[OLPC Networking] Re: [olpc-software] "Mesh" Software

[Aaron Kaplan]

On Thu, 2006-04-20 at 10:53 +0200, aaron wrote:
> On Wed, Apr 19, 2006 at 10:54:24PM -0300, Jim Gettys wrote:
> > Aaron,
> > 
> > Are you willing to mentor someone for the summer of code?
> > 
> > I signed OLPC up for the program; the issue is time/cycles to mentor the
> > summer students.
> >                                   Regards,
> >                                       - Jim
> 
> I can try  to be a good mentor :)
> 

ok, I wrote myself in the wiki page as contact for meshing and google
summer of code.


> 
> > 
> > 
> > On Wed, 2006-04-19 at 12:25 +0200, Aaron Kaplan wrote:
> > > I can propose to use the OLSR routing daemon. It provides quite nicely
> > > for a number of meshes such as freifunk.net, funkfeuer.at (map at
> > > http://sandwich.funkfeuer.at/googletopo ). OLSR.org took a very solid
> > > approach: it integrated developments of other routing protocols (hazy
> > > link state routing etc) on a as-needed basis. It works in practice.
> > > 
> > > 
> > > How about a google sommer of code project "adapting OLSR to the needs of
> > > olpc"? Several things come to mind:
> > >   - IP autoconfig
> > >   - autoservice discovery 
> > >   - autosync 
> > > etc
> > > 
> > > best regards,
> > > aaron.
> > > %
> > > funkfeuer.at community wireless mesh, vienna
> > > 
> > > 
> > > 
> > > On Wed, 2006-04-19 at 06:10 -0400, Ivan Krstic wrote:
> > > > Fabian Deutsch wrote:
> > > > > Maybe a bit out of date? I mean, there has been a lot of movement since
> > > > > 1999 ... especially in this (mesh) area ..
> > > > 
> > > > The paper is perfectly recent enough to illustrate that there are,
> > > > obviously, more than 3 substantially different routing protocols
> > > > available, which was the point I was making.
> > > > 
> > > > I wouldn't call most any of the general-purpose mesh routing work that
> > > > I've seen since that paper was published groundbreaking. A possible
> > > > exception is Biswas' and Morris' opportunistic routing (ExOR), but I
> > > > don't think that'll be of much good given our use case.
> > > > 
> > > 
> > -- 
> > Jim Gettys
> > One Laptop Per Child
> > 
> > 
>