[tor-onions] Presentation on Onion Networking at the BCS

[Zenaan Harkness]

On Mon, Jul 22, 2019 at 08:55:32PM -0400, grarpamp wrote:
> On 7/22/19, Alec Muffett <alec.muffett at gmail.com> wrote:
> > "Why & How you should start using Onion Networking"
> > https://www.youtube.com/watch?v=pebRZyg_bh8
> 
> A fine introduction.
> 
> Yet how do people, including those involved with or using other
> projects in the space, compare contrast and evaluate this with
> "Why and how start using" and writing for... Onion, I2P, CJDNS,
> MaidSafe, IPFS and all the other overlay networks out there
> and forthcoming, all in their respective "non-exit" modes?
> 
> Whether it be for protocol layer capabilities HTTPS/TCP/UDP/IPv6,
> or to achieve application layer... messaging, storage, web-ish, etc.
> 
> And how does each's lack or presence of whatever API
> interfaces, UDP, broadcast, name layers, or other potential
> transport and programming models, lend themselves to app
> development and widespread eventual adoption and use?
> 
> And how, without offering IPv6 or the ultimately better all
> encompassingly wide and modular, even cryptographic,
> AF_OVERLAY interface that all networks could plug into,
> does anyone expect to get everything interoperable and
> working together?
> 
> 
> [Note that comparing "traction" re all other nets
> accessing facebook is false since those nets simply
> do not offer a simple exit mode to do so as tor does.
> What would be fair is if facebook had CJDNS, I2P, Onion,
> etc interfaces, and then comparing those access stats,
> scaled relative to each respective project estimates of
> number of users, project advertising funding impact,
> project *Browser availability, etc.]


A primary foundation is a "trust based" underlying network, like a
friend to friend (F2F) style network - just a simple IP/ETH packet
delivering low layer.

If friend A is physically close, your F2F link to that friend should
be physical;
if not, an encrypted link tunnel to that friend is created.

"Friend" ~= someone who is unlikely to sell you out to govcorp
            should you exercise your ('absolute right' to)
            freedom of speech

With no friends, expect to achieve at best ephemeral access to the
world's information in any anonymous way.

Just above this base layer is chaff fill ≡» such things must be
configurable since some folks will not pay the price, and others will
pay it sometimes, and others still may just want a low-level (e.g.
2 KiB/s) chaff link to cover high-latency low-bandwidth comms, e.g. .

With this base layer in place, an onion/ I2P routing layer can
readily be laid over this - but IP/UDP only of course.


Without the lowest layer done right, there shall continue to be
endless "new" network designs.


Once a sane base level network is readily installable/ configurable,
the next big concept is p2p distributed identity - DHTs, public and
private keys for identifying an entity, website, "name" etc.

And it ought go without saying that any identity implementation which
provides something other than 100% control to the end user, is doomed
to fail - the GNS or GNU Name System is good for thought in this
regard, where the individual is the authority for each target name in
his GNS (DNS like) local name database.

Delegation of authority can be built in, but again, any delegation
must be solely within the hands of the end user or end node; fail on
this, and watch the roll out of "new" systems replace your own
implementation.


Without the fundamental (sane) infrastructure in place, we are doomed
to an endless series of "new" "distributed" "global" interplanetary"
"solutions" ad nauseum.

Once the fundamental sane infrustructure IS in place, then IPFS, GNS,
GitTorrent
https://blog.printf.net/articles/2015/05/29/announcing-gittorrent-a-decentralized-github/
and the like will appear self-evidently either ready in large part,
or needing certain improvements, in order to fit in with the sane
foundation infrastructure.

For whatever reason, folks often begin at the high layers and thus
doom themselves to new solutions displacing their work inside a few
short years.

And some of those who would dedicate swaths of their consciousness to
implementing experiments and test cases on the direct path to sane
infrastructure, are tied up handling the most petty and mundane
bullshit, legal, political and otherwise, that humans have devised to
consume the consciousness of men of good character who would
otherwise bring great technological advancement to this world in a
much shorter time than otherwise is the case.

Cest la vie...