Not Eliza -- more like PARRY:
PARRY was written in 1972 by psychiatrist Kenneth Colby, then at Stanford University.[1] While ELIZA was a tongue-in-cheek simulation of a Rogerian therapist, PARRY attempted to simulate a paranoid schizophrenic.
The Feds' insistence on sweeping up all the innocent-communication "noise" is drowning out the signals. (e.g. "Hey, you might want to keep an eye on those Tsarnaev brothers -- see attached description of the stuff they were doing while they were still here in Russia.")
Indeed. This proves beyond any sane doubt that the targets are not foreigners (who for obvious reasons would ignore any "legal framework" and avoid using defective-by-design NSA-approved encryption). The targets are domestic.
If companies want to take the direction of removing themselves from the encryption picture altogether, that is their prerogative.
And yet that is precisely what the government is pissing and moaning and setting its hair on fire about. Showing that sort of contempt for citizens' private prerogatives is what caused them to forfeit our trust in the first place.
Actually, "both of the above" (foolish and malicious) fits the available evidence best. For instance, Rogers' answers at Monday's cybersecurity forum were both pathetically lame (foolish) and contemptuous of American values (malicious).
And their fruits are rotten and infested with vermin.
That's another advantage of forcing the snoops back to "direct access" methods -- every so often one of them will get caught red-handed snooping on the wrong (i.e. clearly innocent and rich/influential) target, re-focusing attention on them and forcing another round of retrenchment until it blows over.
Now, if a backdoor is found by the bad guys, it will be used almost immediately to destroy a company.
If it's found by really bad guys (e.g. North Korea on a day when Dear Supreme Grand High Panjandrum is feeling especially trollish), it can be publicly circulated to destroy every company.
Anybody can yell "Police" or wear a jacket reading "Police". I recall reading about at least one home-invasion gang doing just that.
Forcing them to switch to "direct access" methods puts pressure on them to follow the law. First, as I noted in my earlier comment, the non-scaling time and manpower costs (each tail, bug, etc requires significant additional resources) forces careful selection of targets. Second, "direct access" methods put the snoops at a nontrivial risk of getting caught and/or leaving recoverable evidence each time they use them illegally.
"Direct access" methods (tailing people, planting surveillance devices, etc) do not scale anywhere near as easily as network surveillance -- each "direct access" target requires a significant fixed cost in resources and manpower. This imposes discipline on the snoops and forces them to pick and choose actual suspects instead of trying to scoop up everything.
Hey, Bob, attend this class, make a list of attendees. Oh man, this is too easy!
It was easy; the instructor gave me a copy of the list. Now, we just need to pull up the physical and e-mail addresses of Mr. Dover, Mr. Jass, Mr. Jablome....
That's one advantage of mostly cooperative board games where there might be a hidden "traitor" among the players who wins if the group loses (e.g. Shadows Over Camelot, Battlestar Galactica). With that possibility on the table, players can't just trust somebody else to make decisions and have to pay attention to what everyone else is doing (usually in these games, exposing the traitor has some reward, at least insofar as it curtails his ability to continue undermining the group).
webmail kinda implies that the provider will either be storing the key or at the very least be able to access it
Obviously they need access to the PUBLIC keys in order to encrypt messages to the designated recipient. The whole point of public-key cryptography is that revealing the public key doesn't compromise security.
Any circuit design must contain at least one part which is obsolete, two parts which are unobtainable, and three parts which are still under development.
