>Isn't there some way to fix the routing daemons to take this into
>account?   The other side of what you're talking about is that if I
>have, for example, an ethernet card that I use when my laptop computer
>is at my desk, and a ricochet modem that I use when it's not, it would
>be nice if the kernel could tell, at the network level, when the card
>has been unplugged, and stop using any routes that go through that
>card.   I'd like the kernel to notice when the card is again plugged
>in, possibly a long time later, and do the right thing then too.   

Ted,

The mosquitonet project has done exactly that, on Linux, on top of
David Hinds' PCMCIA patches, for more years than I can remember.
(the mobile routing is a bit more iffy.)


Isnt about time we caught up?

>It seems to me that it's a lot easier to fix the daemons to behave
>reasonably on inserts and removals than it is to teach the kernel to
>behave reasonably if inserts and removals don't detach interfaces.

I'd agree, except Linux seems to manage fine, and they do throw the
state away.  (Maybe its becuase they bind the device name, in the
userlevel daemon, via the tuple space and/or MAC address?)