On Fri, 18 Feb 2011, Eugen Leitl wrote:
On Thu, Feb 17, 2011 at 03:22:49PM -0600, J.A. Terranson wrote:
I'm tired of hearing the current calls for "mesh networks" myself. As you point out, they are *incredibly* complex, and these "wall-wart servers" aren't going to implement that kind of thing IMO. Christ, large ISPs
I fail to see where they're mentioning the word mesh at all. If I would do it, I would just package Tor, Tahoe-LAFS, I2P and maybe a couple other select goodies (tinc, opportunistic encryption, whatever), and put it on the residential Internet.
You're right - they did not specifically mention "mesh", which probably means this individual is operating even one level of abstraction higher .. in the realm of "I know computers can do wonderful things". ToR in this context doesn't make much sense to me - it's a network designed to run over the public Internet, and the whole point of his "dream" is to get out from under the thumb of the public internet. Did I misread, and he thinks that a bunch of low power micro devices, as nodes on the plain old Internet, are going to be helpful in any way ? Tahoe-LAFS makes a bit more sense, as it makes content available without shackling any one system (operator) with liability for that content.
If you recall I repeatedly suggested that here and elsewhere, to the overwhelming sound of crickets.
Well, if you're suggesting that I have little to contribute, you're right. My interests float into this area, and I've lurked long enough on BATMAN and MANET and OLSR to know when to call bullshit, but that's about it. Make no mistake - I agree with the sentiment that this is possible, and that it's coming, and I will be very happy when/if it does ... I just think some dose of reality is needed for the boingboing set of "technologists".
still have trouble with simple IP, and they have "trained [or] experienced 'engineers'". It's a silly call to arms at this point.
If you're talking mesh, one of the basic requirements is to get rid of central address allocation authority and establish a local-knowledge based routing. If you get that far, the human operator mistakes are removed from the loop.
I was with a group that tried to work out a mesh implementation across a relatively small (~15sq miles) area, and it never came to fruition, despite several years of work on the problems presented.
Do you have a pointer to description of your project, and what went wrong?
I think the BATMAN people have some long, detailed summaries of successful implementations in places like .de and .it and so on. By successful, I mean, they got packets to flow. The deeper issues of establishing everything with no centralized authority, etc., are unsolved.
I can tell you one thing: Eben Moglen is not a stupid man, and he's surrounded with technically capable people. If he can raise the money for 10-100 k wall wart units you can assume people will do useful things with them.
That's good news - better that someone is pushing in this direction than not...