Hi!
On 14:32 Tue 01 Feb , Henry Sinnreich wrote:
> These are useful thoughts at the p2p and application levels.
> There is more to it however:
>
> * Application level - as below, possibly
> * P2P layer - as below, possibly
> * UDP/IP/Data Link/Physical - is missing and here is what may work:
>
> All users to have at least their cell phones equipped with a global standard
> compliant p2p UA and all 5 + p2p layers of the IP protocol stack. Even if
> all the network infrastructure fails due to "main made" :-) or natural
> disasters, users within near wireless range could still communicate as long
> as their battery will last. If lucky, some of the peers may even have
> Internet connectivity, thus serving as gateway for other users, albeit with
> some congestion.
>
> The key is all mobile phones and other devices to communicate in a global
> standards compliant way. This would require all mobile phone and other
> device manufacturer to include such a global standard compliant capability.
> To specify the standard, at least the IEEE, IETF and W3C would need to
> cooperate for consistency across all the 6 layers, considering p2p as a
> (sub) layer as well.
Something similar might not be that far away ;-) . But some details might
still turn out to be different than you think:
- Cell phones are constantly moving, which makes it harder to find routes than
if you use fixed routers. Batteries will not last even a day. If you have
the electricity to constantly recharge, you will likely also have the
electricity for fixed routers...
- Building meshes based on top of IP is IMHO a big mistake. Even if you would
be able to implement what you want, it would be slow and unstable.
- You may want to prefer using this network, even if the cell phone network is
available.
- Such a network has a wide variety of uses, not just for mobile phones.
> Last but not least, to overcome the pushback from many parties who may not
> like it.
There are many phones where you can build your own firmware images. If there
are fixes routers for forwarding the traffic, this should be enough to make
such an interface useful. If it is useful, manufactorers will either include it
or be obsoleted.
> The most promising approach would be to start such a project in academic
> and/or other R&D organizations and make it publicly available, as has
> happened for the early Internet.
>
> There are some IEEE papers on this topic, but accessible only for pay :-(
Are you sure these standard organisations are your friend?
-Michi
--
programing a layer 3+4 network protocol for mesh networks
see http://michaelblizek.twilightparadox.com
_______________________________________________
p2p-hackers mailing list
p2p-hackers@lists.zooko.com
http://lists.zooko.com/mailman/listinfo/p2p-hackers
