[messaging] Namecoin

Joseph Bonneau jbonneau at gmail.com
Fri Aug 22 08:21:04 PDT 2014


Reading Elijah's comments a second time made me realize for the first time
the distinction between decentralized consensus and decentralized
allocation. Decentralized consensus may indeed work-it appears to work well
enough for Bitcoin to function and a Namecoin-like system with more careful
engineering might work and provide efficient proofs of ownership. There
*might* also be decentralized consensus mechanisms

Decentralized allocation may be the deep problem. As Elijah said, people
care about their names, and a name system can enable "human-readable" names
in the sense of being a reasonable small number of UTF-8 characters doesn't
automatically mean the names are human-meaningful, memorable, or acceptable.

Perhaps we can improve on Namecoin here with some sort of automatic auction
mechanism (where the funds go is up for debate) so that you can't just
claim "cocacola" for peanuts without others getting the chance to outbid
you within a period of time. But I'm not sure this works either-I still
suspect speculators will bid up the common names before corporations,
celebrities, etc. get on board to bid. DNS (via the legal system) has
important brakes on speculating, but you could argue even further that
naming systems like Facebook and Twitter (or more indirectly, Google
search) have been even more popular than DNS and largely supplanted it in
people's minds because they have really strong enforcement against
squatters. These are inherently centralized, human-in-the-loop "I know it
when I see it" systems meaning I highly doubt they be effectively
decentralized.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://moderncrypto.org/mail-archive/messaging/attachments/20140822/09595d08/attachment.html>


More information about the Messaging mailing list