[Noisebridge-discuss] Distributed computing and storage (with some major caveats)

Sai Emrys noisebridge at saizai.com
Sat Jun 27 05:25:26 UTC 2009


AFAICT, PIR is only intended to secure the *key* being looked up. Per
the linked page http://www.cs.umd.edu/~gasarch/pir/pir.html:

"Alice wants to query a database but she does not want the database to
learn what she is querying. She can ask for the entire database. Can
she get her query answered with less communication? We model a
database as (1) an n-bit string x=x1 x2 ... xn, together with (2) a
computational agent that can do computations based on both x and
queries made to it. Alice wants to obtain xi such that the database
does not learn i. "

That would be nice, but I think is unnecessary, given that we will
presumably already be ensuring that only the node(s) responsible for
responding to a particular query are able to read the query, and that
in itself may already be sufficient to discover the query's identity
under PIR (as PIR, AFIACT, depends on multiple copies of the db rather
than on a sharded/distributed db).

My interest here (though I think it's slippable, i.e. not part of my
core reqs) is in searching for xi where *x* is encrypted.

- Sai



More information about the Noisebridge-discuss mailing list