3 ms·
Upvote for raising clear points, but at heart I think I disagree with your premises. Your thesis seems to be that by throwing more resources at the problem the
by nkurz 18y ago
Upvote for raising clear points, but at heart I think I disagree with your premises. Your thesis seems to be that by throwing more resources at the problem the vault could be made more secure. Likely true, but at what cost, both in hassle and money?
You seem to be arguing for overengineering without considering the costs. Should the failsafes also have failsafes, should the backups have their own backups? And what if the additional security measures add additional risks? Take #4, "human patrols". Unless implemented very carefully, greater access for security personnel could well add more risk than benefit.
I'm not even sure that one should argue that the vault failed from the point of view of the owners. It was designed well enough for them to do their daily business and such that they were able to obtain insurance on the contents. A vault doesn't have to be impregnable, only visibly difficult to steal from.
- gdee 18y agoAllegedly, the robbers walked with 100M euros without cleaning the vault. Price would have to be really steep for those additional security measures for their cost to be an issue I think. > A vault doesn't have to be impregnable, only visibly difficult to steal from. I'm pretty sure a vault in this situation has to be really impenetrable and definitely not just visibly difficult to steal from. What you're proposing is the exact situation they had. The result didn't look good.
- nkurz 18y agoI'm not quite proposing the exact situation that they had, but I am suggesting that adding more layers of physical security might not be the solution. And if indeed this was really a case of insurance fraud rather than a straight theft, perhaps the outcome was good for everyone except for the insurance company. To the extent that the vault was good enough for them to obtain insurance, the vault was good enough. To obtain better security, the motivation of the participants needs to be analyzed. Perhaps this is a problem best solved through obscurity, with two competing security teams installing two totally separate alarms systems, each standing to lose if their alarm is defeated. Or perhaps skip the alarms, and rely on threats of deadly force. But although any new insurer will properly demand the technological improvements illustrated here, it would be foolish to believe that technology alone will keep out a dedicated team of insiders with full access to the system. Real world security is not equivalent to public key encryption where security can be established by 'proof'.
- gdee 18y agoI fully agree with the notion that (more) technology is not always a panacea but I think it should at least be given a fair chance. That includes implementing it properly. Given that this was not the case there (at that time) I would still bet on technology going forward. And process.
- potatolicious 18y agoWell, there are a few simple fixes that wouldn't have cost much at all but would've significantly improved the security of the vault: 1 - The magnetic plate sensors should NOT be accessible while the vault is closed. There is no reason why this should have been mounted outside the vault instead of inside. 2 - All sensors in the vault were tied to a single circuit . Shorting just two wires was enough to take every single sensor in the room offline. That's ridiculous. 3 - You shouldn't have a simple "break circuit if alarm tripped" design at all. You have a full-blown sensor package mounted on a wall in a secure vault, is it too much to ask that the sensor respond to pings? This is a fairly minimal cost (+$100 per sensor, maybe?) that would also make it impossible to short the sensor circuit or otherwise take a sensor offline without somebody realizing.