Not even dual boot.
Just depending on the password/pattern entered, log in as a different user. It's possible to do in Android as an app. I've been meaning to write one, but it seems like lots of work and upkeep.
What about the people under 18 who I assume could take the poll and also people who didn't go to vote but took a few seconds to take the poll on Telegram?
Also what about all the babushkas who most likely voted for Lukashenko but don't have a smartphone?
This is not to say that more people didn't vote for the opposition than the official numbers state. But Lukashenko still could have won.
There is lots of corruption and bureaucracy and lawsuit-avoidance driven construction. Right now the cost of constructing subway stations in NYC is the highest in the world.
It would be quite laborious to figure out "normal" user traffic patterns and then adjust to those. You would have to collect data on a bunch of users and then shape your own traffic to match.
Only makes sense if you are doing it for a bunch of people and at that point you are another VPN provider.
Regarding SMS interception, you can do it with every other messenger that uses this technique, which is basically every messenger that doesn't use passwords.
Regarding the nonce attack, it looks like the devs responded and said it was because of poor random numbers source on the client, which I personally don't understand as a justification. However, they said they'll remove it in the next update and that nonce has been "0" up until now.
Regardless, all of these messengers for cell phones aren't great if you are paranoid. That's because the hosting company's servers have all kinds of data on you as it is. Your contacts, access to SMS, access to location, camera, mic, photos, and all the files on the device.
This is true for all the messengers that are currently in widespread use.
Problem with such use of statistics is that we underestimate black swan events.
Humans are intuitively better at it. For example, it would be impossible to predict something like 911 event with such frequentist analysis.
Also impossible to predict some sort of nuclear disaster terrorist act that never happened before and could take a million lives.
Another thing to keep in mind is the "missing life" (dying young) and quality of life after disease. That's why something like Alzheimer's seems a lot worse than heart disease.
I think the issue with this approach is it doesn't take away individualism. Everything is still too far away. Most people still have cars and look for parking for hours. The parks are few and something like 30 minute walk away. Your friends and gym are also going to be 30 minutes walk away. It's not dense enough for subways everywhere and buses stop on every block so it makes just as much sense to walk. Nobody walks anywhere besides their neighborhood and there is usually nothing in their neighborhood.
So it's really worse than suburbs and worse than city center that has buildings. Worse than city with buildings because you don't share common resources so everyone is still for themselves. You also don't get nice things like common public playgrounds and parks.
It's worse than the suburbs because you still get the same problems of it being difficult to get places but now it's even harder. You still get "what's mine is mine and I don't want public things" but now you have less of your own stuff and fewer public things because real estate is expensive. You also don't get the diversity of food or clothes that you would get in a city.
I think the ideal is 90k/sq mile in the places where people live with 4-6 story buildings.
Yes, because they try to strip the binaries as much as possible so that the file size is smaller. If you compile a regular "hello_world.c" with gcc -O3, the size is 8 kb. You can def make a malware that is way smaller that does something simple like change some registry value to some uri.
Average size for malware is ~100kb-200kb btw. This is way smaller than almost any software besides some console games.
The reason simple hello world apps are marked as malware by AI-based engines is that small programs tend to be malware. There are probably close to no hello world programs in their dataset.
The problem is that to be a "good" program, yours has to do something other "good" programs do. And there aren't that many of those that use different software development paradigms (not different versions of the same one). So new different programs tend to be marked as malware.
This also brings up a very easy way to circumvent such AVs. Simply modify an existing goodware program and it will be marked as goodware. Add some obfuscation/polymorphism and it would be virtually impossible to detect such malware using static analysis/AI-based AVs.
Having worked on a machine learning-based AV for several years, I'd like to point out that the dataset choice here is extremely important and they seem to have a pretty small one considering the number of possible variations and the choice of model.
What happens in the wild is that one malware author releases a lot of very similar polymorphic or differently-compiled malware so it ends up being trivial to identify it. For example, they could have picked up a small icon that is common to half of the malware or some internal library that is used in a large portion of them. Then a week later the nature of the malware changes and you would identify a lot less.
Another thing to consider is that in many cases, a tiny modification to a known good program can make it malicious. This includes such things as changing the update URI. I don't see how they could catch such malware using this method so the 98% detection seems like a very unrealistic number.
Just to present an example:
One can train a simple logistic regression on some metadata features where the malware comes from one source and easily identify almost all of them correctly, while failing to identify malware from most other sources.
Having said that, it's a pretty cool novel approach and I'd love to try it.
The wikipedia page for it explains it well.