>> Fred >> [email protected] >> Ideally, such mechanisms should completely decouple mobility from routing. >> >> FLT >> Not OK. It should be perfectly OK for mobility to interact FLT >> >> with the routing system as long as the routing churn is FLT >> >> localized and minimized. Strike this sentence.
>[WES] I disagree with your recommendation here Fred, but for purely semantic >reasons. >http://wordnetweb.princeton.edu/perl/webwn?s=ideal >Absent any other considerations, that is, ideally, the proper solution is to >decouple >mobility from routing. In a more practical implementation (ie not a completely >ideal one), >it might be ok for mobility to interact as you are saying, but I don't think >that the >sentence as written says that this is prohibited, only that it's not the most >preferable implementation. If one spends time in MANET networks as I do, mobility and routing are the same thing. By contrast, if one spends time in so-called mobile networks, the mobile entity moves in reference to a comparatively stable infrastructure. In the latter case, communications latencies are reduced for the moving entity if a feedback mechanism exists to update the routing infrastructure to a more current point of attachment. However, as Fred Templin stated, this needs to be done in such a manner that it does not impact routing convergence. _______________________________________________ rrg mailing list [email protected] http://www.irtf.org/mailman/listinfo/rrg
