gnunet-developers
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [GNUnet-developers] Reverse resolution of VPN/GNS


From: Christian Grothoff
Subject: Re: [GNUnet-developers] Reverse resolution of VPN/GNS
Date: Fri, 4 Nov 2016 19:00:29 +0100
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101 Icedove/45.4.0

On 11/04/2016 05:58 PM, carlo von lynX wrote:
> This summer I reported https://gnunet.org/bugs/view.php?id=4625
>
> Apparently this has sparked an exciting philosophical debate on 
> social graph reverse resolution:
> https://gnunet.org/gns-reverse-ideas

Actually, in my mind these are similar, maybe related, but not identical
features.  And, I'm not sure it is correct to say that this was sparked
by #4625, I think they came about independently, and I'm thinking that
conflating them is likely the wrong move.

> I would please ask you to come down from your ivory towers for the 
> following reasons:
> 
> 1. Such a reverse resolution method *would* be a local operation if
> gnunet-social and gnunet-psycstore were actually functional and all
> the appropriate subscriptions in place. In other words: You are
> re-inventing secushare.

Definitively not. Neither the psycstore nor the existing
namestore/namecache have the operation needed for what Martin is
thinking about on the blog.  However, the operation *you* are thinking
about would be local.

> 2. In that blog post you are discussing a "public" social graph like
> PGP's. That is a not exactly futuristic idea and very much inferior
> privacy-wise to the private social graph planned by secushare.

Yes, but some relationships are simply public. Why not exploit that? We
are not proposing that the entire graph be public, but that the user can
choose (non-public being the default, of course).

> 3. To make GNS work with existing applications I simply asked to 
> teach gnunet-exit to return the same names that were used by 
> gnunet-vpn to build those tunnels in the first place. The rest of the
> challenge is then dealt by secushare's pubsub structures.

Yes, I understood your point. Martin wants to do something much larger,
which on top of that likely does NOT even address your goal.  So these
are simply not the same, even though both fall into the category of
"reverse" resolution/lookup.

> So all we need to move forward is:
> 
> 1. The closing of that feature request by implementing just the 
> resolution of *known* addresses, in a simple and fast way. 2. Fixing
> the bugs in gnunet-social or anything below so that we can avoid
> having to use older software just because it works.
> 
> Thanks a lot for the recent fixes in CADET, Bart. Haven't tried out
> if they magically get everything working again, yet, but I am 
> hopeful. Who knows, maybe gnunet-social starts working.

On my system, the CADET test suite still fails, so I'm not too
optimistic :-(.

Attachment: signature.asc
Description: OpenPGP digital signature


reply via email to

[Prev in Thread] Current Thread [Next in Thread]