[messaging] Namecoin

[Tony Arcieri]

I remember staring at Zooko's triangle in the days before Bitcoin and
thinking it was unsolvable, and in that regard, Namecoin was a bit of a
revelation. I had a bit of an "a-ha" moment reading Aaron Swartz's original
post about squaring Zooko's triangle. I'm not sure this approach is the
most practical solution to the problem though.

Perhaps the most common complaint about a system like Namecoin is that it's
immensely inefficient. I don't think that's the biggest problem though.

Key compromise is the biggest worry to me. If you lose a private key
associated with a name, control of the name is irrevocably controlled to
the attacker. Other systems like OpenPGP might answer this with revocation
messages which are created at the same time as the key. I'm not sure if
Namecoin has anything like that.

In the event of key compromise, an attacker can confuse the senders of a
messages to encrypt them under an attacker-controlled key. Hopefully the
original owner of the name can notice this and warn potential message
senders that the private key associated with their original name has been
compromised, and, AFAICT, adopt a new name and tell people to look them up
under that. However, Namecoin as it exists today seems to provide no
solutions to handle key compromise and notifying recipients that a key has
been compromised.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://moderncrypto.org/mail-archive/messaging/attachments/20140820/1e922751/attachment.html>