DIYbanter - View Single Post

Don Y[_3_]

On 9/19/2015 8:28 AM, wrote:
On Sat, 19 Sep 2015 14:54:27 +0000 (UTC), Roger Blake
wrote:

On 2015-09-18, Oren wrote:
Casinos take ~80 photos per minute on the Las Vegas Strip. Street
cameras, red light cameras and such, too.

Fortunately there are countermeasures to facial recognition. Even wearing
large mirrored sunglasses and a wide-brimmed hat can be enough if there is
not already an identified facial image of you wearing those things in the
database. There are even active sunglasses using near-infrared LEDs under
development that will prevent the cameras from recognizing that a face
is present.

If you were wearing that the last time you were in the casino, they
have it.

Or, they may insist that you *remove* it! E.g., walk into a bank
with a cap and sunglasses and you'll quickly draw attention to
yourself!

As the police state and corporate interests continue their inexorable
encroachment on our liberties in the U.S. It is definitely going to
become more difficult to preserve a modicum of privacy.

What privacy?
Everyone got their panties in a wad over the NSA but Goggle knows a
whole lot more about you than the NSA and they read/analyse every byte
of thor Gmail if you use it. I imagine most "free" email services do.

The difference is, you have the option of *not* using their services!
There's no way to "opt out" of NSA surveillance!

I self-censor what I discuss with folks based on their email provider.
This isn't foolproof; google handles mail for some ISP's without
applying their name to the domain!

We are the most spied on populatrion in the history of the world and
the computer power makes all of that information searchable, unlike
the Stasi in East Germany that collected reams of data that they could
never actually use.

China and Britain have excessive monitoring capabilities.

As to "never actually use", I contend that the volume of data collected
defies anything other than *targeted* use -- looking for something *specific*.

You can use Big Data to identify trends. But, only if those trends manifest
in reasonably "static" ways. E.g., if someone placed a Craigslist ad in
January using a particular phrasing; then visited the Bolshoi ballet 9 months
later (and happened to use the men's lavatory on the second floor);
then placed a phone call to a particular phone number 2 months later,
you'd be hard pressed to extract this relationship from the amount
of data you've collected. Even if it was a 100% positive indicator
of something of interest to you!

Instead, you look for more static data that "stands out" by itself in
shorter terms or with less associated "noise" (e.g., wanting flying
lessons and being of middle eastern descent)