Re: Nontrivial capability graphs with mistrusting parties

Marc Stiegler <[email protected]> Mon, 4 May 2015 10:51:39 -0700
Newsgroups gmane.comp.capabilities.general
Message-ID <CAK=cCVXQrJ-T_v5pSJbkhn5Mbz_B11GHPzbB6-_fH4sueQh45A@mail.gmail.com>
So, this has been a topic of conversation for some of us for over a decade.
A couple of us even did a startup company to be the third party to host
this interaction :-)

AFAIK,the only general-purpose solution that has arisen since then that
offers an elimination of the trusted third party is the bitcoin blockchain.
But one of its obvious limitations is, it is the opposite of secret,
everything in the blockchain is published to the universe :-)

--marcs

On Mon, Apr 27, 2015 at 12:00 PM, Anton Burtsev <[email protected]>
wrote:

> Hi all,
>
> I would like to create relatively complex communication graphs of in
> face of multiple mistrusting parties (edges are capabilities, nodes are
> capability objects).
>
> For example, both A and B have secrets on which they want to set up a
> joint computation in such a way that none of them can access results of
> a joint computation without being accounted and declassified by another
> party first. In other words, they want to construct an object graph
> like:
>
>  Secret A -\                      / - Accounting/declass B - A
>             - Joint Computation -
>  Secret B -/                      \ - Accounting/declass A - B
>
> One might say that there is a trusted third party that creates such
> graph, but maybe there is a more elegant solution that doesn't require a
> third party?
>
> Thank you,
> Anton
>
> _______________________________________________
> cap-talk mailing list
> [email protected]
> http://www.eros-os.org/mailman/listinfo/cap-talk
>

_______________________________________________
cap-talk mailing list
[email protected]
http://www.eros-os.org/mailman/listinfo/cap-talk