Hacker News new | past | comments | ask | show | jobs | submit login

"IRC. Everyone who wants to make sure their messages land use IRC as a backchannel now."

No, damnit. IRC is logged, double logged, and then has all of the logs shipped offsite to be engraved in bedrock. IRC servers, because of their history in pirated content, are very heavily controlled by law enforcement. A huge number of raids have been carried out around the world based off what people have said on IRC.

IRC is very much not the solution to the widespread monitoring of facebook.




A lot of my friends who use IRC as a primary communication channel will stick to invite-only self-hosted SILC IRC networks. These are essentially 1 step above darknet (eg: local IRCd), and are a far cry from the efnets of the IRC-verse.


All it takes is a warrant and a plea bargain, and your private 'darknet' (haha) irc server is now a honeypot collecting all the traffic you thought was private. The more popular an easily controlled solution like private IRC servers become, the greater the advantage for law enforcement and others to control them.

End-to-end encryption is a requirement for secure communication. Anything less is the same as scribbling on paper and tossing it to the recipient, and hoping nobody catches it on the way.


PRQ (http://prq.se) serves as a decent host insofar that it's harder for the server to be seized--it works for us. Obviously there is no perfect solution, but if you don't have a target on your back then this is one way to decentralize communication with a group of friends.

To take it a step further, end to end encryption is only good when you can protect your private key and passphrase... All you have to do is take a look at Rakshasha (https://www.youtube.com/watch?v=yRpilXPv8pU) for a practical example of how compromised we all really are.


As long as we can all agree not to get complacent, it's a step in the right direction. Arguing that people need to go form 0 to 60 in under a second isn't going to get us anywhere.

For example, if you want to convince someone to become a vegan, which path is more likely to succeed:

  Omnivore => Vegan
or:

  Omnivore => Vegetarian => Vegan
or even:

  Omnivore => No red meat => Only fish => Vegetarian => Vegan


On the other hand.. what's easier, convincing someone to change a habit once, or convincing them four times?


Some diet habits are harder to persuade, and many won't stick with them. So anything is something.



pyre assumes (and I agree with him) that the difficulty (D) of convincing someone to change the habit satisfies:

D(Omnivore => Vegan) > D(Omnivore => No red meat) + D(No red meat => Only fish) + D(Only fish => Vegetarian) + D(Vegetarian => Vegan)

Often intermediate steps are significantly easier to implement than just implementing the end result. It's the same reason for which people tend to break big scary tasks into sequence of many easier ones.


That's a fascinating question! Do you know many vegans? I know a lot of them, and I'd say on the order of 75% of them went directly from omnivore to vegan.

They hadn't thought about the ethics of meat before, when they did, they chose to be vegan.

So it turns out (in my experience) telling rational people "do the thing that makes sense, even if it's a little extreme" is much more likely to result in the most extreme change of behavior. But we're deep into anecdotal/psychological theory land here, so I claim no authority beyond N=30 vegans.


Yes, that is what SILC is for. It has been around for really long.


I think that statement was more concerned with censorship than privacy.


Are they not one and the same? Take a look at Edward Snowden.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: