HN Theater @HNTheaterMonth

⬐ Intermernet

This talk is (as expected) a very level headed summary of why exceptional access in encryption is a very bad idea.

The only point I think it's worth adding to those in the talk (which may be covered in the original paper) is one of temporal trust. That is, even if you manage to convince yourself that there could be a way to safely solve the technical problems of setting up exceptional access in the current world, and you decide that all LE agencies that get access are currently trustworthy, and you believe that other issues such as LE agency corruption and spying are currently minimal to non-existent, There is no way to ensure that this "ideal" (and somewhat naive) world will continue into the future.

I think any person or government that tries to argue that exceptional access is a good idea either hasn't considered historical precedent, hasn't thought the situation through to it's logical conclusion, or is being deliberately disingenuous.

⬐ diskcat

b-but terrorists and pedophiles

⬐ a_imho

lovejoy's game: when you invoke lovejoy's law, you lost the argument

⬐ pdkl95

A warning: beware legislation (or non-legislative regulation or order) that attempts to accommodate the problems inherent in exceptional access. Tech-focused people often use strict logic-based categories when considering if something is good/bad/useful/whatever. "Exceptional access" would create serious problems in many areas, so the entire concept should be abandoned. While this is true, legislators often add exceptions to legislation.

For example, it would be a lot harder to argue that there is a national security risk in legislation that only mandated exceptional access to the crypto used in consumer-level phones. Law enforcement, business, etc gets real crypto with the excuse "why would a business owner be a terrorist?" or similar nonsense.

> deliberately disingenuous

At a minimum, this has to be the case for the people that were arguing against encryption since the first crypto war.