🚀 go-pugleaf

In article <722284839snx@crynwr.com>, nelson@crynwr.com (Russell Nelson) writes:
> ...
>    >Something that might hurry up the decades long death of token ring
>    >might be higher speed LAN's.  The short term winner(s) in those wars
>    >might have some effect on the numbers of new ethernet and token ring
>    >networks.  I thinking of cable plants and hubs that can be used for
>    >for both high and low speed links to the desktop.
> 
>    Token-ring won't die.  Eventually, FDDI will be much more common
>    than E-net and TR.
> 
> Both Token-ring and FDDI will die.  Did you miss all the hoopla about
> 100Mbps Ethernet?  It was in all the papers.

Everything dies eventually.  Everything.  However, I don't think those
babbling that new buzz have a clue about what will kill FDDI.

Until a few months ago, I was wandering around agreeing with the then
current buzz that ATM will replace FDDI.  I'm not so sure today.  ATM
is not developing as fast as it must to be healthy in the LAN market.
I assume ATM interfaces that require host programmed I/O for each byte
are only temporary stupidities.  Something else about ATM as a LAN has
begun to smell bad, but I can't quite put my nose on it.
Maybe it's just post-buzz depression.

It will be a at least 6 to 18 months before more than a few dozen
people in the world have informed technical opinions on how whether
100MHz ethernet works under any interesting circumstances.  Until then,
almost all opinions about it are not even worth pubishing in the
netnews.  The wonderful stuff you hear now is green manure, intended to
fertilize standards committees and the budgets of those few people
actually working on it, or to demonstrate that the speaker is up on the
latest buzz.

100Mhz ethernet will remain a maybe-or-maybe-not until you can order
and have delivered within 4 weeks 100 MAC and support chips.  Until you
can build 100 prototypes, the hype is empty.

Sheesh!  100Mbit/sec FDDI-over copper is 2 or 3 years old, and the
current, almost standardized scheme is very different from the previous
proposals.


> ...
> I'm contracting with someone to write an FDDI packet driver.  The
> silly thing needs a 180Kb TSR to manage the FDDI connection.
> ...

That so many people in the 1990's can't program themselves out of an
elevator in less then 5 years with 500 people and 50MBytes of code says
little about anything technically interesting.

You don't need 180KByte of code to run an FDDI interface, but it does
require vastly more than makes good sense.   Send your thanks to ANSI.

Remember that before 100MHz ethernet is real, it will have to survive
that same Standards Process.  Remember how many decades ISDN and FDDI
have been plodding through committees.  Remember how many years 10BaseT
required.  By the time 100MHz ethernet is half standardized, 100
Mbit/sec will not be fast.


Vernon Schryver,  vjs@sgi.com