morgan@JESSICA.STANFORD.EDU (02/09/90)
Wolfgang Naegeli <Wolfgang_Naegeli.ED_TSRS@qm01.ctd.ornl.gov> writes: > If not worthless, the Kinetics boxes still seem to be overpriced, even > if they were an order of magnitude more reliable than they have been. You know, I always used to look at them and think "damn, I could build one of these for a couple hunnerd bucks. What a ripoff." Evidence to the contrary, I suppose, is that now there are several such hardware devices on the market (Gatorbox, Shiva EtherGate, Webster Multigate) and those I am familiar with (Gator and Shiva) are about the same price as a KFP. If it were really that easy to build a cheap LocalTalk/Ethernet/AppleTalk/IP gateway, somebody in our wonderful free-enterprise world would have done it. Maybe the Hungarians *8^)*. In terms of KFPs in particular, I think they were much more of a burn when we had to run university-written software in them. When I think of the complexity of writing and maintaining a program like K-Star, it seems more reasonable to think of the package as a $1K piece of hardware and a $1K piece of software (university prices *8^)*. Remember also that as many as we see around us (entirely too many around here) the market for these things is still pretty limited, if only because they're so damn hard to set up. - RL "Bob" Morgan Networking Systems Stanford
roy@phri.nyu.edu (Roy Smith) (02/10/90)
> If not worthless, the Kinetics boxes still seem to be overpriced, even > if they were an order of magnitude more reliable than they have been. They may be overpriced, they may be a real bitch to configure, and, at least when we got ours, had to rely on OPS (Other Peoples' Software) to work, but I don't think you can knock on them for reliability. After our box got over its infant failure mode (bad power supply), it's been rock steady. We've had our KFPS-2 for almost 3 years. Other than when we had a power failure, I can remember having to reboot it once. It's basicly the most reliable piece of equipment we have. -- Roy Smith, Public Health Research Institute 455 First Avenue, New York, NY 10016 roy@alanine.phri.nyu.edu -OR- {att,philabs,cmcl2,rutgers,hombre}!phri!roy "My karma ran over my dogma"
veizades@apple.com (John Veizades) (02/13/90)
MacTCP 1.0.1 will be available shortly (it is going final at the end of this month) and will be updated to all license holders. It is now in beta test at a number of sites and solves many of the problems described in this list over the last week. John Veizades... MacTCP engineer.
rapatel@pilot.njin.net ( Rakesh Patel) (02/13/90)
Are there any plans to support SLIP w/header compression or PPP in future releases of MacTCP? Rakesh Patel.
urlichs@smurf.ira.uka.de (02/15/90)
In comp.protocols.appletalk, <Feb.12.21.05.50.1990.16054@pilot.njin.net>,
rapatel@pilot.njin.net ( Rakesh Patel) writes:
<
< Are there any plans to support SLIP w/header compression or PPP in
< future releases of MacTCP?
<
Please define "support". :-(
MacTCP includes code to support alternate link-level interfaces.
"Of course", the interface betwixt MacTCP and its low-level code is totally
undocumented. Apple!?
--
Matthias Urlichsveizades@apple.com (John Veizades) (02/16/90)
In article <1532@smurf.ira.uka.de> urlichs@smurf.ira.uka.de writes: > MacTCP includes code to support alternate link-level interfaces. > "Of course", the interface betwixt MacTCP and its low-level code is totally > undocumented. Apple!? > > -- > Matthias Urlichs There is a forth coming document that will desribe how to write alternate link access protocol moduals for MacTCP (SLIP, PPP, Lanstar, Arcnet, etc.). I should have it finished soon, working out some of the last bugs, and it will be available as a tech note. More news when there is news on this one. John Veizades... Apple Computer, Inc.