting for Paul- the doors are scratched, the leather cut up from the smashed glass, and his pride and joy defiled.
What's upsetting for me is that the building has all the components for a good security system. RFID swipe tags for tenants' access, pin numbers for tradespersons to use temporarily, mag locks on most doors, cameras on gates and scalable walls. Put all that information together and its easy to investigate the theft: how the crook got into the building, who he might have seen while he was here, how he got out again. In this case, though, it all falls apart when you go to put the information together. The swipe-key controller is a separate system and requires a Chubb Security officer to retrieve the access data. Manual door releases are not logged. The video system records everything but exporting the video is difficult in the extreme.
If I were designing the system, I'd have all the cameras, access control, power control logging every single little event into a webpage. It would then be simple to follow someone, even an unauthorised person, by video and by revealing which doors were opened and when, which lights were turned on, which lifts operated. Might not help with catching him once he's gone, but at least you could talk to everyone he came into contact with and get a good description.
Such a system could offer other insights too: does one particular stairwell or door get used far more than the others? It'll need extra maintenance then. One bathroom more popular? Clean it more often. Which bulbs would be best replaced with energy savers, which left as regular cheap globes? Could the phasing of the carpark traffic lights be tuned to favour the more frequently used floors? How about something real fancy: If you turn off the lights, set the alarm and lock the office door at 6pm, what if there were a lift waiting for you on your floor? And the lights on just your floor of the carpark were on?
When I am king, I shall have these toys. Maybe I'll become king by building them. In a lot of cases the infrastructure is there already, its just a matter of being able to imagine how the data might be used to make things better. In the hands of benevolent computers, of course.
Nasty and upsetting for Paul- the doors are scratched, the leather cut up from the smashed glass, and his pride and joy defiled.
What's upsetting for me is that the building has all the components for a good security system. RFID swipe tags for tenants' access, pin numbers for tradespersons to use temporarily, mag locks on most doors, cameras on gates and scalable walls. Put all that information together and its easy to investigate the theft: how the crook got into the building, who he might have seen while he was here, how he got out again. In this case, though, it all falls apart when you go to put the information together. The swipe-key controller is a separate system and requires a Chubb Security officer to retrieve the access data. Manual door releases are not logged. The video system records everything but exporting the video is difficult in the extreme.
If I were designing the system, I'd have all the cameras, access control, power control logging every single little event into a webpage. It would then be simple to follow someone, even an unauthorised person, by video and by revealing which doors were opened and when, which lights were turned on, which lifts operated. Might not help with catching him once he's gone, but at least you could talk to everyone he came into contact with and get a good description.
Such a system could offer other insights too: does one particular stairwell or door get used far more than the others? It'll need extra maintenance then. One bathroom more popular? Clean it more often. Which bulbs would be best replaced with energy savers, which left as regular cheap globes? Could the phasing of the carpark traffic lights be tuned to favour the more frequently used floors? How about something real fancy: If you turn off the lights, set the alarm and lock the office door at 6pm, what if there were a lift waiting for you on your floor? And the lights on just your floor of the carpark were on?
When I am king, I shall have these toys. Maybe I'll become king by building them. In a lot of cases the infrastructure is there already, its just a matter of being able to imagine how the data might be used to make things better. In the hands of benevolent computers, of course.
