[p2p-research] P2P Email

[M. Fioretti]

I said I would not comment on Marc's alternative to email and I
confirm it. It may very well be the best thing since slided bread for
all I know, but it would need much more time than I could afford now
to make an informed opinion.

However, there are one or two things I feel I should make clear about
the current system and the way I and others use it, just for the
record.

On Sat, May 09, 2009 06:39:12 AM -0700, marc fawzi wrote:

> my experience with setting up my own SMTP/POP server on my own
> machine involved configuring my router/NAT and firewall which is
> beyond what the average person is willing to do.

when I decided to go the opposite way of Gmail and similar, that is to
take my online communication in my own hands and contribute to a more
decentralized email infrastructure asap, I didn't stop for one second
to check how difficult it would be to configure my router, NAT,
firewall and so on, on my own home computer, for at least three
reasons, in more or less increasing order of importance:

1) I was sure that if I could sort out something like SMTP and
   POP/IMAP server configuration (which I eventually did) I could also
   work out without problems that kind of stuff, at least in the case
   of my own home configuration

2) laws and market conditions: today there is no guarantee, in mine
   and many other countries, for people who want to run servers or
   services of any kind directly from their own home. If you have a
   private/residential contract for Internet access, the ISP can close
   any port they want (and the most frequent case is just the outbound
   email port) if they feel your usage is outside their "average Joe
   at home" profile, and there is nothing in the contract that forbids
   that. To have legally valid guarantees that this won't happen you
   must subscribe much more expensive contracts.

3) reliability. Even if my ISP had come home to configure my router,
   nat, firewall etc, gratis for me, I would have still gone for the
   "virtual server in datacenter" solution, for all the reasons I've
   already described.

Marc's vision may turn up the perfect email replacement in the next
decades, I have no idea.  I was just very puzzled to realize that he
got the idea, or so it seems, from struggling about what was (and
would still be, if I still needed a personal solution today) to me a
non-existant issue, for all the reasons above.

> So my argument with Marco was not to say that inter-mediation is
> bad, but to say that the solution he uses can never be pervasive and
> thus can never be "true p2p" ...

does **everything** we need *have* to be "true p2p"? is it convenient?
In a sense, this is the same issue discussed in case 2 of
http://p2pfoundation.net/Thoughts_on_P2P_production_and_deployment_of_physical_objects

and all this discussion was another example of confusion between
Research and Development as explained in the same page (nobody's
fault, of course). I developed a working solution with existing
tools, while what Marc talks about is more in the research area.

marco
-- 
Your own civil rights and the quality of your life heavily depend on how
software is used *around* you:            http://digifreedom.net/node/84