From: TOMSON ERIC <Eric(_dot_)Tomson(_at_)siemens(_dot_)atea(_dot_)be>
The "best" doesn't always win.
The "cheapest" often does.
Look at VHS over Betamax.
Look at Token Ring over Ethernet.
In fact, VHS was better than Betamax for what people wanted, and
CSMA/CD was better than 802.5 on almost any grounds you care to
name that weren't salescritter nonsense.
Yes, by all accounts Betamax had slightly prettier pictures but trailed
in recording time. Most people still record VCR tapes at super-low
resolution. I find the typical SLP tape is not just a waste of time but
painful to watch, but most of the rest of the world disagrees.
The 100VG-AnyLAN, FDDI, and 802.5 salescritter songs were based on
mistakes in textbooks such as confusing CSMA/CD with ALOHA and
distortions such as equating low standard deviation in latency to
bounded latency and bounded, random backoff to unbounded latency.
Worrying about whether an OPES WG would give IETF sanction to evil is
a waste of time. Whether the old telephants and those hoping to create
new telephants succeed, it will not be because the IETF does or does not
sanction interception proxies. Instead of nattering about whether OPES
is 50% or 90% down a slippery slope, you'd do more good for the net by
teaching your own systems about SMIME, PGP, IPSec, and STARTTLS. Such
as those are the only hope of protecting the notion of dumb networks
and end-to-end services.
That so few people care about actually doing something by ensuring their
own traffic is invulnerable to involtary improvement by such as OPES but
so many want to argue about the virtues of OPES is not a good sign about
the IETF.
Vernon Schryver vjs(_at_)rhyolite(_dot_)com