What's the best real-world (i.e., non-propellerhead) solution to this?
There isn't one. I've looked at creating one, and while its a sizable engineering job all the pieces are available. What it isn't is monetizable. Not like 'make me a gazillion dollars' monetizable, but like 'pay me a living wage to work on it' monetizable.
The key (and its a horrible pun) is the key. You can build zero knowledge proof [1] key exchangers now (patent expired :-) and a relatively inexpensive 'key' based on either USB or bluetooh communications (see Yubikey [2] as an example). Such that email to a new third party could be done in an encrypted way such that the message could only be read when that exact party was reading the email in a reader that could get the keys to unlock it from the physical key.
Like most such systems it only 'makes sense' if everyone (or at least a large fraction of everyone) has one.
To get initial adoption it needs support in 'free' tools which means it needs to be open in the sense that folks can trust what it does, and implement a compatible protocol without paying you anything.
To earn a spot in your pocket/purse/pack it needs to be flexible enough to accomodate other uses. To get those other uses the folks who provide them need to be able to support it for 'free' since their customers won't be paying them to put it in.
So a large investment in propellor heads to make it usable by the rest of the world and achieve critical mass for adoption. Oh, and if you do start getting traction the governments of the world are going to want to disappear you (which was one of Bruce's points)
There isn't one. I've looked at creating one, and while its a sizable engineering job all the pieces are available. What it isn't is monetizable. Not like 'make me a gazillion dollars' monetizable, but like 'pay me a living wage to work on it' monetizable.
The key (and its a horrible pun) is the key. You can build zero knowledge proof [1] key exchangers now (patent expired :-) and a relatively inexpensive 'key' based on either USB or bluetooh communications (see Yubikey [2] as an example). Such that email to a new third party could be done in an encrypted way such that the message could only be read when that exact party was reading the email in a reader that could get the keys to unlock it from the physical key.
Like most such systems it only 'makes sense' if everyone (or at least a large fraction of everyone) has one.
To get initial adoption it needs support in 'free' tools which means it needs to be open in the sense that folks can trust what it does, and implement a compatible protocol without paying you anything.
To earn a spot in your pocket/purse/pack it needs to be flexible enough to accomodate other uses. To get those other uses the folks who provide them need to be able to support it for 'free' since their customers won't be paying them to put it in.
So a large investment in propellor heads to make it usable by the rest of the world and achieve critical mass for adoption. Oh, and if you do start getting traction the governments of the world are going to want to disappear you (which was one of Bruce's points)
[1] http://en.wikipedia.org/wiki/Zero-knowledge_proof
[2] http://www.yubico.com/yubikey