Sprints definition on NAPs (question)
jerry at fc.net
Wed May 1 07:10:09 UTC 1996
>|} > the Sherman Act (if memory serves). These types of problems can be quite
>|} > nasty, involving treble punitive damages.
Unfortunately for Nathan, this above is wrong.
There are very real engineering reasons for not peering
if someone is at one NAP/MAE. Also since Sprint and MCI
do have published policies, if they made exceptions to them
they could get sued for discriminating against some competators
(not all, makes a big legal difference).
So in fact, unless Sprint and MCI want to give away
service to all people that connect to the MAEs/NAPs,
they MUST have a policy, and MUST abided by it.
(And as soon as that happens, I know of a Texas company
that will drop lines into MAE-East and force peering
with Sprint and MCI, etc., needless to say I don't see that
happening, so I will have to build a backbone to three NAPs
just like everywhere else.)
And there is the issue of actually having peering capacity
available. (Not only do some want free service to the
carrier's customers, but they want the carrier to replace all
of the carrier's routers).
I understand that when capacity is available, a number of the carriers
would not be adverse to discussing having someone that does
not meet the full requirements for peering, PAY to get peering,
thus offseting the backbone costs. (This should cost less than
full transit since its just inside the carrier's backbone, but
this partly depends on the true incremental cost of the paths
>|} Ya, but Sprint has more money then us, and money wins. :-)
>More importantly, Sprint (or any "larger" carrier) has content, and
>customers that YOU (being a "smaller" ISP) want to provide to your
>customers. Typically the larger folks are happy to get to ISP #1 via
>their single transit route because there's less load on their routers
>(being border or otherwise), fewer paths, etc.
Jeremy Porter, Freeside Communications, Inc. jerry at fc.net
PO BOX 80315 Austin, Tx 78708 | 1-800-968-8750 | 512-339-6094
More information about the NANOG