[serval-project-dev] Re: Serval Mesh

[Paul Wolstenholme]

Congratulations to the team on all the good work done to date.

Paul Gardner-Stephen wrote "we also have ideas about how to provide
financial incentive for  users to participate fully in the mesh, but I
am not in a position to describe that fully right  now."
I have been musing about exactly that issue (encouraging fair
participation in a mesh network)  for some time.  This therefore seems
to be a good time to introduce myself to the group and  offer an
alternative point of view:

My background is that of an electrical engineer involved in
communications and control systems,  including product design.  Don't
let this first posting fool you. I have no political leanings  or
agenda and don't believe my suggestions here would apply to human
society - just to a  'society' of meshed nodes...

When designing a communication protocol, it is convenient to draw from
and adapt our existing  experience.  I prefer to think of how
cooperative people might interact with each other as a  model for how
mesh nodes might interact cooperatively.  The use of currency in our
society is a  very familiar concept and we understand how it can
encourage a degree of fairness in how we interact.  Paul G-S has
already noted some of the difficulties in implementing a currency
electronically.  I would like to add my observation that the use of
currency doesn't result in  the fair sharing of resources anyway since
it results in very rich and very poor people.  Perhaps we should  be
looking for a better solution both in terms of ease of implementation
and the ability to  achieve the desired outcome.

So what traits of cooperative people can be applied to mesh nodes?
Some people like to give away a proportion of what they have or
receive for the purpose of helping others.  The open source community
who give the fruits of their spare-time labours for all to use are
certainly an example.  Some believe that new arrivals from other
places should be considered worthy recipients for such generosity.
Indeed, the open source community is known for welcoming  newcomers
and investing time in helping them learn to cooperate within the
community.  If mesh network nodes behaved in the same sort of way, we
start to think of nodes that roam to a new network as a guest to
welcome rather than a foreigner who shouldn't participate until his
currency is converted.

If we accept the concept of mesh nodes giving some service to other
nodes for free, we are free to look at payment for service in a
different way.  By giving service (such as relaying  messages) without
agreeing a price and demanding verifiable payment beforehand, I expect
the protocol could be simplified.  That is not to say there is no
payment.  It is fair that the recipient of service offer a similar
amount of service in return.  Users that do play fairly in all
respects increase the trust that the service provider has in the
service user.  It would be fair for a node to save most of its
resource (bandwidth or battery energy) to serve nodes it trusts.  The
resources a node gives away without being paid in return could be
limited to a  fraction of the available resource.

So the next question is as to what should constitute payment.  For
communication within a mesh network, a promise to relay messages
should be payment enough for messages relayed.   But how long does the
promise need to be valid for?  Traditional currency holds its value
indefinitely while slowly losing value through inflation - all of
which is difficult to model electronically.  Realistically, a node
that promises to relay messages isn't necessarily capable of keeping
such promises for longer than the charge in its battery.  Perhaps all
such promises should be time-limited and expire if not used.  I
believe the time limit will help with record keeping - although I'm
not sure how long the time limit should be.  The more enduring record,
however, should be the level of trust that one node has for another
following many successful transactions.

If the nodes in our mesh network behaved in this way, it would
encourage a system where high users of communication traffic (meaning
nodes sending lots of messages or sending messages over  many hops)
need to provide high levels of service to others in order to be
trusted enough to get the service they want.  If trust is built
slowly, greed would be discouraged in the long term.  This would not
stop a highly trusted node from successsfuly requesting high levels
of  service on the odd ocassion when it is required.

I could go on with musings about sharing credit (or promises of
service) with those that are  highly trusted or giving it away before
it expires, but these are just thoughts for now.   Proper modelling
and implementation are beyond what I can envisage so far.

I've never claimed to be good at presenting my ideas but I would be
interested in hearing the shortcomings in this concept - and how they
might be overcome.  Personally, I think using the concept of trust
means we can get away with short-term promises of service rather than
standardising a currency.

-- 
You received this message because you are subscribed to the Google Groups "Serval Project Developers" group.
To post to this group, send email to serval-project-developers at googlegroups.com.
To unsubscribe from this group, send email to serval-project-developers+unsubscribe at googlegroups.com.
For more options, visit this group at http://groups.google.com/group/serval-project-developers?hl=en.

----- End forwarded message -----
-- 
Eugen* Leitl <a href="http://leitl.org">leitl</a> http://leitl.org
______________________________________________________________
ICBM: 48.07100, 11.36820 http://www.ativel.com http://postbiota.org
8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE