Good intent and somewhat competent

[Karl Denninger]

> Were there to be a major shift in the allocation requirements that would
> lead to one or more providers grabbing a huge block of address space
> without generally agreed upon justification, the other large providers
> (i.e. those who exchange BGP routes at the NAPs) would just not accept
> advertisements for the strongarmed routes.  When Sprint stopped accepting
> new route advertisements that were smaller than a /19, they didn't just
> wake up and do so, they had been warning people for months that if people
> continued to advertise /24s and such, that the routing tables would become
> so large that no router could handle them.  People continued to advertise
> small blocks, so they did what they felt was justified, and I don't think
> any large provider felt that Sprint was wrong, even if they may have been
> ecstatic that Sprint took the fall as the "bad guy out to kill the small
> ISP".
>
> Jeremiah
>
>
>       ________
>       \______/                  Jeremiah Kristal
>        \____/                   Senior Network Integrator
>         \__/                    IDT Internet Services
>          \/                     jeremiah at hq.idt.net
>                                 201-928-4454

Of course, the REASON we have this problem goes back a few years... were you
on the net then?

Remember the CISCO AGS+?  Used to be the workhorse of the Internet.  16MB of
RAM, 68040 processor.  Not a bad box (We still have some in service as
interior routing devices).

HOWEVER - its downfall was not just RAM space, but CPU horsepower and
ARCHITECTURE.  A basic architecture that was replicated not once, but TWICE
by CISCO since they found out that it was insufficient (first in the 7000
series, and then again in the 7500!)  The first replication was bad enough
-- the second, IMHO, is inexcusable.

CIDR was designed and pushed by CISCO engineers.  It was done due to the
fact that *CISCO DID NOT MAKE A DEVICE AT THE TIME WHICH DID NOT HAVE
THOSE LIMITATIONS*.  Unfortunately, neither did anyone else!  IF they had,
CISCO likely wouldn't HAVE a backbone business right now -- and we wouldn't
be stuck with route aggregation concerns.

So here we are in 1996.  Several years later.  CISCO *STILL* doesn't make a
router with an intelligent architecture which can actually handle the
offered loads.  And guess who's name is on some of the more-recent RFCs
regarding address allocations and such?

CISCO employees.

The "why" is left to the reader.

BTW, that monopoly is about to be broken.  Despite the fact that this
industry has pampered a company that is stuck selling 1970's technology in
1996 (when IMHO it should have forced them out of the market or forced them
to adopt solutions which would WORK) it still is happening -- some people
ARE in fact waking up to the opportunity that is present despite the
railroading of the standards process.

Of course, we also now have "BCP" documents and business practices which
IMHO act to restrain trade and possibly violate anti-trust laws...

--
--
Karl Denninger (karl at MCS.Net)| MCSNet - The Finest Internet Connectivity
http://www.mcs.net/~karl     | T1's from $600 monthly to FULL DS-3 Service
                             | 99 Analog numbers, 77 ISDN, Web servers $75/mo
Voice: [+1 312 803-MCS1 x219]| Email to "info at mcs.net" WWW: http://www.mcs.net/
Fax:   [+1 312 248-9865]     | 2 FULL DS-3 Internet links; 400Mbps B/W Internal