|
|
|
|
|
|
|
|
|
|
|
|
| home | help | back | first | fref | pref | prev | next | nref | lref | last | post |
Date: Tue, 14 Jan 1992 08:54 EST From: BRIAN KAHIN <KAHIN@HULAW1.HARVARD.EDU> To: becker@ans.net Cc: com-priv@psi.com X-Vms-To: MAILER%"becker@ans.net" Jordan, I am confused by this statement that you just made: >You are correct in presuming that ANS will assess a charge to the intermediate >network that hosts a commercially classified subscriber(s). The charge will >be based upon the size of the pipe, and the total percentage of CO and RE >COMBITs (which are directionless) generated by the intermediate network. If >100% of the subscribers hosted by the intermediate network are classified as >CO, then they will pay a flat predictable rate. If 100% of their subscribers >are classified as RE, then they will pay a lower predictable flat rate >(nothing in the case of an NSF sponsored midlevel network). It was the last sentence that threw me. Are you talking about a charge for use of the noncommercial ANSnet rather than CO+RE? How do you define an "NSF sponsored midlevel network"? I thought that NSF had to give permission for midlevel nets to use the NSFNET backbone (subject to the NSF AUG) -- and had done so for PSI, AlterNet, and BITNET, as well as the regional networks. As these all "NSF-sponsored"? Or do you have a narrower definition in mind? Brian Kahin
|
|
|
|
|
|
|
|
|
|
|
|
| home | help | back | first | fref | pref | prev | next | nref | lref | last | post |