How not to free the Internet

Good afternoon, dear habrovchane. Waking up today at fourteen o'clock in the morning, I found two articles on Habré: The US government betrayed the Internet. We need to get it back into our own hands and How to specifically release the Internet . Frankly, I was surprised by their promise to such an extent that I decided to create this post without even finishing my morning tea.

I feel very uncomfortable, being forced to argue with Bruce Schneier. But there is an opinion that he, despite his enormous authority, is now critically wrong.



What exactly is proposed to us in these two articles? This can be described in three words: "the protection of everything somehow." Any data on the Internet should be encrypted, but not so well that the security forces with a strong desire could not crack the cipher. I affirm that this is an approach diametrically opposed to the right one.
')
Let's face it: a significant (if not more) part of the information passing through the Internet is not private, and most of the private information is not secret. Personally, I can go to the main square of my city and recite 90% of my correspondence over the past week with a feeling. Another 9% is what I consider to be personal. I will not share this with everyone, but I’m clearly aware that if these data are in a huge pile of other possible data available to the special services, then nothing terrible, bad, or even unpleasant will happen. Most likely, they simply do not pay attention.

The remaining percentage is something that you should probably encrypt. If we assume that the figures given by me are true for the average person, then “protecting everything” in practice means a one-hundred-fold increase in the costs associated with information security.

Now from “everything” we will move on to “somehow”. Bruce Schneier and the author of the second article, chainik , agree on allowing the security forces to decipher the messages of the “scoundrels”. Here, first of all, I will take the word “rascals” in big, big quotes (I’ll have to do this in my mind, because I didn’t find where the button “increase font size” is in Habr). Secondly, I will allow myself to refer to another article on information security: Tools for treason .

If your algorithm does not allow the pedophile to irreversibly spoil your hard drive and avoid punishment, it will not help people who are under ideological pressure to keep prohibited books. If your instant messenger application does not allow you to safely plan a terrorist attack at the World Championships, then it will not help the activist talk about human rights violations. If your map does not allow poachers to catch rhinos without attracting the attention of environmental organizations, it will not be able to be used by national minorities to avoid “cleansing”. The power of the tool determines the very possibility of these things, and this is a very, very old question about the purpose for which this tool will be used. And the answer, as usual, will be "for those and for others."

The main direction of the state’s work is homeostasis, maintaining the status quo. And the possibility of targeted decryption in the first place will turn against those who are trying to change the existing order of things. Against people like Snowden and Assange. Simply put - against the "political". For scoundrels (without quotes) is simply not enough resources.

The report is over.

PS Weak encryption "just in case" - in principle, not a bad idea. But not at the expense of anything. I want the photos of cats to be loaded and sent as quickly as possible, and I don’t mind that the special services or even Putin himself could see them.

PPS Opinion about the leak into the network of "compromising" materials I have outlined here .