I guess I never actively thought about it, but instinctively considered the screen to be off, but the phone as a whole to be on. I grew up with desktop computers and on those if you turn off the video screen, it doesn't impact on the rest of the system. I simply applied the same mental model. But now that you mention, it makes all sense for phone makers to implement measures to lower consumption like what you described.
There's not a clear, explicitly stated indicator to the user that there's a standby mode so it being ignored is understandable because the other option is me being dumb and I'm not dumb, duh
Yeah, what did you call it before smartphones when the phone was on but wasn't on? Heck in the old days the screen wouldn't turn off, but the backlight would
@Bob someone should sniff the traffic at the router level (where nothing can be hidden/obscured) and see if the image data is actually transmitted over the wire or not
someone was saying they have JS (or Swift or Java if it's a native app on your phone?) that will do the hashing locally
@djsmiley2k one-way hash functions in general -- insofar as they are cryptographically secure -- never need to be "hidden" from adversaries; the whole point of hash functions is that there is no efficient (meaning, computable) algorithm for associating any specific input data with any particular hash, or going from a hash back to the original data
if I have a one-way hash function that I trust to be a correct and cryptographically secure hash function, I should be able to hash my most valuable password, tell you which website it's for, and post it right here for you to see and not worry
well, true -- but in general it'll probably be quite easy to get around any "fuzzy matching" hashing algorithm anyway, because human creativity far outstrips computational similarity
to truly block the image, they'd have to keep a copy of the actual image data on their servers, and use an "ideal" (read: 99.9999999% perfect) Computer Vision system to look at (in the same sense as a human being would "look at") other images and determine if they're fundamentally "The Same Image" (even altered)
> According to a Facebook spokesperson, Facebook workers will have to review full, uncensored versions of nude images first, volunteered by the user, to determine if malicious posts by other users qualify as revenge porn.
Schneier is smart, and he might be making inferences on what they'd have to do to make the system do what FB claims it'll do, but no one really knows what they'll be doing unless they're transparent about it I guess
> What that and other explanations do not necessarily make clear, however, is that prior to making that fingerprint, a worker from Facebook’s community operations team will actually look at the uncensored image itself to make sure it really is violating Facebook’s policies.
yeah, I mean, that check has to be there to prevent someone from taking my professional studio pic and claiming it to be "revenge porn" so I can't use it :P
unless they have True Computer Vision that can tell them with extreme accuracy whether a pic represents something intimate, but good luck training that dataset
also, I think the people whose job it is to manually review these images are going to need psychiatric help, as people are going to submit images that are intentionally disturbing for them to look at
@Bob that's another thing -- certain classes of image data (such as what you just said) are illegal to possess for any reason unless you're a law enforcement officer on duty confiscating something under a warrant (IIRC there are laws that prevent, for example, cops from being liable when they raid an illegal drug factory, etc. that cover these scenarios)
but Facebook isn't associated with government or law enforcement, so them possessing the images for any reason makes them liable
the images are sent directly to facebook but that's after a request/complaint at e-safety
I didn't even know that existed
> Facebook is piloting the technology in Australia in partnership with a government agency headed up by the e-safety commissioner
> In the Australian pilot, users must first complete an online form on the e-safety commissioner’s website outlining their concerns. They will then be asked to send the pictures they are concerned about to themselves on Messenger while the e-safety commissioner’s office notifies Facebook of their submission.
oh, hm. it is supposed to be proactive
> it would allow victims of “image-based abuse” to take action before pictures were posted to Facebook, Instagram or Messenger
hmmmmm
yea that makes it harder
"they're not public yet but should I share them with facebook just in case others post it?"
@allquixotic something they'd have to work out with the us gov, presumably, before they try it there
I'm using gntp-send to send messages to a growl for windows instance. GFW hasn't been maintained a while so there dosen't seem to be support for windows 10 style notifications. How would I get GFW to work with the new style notifications?
I use a SSH client (bitvise SSH client) to connect to another Windows machine and launch a process. The processes take a long time to complete and I want to disconnect from the remote machine while keeping the process running. How can I do this? When I close the terminal or terminate the SSH sess...
I'm using gntp-send to send messages to a growl for windows instance. GFW hasn't been maintained a while so there dosen't seem to be support for windows 10 style notifications. How would I get GFW to work with the new style notifications?
