3 ms·

The fact is, outside the echo chambers of HN and reddit--and even within, by a silent minority at least--people simply don't worry about NSA / etc concerns enou

The fact is, outside the echo chambers of HN and reddit--and even within, by a silent minority at least--people simply don't worry about NSA / etc concerns enough to let it affect their day to day life. I'm one of those. This isn't to say I endorse in any way what's going on (and I've written to my congressional representatives to that extent), but whether rationally or irrationally I am really not concerned about NSA surveillance affecting me personally. Will it negatively affect the nation as a whole, sure, but I honestly don't think wearing Glass (once the price drops enough I'd even consider buying one) will pose an real threat to me. So yeah, I do think it's a remotely good idea, as long it's useful to me otherwise.

xyzzy123 13y ago
I understand your position and I think it's pragmatic. My concerns are mainly societal, as widespread use would provide unprecedented new surveillance capabilities. The main difference is that cellphones mostly only compromise your own privacy, whereas Glass users are in a position to enable data collection on everyone around them too. Data from Glass may already be available to existing collection efforts. It's just an Android device. Glass cloud syncs by default and uses the regular (insecure) Android backup mechanism. Google have been careful to point out that Glass does not support facial recognition. Sure, but image recognition is much better done in a server farm across bulk collected data anyway. If glass is widespread, it becomes plausible to track targets (non users, vehicles etc) through "crowd-sourced" surveillance images. This gets really interesting when you have fine-grained location data and the ability to correlate between multiple users (whose path in the world is known). Most of the time the system could just work off passively collected (synced) data. There would be no indication on the device itself that this is going on, but it would not be real-time. However if a target is sought in a certain area, you could go active. Software could be pushed to nearby Glass units to conduct a high intensity search[1]. In a metropolitan area, you could turn on thousands of eyes at will[2]. Naturally you could also collect audio or other sensor data if you needed it. What amazes me is that this is not science fiction - it's an obvious extension of existing collection efforts. A lot of the required infrastructure already exists. Given what we know to date, I view efforts to develop such a system as basically inevitable. The idea is too obvious, too tempting. That's why I'm worried about Glass. [1] Take lots of pictures. Be willing to drain battery. You might want to do some local processing to minimize load on the cell network. [1] Google can push software to Android devices. Carriers can push updates. There are probably a bunch of other ways to get software on there.

xyzzy123 13y ago

I understand your position and I think it's pragmatic. My concerns are mainly societal, as widespread use would provide unprecedented new surveillance capabilities. The main difference is that cellphones mostly only compromise your own privacy, whereas Glass users are in a position to enable data collection on everyone around them too. Data from Glass may already be available to existing collection efforts. It's just an Android device. Glass cloud syncs by default and uses the regular (insecure) Android backup mechanism. Google have been careful to point out that Glass does not support facial recognition. Sure, but image recognition is much better done in a server farm across bulk collected data anyway. If glass is widespread, it becomes plausible to track targets (non users, vehicles etc) through "crowd-sourced" surveillance images. This gets really interesting when you have fine-grained location data and the ability to correlate between multiple users (whose path in the world is known). Most of the time the system could just work off passively collected (synced) data. There would be no indication on the device itself that this is going on, but it would not be real-time. However if a target is sought in a certain area, you could go active. Software could be pushed to nearby Glass units to conduct a high intensity search[1]. In a metropolitan area, you could turn on thousands of eyes at will[2]. Naturally you could also collect audio or other sensor data if you needed it. What amazes me is that this is not science fiction - it's an obvious extension of existing collection efforts. A lot of the required infrastructure already exists. Given what we know to date, I view efforts to develop such a system as basically inevitable. The idea is too obvious, too tempting. That's why I'm worried about Glass. [1] Take lots of pictures. Be willing to drain battery. You might want to do some local processing to minimize load on the cell network. [1] Google can push software to Android devices. Carriers can push updates. There are probably a bunch of other ways to get software on there.