On Thu, 2013-07-25 at 06:38 -0400, several spaketh:
I guess it will kind of boil down to applications. And a way to advertise them to everyone else who has access to the network.
Actually you need to reach those who don't have access as well, and those could produce some unknown key players down the road!
I envision those benefactors (not necessarily ISPs) who have access to fiber optic networks, ipv6, and some spare bandwidth, with facillities that could support a couple of servers, and last-mile connections to: HSMM (now called Broadband-Hamnet); cable; or even fiber; to local groups of hams who might organize to provide that last-mile internet access to remote sites (and/or their QTH). Bandwidth needs should be minimal, amateurs are licensed, and for the most part would be willing to agree to avoid streaming entertainment, etc.
Isn't that what local gateway operators do? Provide connectivity via RF into the amprnet? The RF is a given IMHO. It'll pump your protocols through regardless of if it's 1200 baud 2m or HAMNET.
This is not intended to replace their commercial ISP but to provide a backup, with, of course, 44-net addresses. ( I currently pay two independent ISPs $39 a month for the privilege of testing routers, OpenVPN, and having a pitifully slow point-2-point RF connection for when the cable is being hogged by Netflix ;)
The average "joe ham" will think: "for 39 bucks, I can watch whole length movies vs for 100 bucks I can have faster HAMNET and can't watch full length movies". It's a matter of cost effectiveness.
A potentially useful quote from Broadband-Hamnet:
- It is NOT intended to replace your personal Internet connection. It is a
“feature” that HSMM-MESH™ can be connected to the Internet. General Internet access is not its primary purpose. Use your personal Internet connection under Part 15 to serve that purpose.*
In a sense, Brian Kantor indirectly said that here as well.
One application that I can see being enabled by regional peering could be analysis of the RF network for a service such as APRS, where duplicate AX-25 packets are collected and stored, rather than hitting the bit-bucket.
Perhaps I'm wrong here, but I would think those who are licensed and want to participate in APRS are already doing so... and I'm sure many of them aren't even on the amprnet.
I also see a need for grass-roots Internet access that can be deployed in the field to get e-mail into disaster areas - or APRS into remote areas to support Search-And-Rescue (SAR)
That's why I completed the work started by Hessu and Marius in regards to axMail-Fax... however there's already a system out there called Winlink 2000 that uses HF, FM, and Internet to handle email. When I query those who run it I get almost always these 3 identical comments:
1) It's windows, there's no learning curve of nos or linux involved. 2) I can use the resources I have already without any _additional expenses_. 3) I don't have time to learn amprnet.
... then there was the issue raised about 3-party relaying, so now we have governmental restrictions shying us away from certain applications.
How we route is all well and good, but if you're not: - providing useful services, some of which may be unique - have a user base to use this network - cost effectiveness (especially to the seniors who may be on fixed incomes)
We all know how to build a bridge (network routing) but it appears we're discussing of doing such without having a road to connect both ends of it to.