> There has been a long-standing assumption that if you exercised a path in the > (recent) past you can probably assume that the properties haven't changed > much if you want to start exercising it again. This is why heuristics like > caching path properties (RTTs, etc.) are often of benefit - often, but not > always, and maybe never in some scenarios (e.g., overcommitted CGNs.)
This is rough assumption for WiFi. I just measured one laptop on a WiFi n network going from no service to 1Mb/s to 10Mb/s moving only a few meters, and then 100Mb/s moving some more. You can rapidly transition between a network being available and not available, and if you do not handle some heuristics, you can get some bad artefacts. Mobile can of course also vary a lot when going from 4G to 3G to no signal in a tunnel on a train. Mikkel
