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

[freehaven-dev] Re: [Freenet-chat] MojoNation



On Tue, Aug 01, 2000 at 03:19:07PM +0100, Ian Clarke wrote:
> I think that mojonation is mainly concerned with sharing and being
> rewarded for sharing resources such as bandwidth, processor power, and
> disk-space.  It is a nice idea, in theory, although bandwidth, processor
> power, and disk-space are all getting cheaper by the minute, and there
> are very few applications which would require large amounts of
> processor-power over a slow bus (rendering 3d films might be one).
> 
> I feel that since bandwidth, processor power, and disk space are now so
> cheap, I don't think it is really nescessary to reward people
> financially (just look at distributed.net or SETI at home - neither of
> which offer to pay anything).  Freenet will operate due to people's good
> will, and desire to have a local cache which will speed up accesses of
> information.
> 
> Ian.
 
I think you're misinterpreting the rationale for mojonation's 'economy'.

One of the issues common to each of these publishing services is that we
need to solve the accountability problem: we need to make sure people
don't flood the system and use up all of our resources. The resource most
commonly addressed is space, but other resources to address include
bandwidth between servers and clients, and bandwidth between servers and
servers.

Projects take many different approaches to solving this problem:

* Freenet dumps unpopular data on the floor, so people flooding the system
  with unpopular data are ultimately ignored.
* Gnutella doesn't 'publish' your documents anywhere except on your
  computer, so there's no way you can flood other systems.
* Publius limits the submission size to 100k and hopes that nobody
  will fill them with garbage (it remains to be seen how successful this
  will be -- it sounds dubious, doesn't it).
* Free Haven requires publishers to provide reliable space of their own
  if they want to insert documents into the system. This economy of
  reputation tries to ensure that people donate to the system in
  proportion to how much space they use.

The micropayments scheme in Mojonation is designed to make sure that
users can't get `too much' without providing something in return. If they
tweak the parameters correctly, normal users will always have enough
mojo to do normal behavior (browse a while, pick your file, download
it, repeat). But if they try to attack the infrastructure (for example,
by doing repeated high-bandwidth reads or writes) then they will find
that in order to sustain the attack they must "pay" for it by donating
comparable resources of their own. (Freenet also addresses this in part,
by caching popular data near the attacker, so he can't attack the entire
network "very quickly".)

(And at the same time as addressing availability issues like the above,
there's a whole new class of anonymity issues -- but that's a different
story.)

I spoke with Jim McCoy a couple days before their presentation at DefCon;
the 'financial rewards' you mention (converting mojo into US currency)
are something they might do in the future if the basic architecture works
out. It's not integral to the system -- the basic Mojonation design does
not interact with any sort of external currency.

Anyway, I think your analysis above criticizes something not really
important to the system, and trivializes some of its really cool aspects.

--Roger

PGP signature