[Cryptography] GNU's "anonymous-but-taxable electronic payments system" Heh.

[Jeff Burdges]

On Thu, 2016-06-09 at 10:34 -0700, Ray Dillinger wrote:
> On 06/07/2016 09:14 AM, Phillip Hallam-Baker wrote:
> > This is something that I see as a trend in modern crypto that is 
> > worth reminding people of.
> > 
...
> > 
> > It is very easy for people to look at the legacy crypto applications
> > and assume that those represent timeless design truths. They don't. 
> > Those systems were designed around constraints that no longer exist.
> 
> Honestly, it would be better if software designers hadn't taken that
> "ubiquitous network connectivity" thing to heart.  There is nothing
> better for security and privacy (and in many cases functionality!)
> than avoiding unnecessary connections.

I agree in general.  In Taler's case however, we've a RESTful protocol
without too many rounds that gives parties some flexibility.  

We do not require that customers be online to spend money, only that
merchants must be online to prevent double spending.  In other words,
merchants must contact the exchange before delivering goods, or else
must institute their own identity checking and debt collections. 

At present, I'm unsure exactly how our merchant software behaves when if
the exchange is down.  In principle, it could continue carrying out
transactions, but mark the orders as pending, and then go collect the
coins and mark the orders as paid later when the exchange comes back.
Internally, I've mentioned this as potentially useful but maybe
something breaks currently. 

We do require that customers go online to withdraw or refresh money of
course.  Yet, Taler could be adopted to point-of-sale systems with
offline customers.  We could even batch withdrawal and refresh
operations to occur through a merchant's connection, so like when you
spend money at the corner shop you also get you freshing anonymized
change from the exchange.  Anyone interested in a point-of-sale system
protocol for that is welcome to contact me to chat about it.  :) 


Just an aside where I agree with you :  We need an offline or at least
highly asynchronous alternative to the web.  I'm thinking roughly
Yawning Angel's anti-captcha extension CFC for FireFox / Tor Browser,
but using distributed storage and an asynchronous mix network, instead
of the centralized servers of archive.is and Tor.  I'll clip message to
tor-dev about it and forward it below.  Arguably, the web standard could
be radically trimmed along the way, possibly losing security properties
of web-to-async-web gateways.  I think Taler could be adapted to such a
setting. 

-------- Forwarded Message --------
> From: Yawning Angel <yawning at schwanenlied.me>
> Reply-to: tor-dev at lists.torproject.org
> To: tor-dev at lists.torproject.org
> Subject: [tor-dev] Request for feedback/victims: cfc-0.0.3
> Date: Mon, 18 Apr 2016 22:16:31 +0000
> 
> Thanks for the feedback/bug reports so far.
> 
>   [ PEOPLE THAT HAVE BIG SCARY ADVERSARIES IN THEIR THREAT MODEL
>     STILL SHOULD NOT USE THIS. ]
> 
> This is the third release of the archive.is based captcha-be-gone
> extension for Tor Browser.
> 
...
> 
> Source: https://git.schwanenlied.me/yawning/cfc
> XPI: https://people.torproject.org/~yawning/volatile/cfc-20160418/
> 
> Regards, 
> Yawning Angel

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: This is a digitally signed message part
URL: <http://www.metzdowd.com/pipermail/cryptography/attachments/20160609/8adfb599/attachment.sig>