4 ms·

I don't doubt that Signal has put a significant amount of effort into making sure its blurs can't be reversed. But when I look at the results in the photo, I do

I don't doubt that Signal has put a significant amount of effort into making sure its blurs can't be reversed. But when I look at the results in the photo, I don't understand why they would put in that effort. Why derive a blur from the base photo at all? Things that seem way easier to me: A) blacking out the face entirely with a solid color, B) if that looks ugly, replacing it with some kind of clip-art, C) if that still looks ugly, replacing it with a gradient D) if that still looks ugly, replacing it with a pre-blurred face from a generic set of buckets. I sort of get the aesthetic argument, but I also really don't, because the way Signal is blurring faces is ugly, at least in the photo they show. It's not a seamless thing that blends into the background and looks way better than a solid color. It's giant squares, and the amount of blurring means that the contents are basically indistinguishable from a radial gradient to my eyes anyway. Am I missing something? Would a gradient really look any worse than this? Is there some kind of use-case where blurs give aesthetically much better results than what we're seeing in the photo? Are the concerns I'm seeing below about de-masking just fear-mongering? Are blurs in general just pretty safe, fast, and easy to do? Moxie isn't stupid, I assume in situations like this he knows what he's doing.

matmann2001 6y ago
Also consider the social effect. To the ordinary eye, it's not obvious that this blur is resistant to reversal, unlike standard blur effects. However, someone seeing the use of this blur effect, without understanding what it is, may come to the false conclusion that a non-Signal blur is safe too.
- danShumway 6y ago
  That's a good point. I've already seen at least one other person on this post comment to a Github repo that blurs faces. It might be secure, it might not be. I wouldn't trust it by default. With a static overlay, the method is simple enough that I can evaluate the security. With a blur, I don't know the difference between a good one and a bad one, so I can only trust the reputation of the author. I like to be able to look at the output of an anonymizer and to be able to tell myself at a glance whether it worked.
- tripzilch 6y ago
  I'm not even sure the Signal blur is safe, look at the comparison in this comment: https://news.ycombinator.com/item?id=23422993 https://news.ycombinator.com/item?id=23422993
dgco 6y ago
In https://arxiv.org/abs/1905.05243 https://arxiv.org/abs/1905.05243, we presented a measure of the effectiveness of eight face obscuration techniques. We did so by attacking the redacted faces in three scenarios: obscured face identification, verification, and reconstruction. Based on our evaluation, we show that the k-same based methods are the most effective.
billme 6y ago
>> “I assume in situations like this [Moxie] knows what he's doing.” If true, I sincerely would like to know why he is suggesting a non-violent movement should cover their face with military camouflage, which to many is literally a symbol of violence: https://signal.org/blog/images/blur-face-covering.jpg https://signal.org/blog/images/blur-face-covering.jpg ————— Moxie is Signal’s Founder & author of the blog post: https://en.wikipedia.org/wiki/Moxie_Marlinspike https://en.wikipedia.org/wiki/Moxie_Marlinspike He also pops up on HN: https://news.ycombinator.com/user?id=Moxie https://news.ycombinator.com/user?id=Moxie
- mckirk 6y ago
  Presumably because that kind of camouflage confuses face-tracking technology the worst.
  - billme 6y ago
    Nonsense. Source? Well defined 2D pattern stretched across as 3D surface would be easier to circumvent than a generic non-stretching mask. Beyond that, thermal & inferred imaging of particular covered faces have easily been able to identify an individual. Lastly, as I mentioned in my other comment, identifying a person is easily done via number of methods. Please refrain from “presumably” comments, they help no one.
giancarlostoro 6y ago
I was just commenting the same though, I like your ideas more too. I guess for most people it would look more aesthetically pleasing as you point out.

matmann2001 6y ago

Also consider the social effect. To the ordinary eye, it's not obvious that this blur is resistant to reversal, unlike standard blur effects. However, someone seeing the use of this blur effect, without understanding what it is, may come to the false conclusion that a non-Signal blur is safe too.

danShumway 6y ago
That's a good point. I've already seen at least one other person on this post comment to a Github repo that blurs faces. It might be secure, it might not be. I wouldn't trust it by default. With a static overlay, the method is simple enough that I can evaluate the security. With a blur, I don't know the difference between a good one and a bad one, so I can only trust the reputation of the author. I like to be able to look at the output of an anonymizer and to be able to tell myself at a glance whether it worked.
tripzilch 6y ago
I'm not even sure the Signal blur is safe, look at the comparison in this comment: https://news.ycombinator.com/item?id=23422993 https://news.ycombinator.com/item?id=23422993

dgco 6y ago

In https://arxiv.org/abs/1905.05243 https://arxiv.org/abs/1905.05243, we presented a measure of the effectiveness of eight face obscuration techniques. We did so by attacking the redacted faces in three scenarios: obscured face identification, verification, and reconstruction. Based on our evaluation, we show that the k-same based methods are the most effective.

billme 6y ago

>> “I assume in situations like this [Moxie] knows what he's doing.” If true, I sincerely would like to know why he is suggesting a non-violent movement should cover their face with military camouflage, which to many is literally a symbol of violence: https://signal.org/blog/images/blur-face-covering.jpg https://signal.org/blog/images/blur-face-covering.jpg ————— Moxie is Signal’s Founder & author of the blog post: https://en.wikipedia.org/wiki/Moxie_Marlinspike https://en.wikipedia.org/wiki/Moxie_Marlinspike He also pops up on HN: https://news.ycombinator.com/user?id=Moxie https://news.ycombinator.com/user?id=Moxie

mckirk 6y ago
Presumably because that kind of camouflage confuses face-tracking technology the worst.
- billme 6y ago
  Nonsense. Source? Well defined 2D pattern stretched across as 3D surface would be easier to circumvent than a generic non-stretching mask. Beyond that, thermal & inferred imaging of particular covered faces have easily been able to identify an individual. Lastly, as I mentioned in my other comment, identifying a person is easily done via number of methods. Please refrain from “presumably” comments, they help no one.

giancarlostoro 6y ago

I was just commenting the same though, I like your ideas more too. I guess for most people it would look more aesthetically pleasing as you point out.