Cell Relay Archive[Date Prev][Date Next][Thread Prev][Thread Next] [Date Index][Thread Index][Author Index][Subject Index] Re: A lesson from the ATM/IP wars
In article <383B3277.FB67E7F1@cs.wustl.edu>, Jonathan Turner <jst@cs.wustl.edu> wrote: > I don't think it's the end systems that are hard > to change. In the approach I describe, all you need > to do to the PCs in typical corporate environments > is install some additional software. So long as it's > appropriately shrink-wrapped or downloaded from the > net, this should not be a big obstacle. Even installing > a NIC isn't a big deal. The bigger problem is > the connectivity between a PC and an ATM switch. > But in the proposed scheme, that isn't essential. I have to say that I'm in complete agreement with Mike Heard and Paul Koning on this matter. Changing applications that work today to add useful QoS knobs is probably out of the question. To create new applications that require fundamentally different QoS knobs, such as those provided by ATM vs IP enhancements, is probably also not easy. On the other hand, to provide an infrastructure that can carry IP enhanced QoS _better_ than connectionless datagram infrastructure can do _might_ be one good avenue. And another would be, as I proposed before, for telcos to sell new services that are independent of IP, but that could make use of the same infrastructure. -- Bert manfredi@arl.bna.boeing.com Sent via Deja.com http://www.deja.com/ Before you buy.
|
|