On 02/04/2011 01:46 PM, Serguei Osokine wrote:
> Most certainly. All I'm saying that the system that "offers to set up
> a DHT or even, ad hoc mesh network -- or even a "sneakernet" -- if it
> detects the internet has stopped functioning" most likely won't be
> used by anyone if it does not offer any clear advantages even when
> the Internet is present. People would deploy it only if they can use
> it to stream movies or something even when the Internet is working.

I think we agree: it needs to be *at least* as valuable as any 
internet-based alternative, when the internet is available.  So if it's 
a mass-communication tool like twitter, for example, I think it should 
post to the *real* twitter when possible.


> Besides, if you won't excercise your mesh code even under the normal
> conditions, it most likely simply won't work when you'll need it to.

This is a reasonable concern; I'd solve it by doing everything through 
the mesh *and* the internet, such that both are constantly used. 
However, given that the internet is generally a superset of the mesh, I 
don't think most people would even realize the mesh was there until 
internet dies and everything miraculously keeps working.

For example, imagine you and I were both using this tool.  We're each 
able to tweet to the world using the tool via the internet and the world 
gets it (even those who aren't using the tool).  The internet dies, and 
you and I are still able to tweet to each other -- to you and me it 
looks like nothing has happened -- even though for some reason the rest 
of the world is silent.

So this mesh-tweet feature would be continuously used even when we have 
the internet, but odds are the *real* twitter is actually faster such 
that by the time you get my mesh-transferred tweet, you've already got 
it from the real twitter.


I think this "use the internet when you can, use the mesh always" 
approach can probably be extended to BlueTooth mesh and sneakernet as 
well.  And for some applications, especially piracy, those methods might 
actually be superior to the internet.  But the user needn't be involved 
in the decision: maybe it'll be faster to tweet to you over bluetooth 
than the Internet if we're right next to each other, maybe not.  The 
user shouldn't pick, it should just try both routes and the first wins.


> In fact, in places like airplanes and beaches you might find yourself
> pleasantly surprised even long before the Internet stops functioning.
> Sure, it will also help demonstrators to organize flashmobs in Upper
> Lampusia - but only if they will already have it working by the moment
> when they will actually need it.

That's a good point -- if the mesh is always on (even if not really 
needed), then it'll already be tested and ready to roll in the event the 
internet temporarily becomes unavailable for more mundane reasons than 
mass protest.

-david
_______________________________________________
p2p-hackers mailing list
[email protected]
http://lists.zooko.com/mailman/listinfo/p2p-hackers

Reply via email to