[OTR-dev] Fwd: Ensuring transcript soundness in a multiparty chat setting

[Ximin Luo]

On 17/12/13 00:19, Trevor Perrin wrote:
> We should all pause and read this paper:
> 
> http://research.microsoft.com/en-us/people/mickens/thesaddestmoment.pdf
> 

Haha, yes I saw this on Hacker News too.

However, I'm pretty sure that (the consistency part of) what we're talking
about is emphatically *not* the Byzantine problem. (I mentioned this in an
earlier email but it probably got lost with the rest of what I wrote.)

What makes the Byzantine problem so hard is that the parties cannot make a
commitment until they are sure everyone else has made the same commitment.

By contrast, we (and git) do *not* have this problem. When A commits a new
message M on top of parents (P[i]), this is *independent* of everyone else and
is a firm commitment that will *never change in the future*.

Likewise, when I receive message M from A with parents P[i], I am *certain*
that A wrote this message and that they have seen the messages P[i], and no
future information can possibly change my knowledge regarding this matter.

(These properties are why I believe my "agreement" protocols mentioned
previously work, but no-one has reviewed them yet, and why this
periodic-broadcasting eventually achieves consistency.)

X

-- 
GPG: 4096R/1318EFAC5FBBDBCE
git://github.com/infinity0/pubkeys.git

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 897 bytes
Desc: OpenPGP digital signature
URL: <http://lists.cypherpunks.ca/pipermail/otr-dev/attachments/20131217/b3400912/attachment.pgp>