The large ISPs and Peering

Joe Provo joe.provo at
Wed Jul 25 21:08:32 UTC 2001

On Tue, Jul 24, 2001 at 07:50:04PM -0700, Peering Resistance wrote:
> This is an interesting tale, and one that everyone
> involved with the ISP world should know about.

This also isn't much news. It ain't collusion to keep costs down, nor is
it collusion to offer cheaper transit services to your customers when
they're in the same spaces as you...

Considering how much hay some folks STILL make of the old head-of-line
blocking at the MAEs and tit-for-tat interconnection upgrades, why is it
a Bad Thing?

There's a presumption that this indicates a major change; nope. Just
folks sharing more of their criteria and realizing that some of the
places in which they already inhabit/are planning to inhabit fir the
bill.  There also appears to be the presumption that one would ONLY be 
able to get service/connectivity/etc from these big dogs at these sites. 
I hadn't heard that.  Of course there would be a thin-end-of-the-wedge,
slippery slope argument if that were true, but it aint. The previous
poster nailed it when pointing out the desire to fire-and-forget cost-
effectively for fat PIs. 

> 1) What are the selected sites?
I'm sure there will be a press release of some kind.

> 2) How do the rest of us play?
What do you mean by 'play'?  If you meet a networks' peering bar, ask 
them. Most large networks have publicly-posted peering requirements. If
you can't find them, ask. If you're a customer-level and not a peer,
then don't.  If you're a colo operator in a non-fiber-rich city upset
that you've been passed over, get some facilities.

> 3) Why wasn't this process more open?
What degree of openness would satisfy you? Business (just like people)
need to decide where to spend their effort. I'm sure it's a waste of
time to send the RFP to folks that don't have facilities in any of the
cities you have in common with your top N (5? 8? 10?) peers. So I
wouldn't expect the Hills of West Virginia IX [no offense to the West
Virginians] to get a copy. 

Likewise, I'm sure the businesses have their own definitions of return-
on-investment for defining the Top N they want to play with.  I'm sure 
the bean counters wnt a small N and the engineers want a large one.
There are very few engineers that want to NARROW their choices, and
very few bean counters that want to spend money, in this or any economy.
For many folks, typically those who post anonymously in order to stir
controversy, there is no satisfactory level of "openness".  

To quote Barry Shein "Yes, keep an open mind, but not so open that your
brains fall out of your head."  


Joe "you may revisit the X-Disclaimer now" Provo

Joe Provo                                            Voice  508.486.7471
Director, Internet Planning & Design                 Fax    508.229.2375
Network Deployment & Management, RCN                 <joe.provo at>

More information about the NANOG mailing list