Bill: Thanks, but I'm looking for something worldwide.
Jean: A survey site is never going to be even remotely correct. People are too busy to bother entering data more than they have to. I was hoping for a method that uses known good data.
Marius: Thanks, but like my comment to Jean, that site is never going to be correct. For example, I have run 6 BBSs for several years now and I didn't even know about the site. And currently it lists only 7 gateways in the United States. We have more than that just in our one county in California. Not very useful.
For AMPRnet: What do you all think of the following logic? 1) Take the number of gateways listed in the portal. 2) Multiply by some factor to account for dead entries. This factor would be less than 1. But it's probably not much less than 1 since we just went through the transition to the portal and that must have cleaned out lots of dead stuff, right? So, maybe 90% are good, or a factor of 0.9? 3) Multiply by another factor representing the number of BBSs per gateway. This factor is greater than 1. But I think maybe it's not much greater than one. For example, we used to have 4 BBSs behind one gateway. But in order to gateway inbound e-mail without violating FCC Part 97 (discussed here long ago), we added Internet connections at each site. Once we had Internet at each site, we made each machine its own gateway. So I'm thinking the factor is really close to 1, maybe 1.1. 4) Given both factors, I conclude that a reasonable estimate of the number of BBSs reachable via AMPRnet is approximately equal to the number of gateways in the portal. Thoughts?
For the BBS Network: I still don't have a clue as to how to estimate this.
Michael N6MEF