| However, that restriction might not be feasible; e.g.
| imagine a car rolling down a highway, connecting up to different wireless
| ISP's as it goes along.
Inside the car are a bunch of individual computing elements -- a PDA,
a smart keychain, a laptop, the digital audio player, ...
All of them move as a unit, unless one falls out of the car
and is left behind. Consider taxis, or one or more passengers
hopping out in front of the restaurant while the driver hunts
around for a parking spot.
Landmarks come and go both for individual computing elements,
and for clusters of them -- the car driving around, a site
site switching from one ISP to another, you name it.
As Avri noted, we have evolved ways of dealing with many of
these problems without any sort of architecture. There are
obvious drawbacks to this approach, but the fact that it
can be done is a useful and important feature of the collection
of protocols and practices in the current I* routing architecture.
| If you don't want to live with that restictions (limited network mobility),
| then I would tend to think that perhaps mobile networks will need support from
| the overall architecture (e.g. in the form of a "moblile network" object which
| can group a number of endpoints together, among other properties).
Mobility is fact -- I don't think there's any piece of the
Internet which does not move (perhaps in slow motion) with
respect to the remainder of the topology. It's all relative.
| My crystal ball has clouded over.
Nobody said this stuff would be easy. :-)
Sean.
This archive was generated by hypermail 2b29 : Mon Aug 04 2003 - 04:10:04 EDT