- AMPRNet / HamNet routing is quite complicated for a non-IT guy. BGP
requires huge equipment and skills. IPIP requires hacking protocol redirect on Internet boxes. Those are not easy things for people operating a voice repeater or hotspot. They just build a Pi image, plug the machine, and it works. Why should they bother with complex addressing ?
We have quite some repeaters that are connected via AMPRnet. We normally use MikroTik routers. I do not consider these "huge equipment" and they are not difficult to configure with BGP. I have some example configs for setting up an endpoint with L2TP/IPsec tunnel to our gateway router and using BGP to advertise their own subnet. This is much easier to get going than IPIP, for example there is no need to touch the existing internet router (open ports/protocols not required). This even makes it suitable for installation on buildings where the owner may make available some guest internet access but would not want you to tweak their network to pass IPIP. The advantage of using BGP even in this trivial case is that the network can now be extended when the opportunity arises without having another hurdle of complexity. A WiFi link to another station can be added, e.g. in some cases people have an internet connection at an amateur nearby the repeater, and then a WiFi link to the repeater itself.
I would be all for rolling out such a system worldwide to replace the IPIP mesh. Routers (e.g. MikroTik CHR that can run as a VPS) in datacenters all over the world interconnected with a static tunnel mesh and offering VPN service for local amateurs to connect, and routing using BGP on private AS (this only routes AMPRnet, not full internet). In different places those routers could have the AMPRnet subnet(s) for that region announced on internet, like we do for 44.137.0.0/16 and others do for other country networks. And each of those can offer different VPN technologies so you are able to follow the trend of the day without having to do a migration in the entire network.
Rob PE1CHL
AMPRNet / HamNet routing is quite complicated for a non-IT guy...
The advantage of using BGP even in this trivial case is...
I don't think the most important question is about selling BGP or any particular technology (I'm well versed in internetwork engineering and worked in that field professionally for many years; I'm in academia now).
I'm writing this because education was a specific question in the survey.
The reason that we have amateur radio is to enable experimentation with using the radio spectrum in a way that is otherwise not permitted or practical. With the Internet, there are certain things that are only possible to experiment with if you have your own addresses and other network numbers. AMPRNet is (perhaps that's too strong, and we could say, "can be") a way of enabling a kind of experimentation on the Internet similar to what we do with the radio spectrum.
Putting the Internet on a similar footing to radio as an object of experimentation is not what every amateur radio operator wants to do -- for some, their interests and even their identity is bound up in radio. The Internet is something else. For others, the interest is in ways of communicating and they have a very similar character.
Much of what we learn from radio is about point to point communication. A QSO doesn't classically involve an intermediary. But then we have repeaters and digipeters that pass traffic. Those are single intermediaries and are very simple to understand. Beyond that, sometimes we invent ways of networking repeaters (typically very differently from the way that packet-switched networks function). That starts to get into territory that is not well-understood by many of the users, or indeed some of the operators of the repeaters.
We also have an old practice of manual message passing. That process resembles packet-switched networking in some ways, though the rule for how to decide where to pass a message to is not always clearly defined.
There is, it seems to me, a cluster of topics that can be described as, "how can messages be passed across a network", that most amateur radio operators, and most users of the Internet are never exposed to. How does these actually work in practice? Where does this mysterious thing called BGP that people seem to think is so difficult fit in? How can messages be passed between all of these different networks to get to their intended destination? This is a gap that could usefully be addressed by developing material within or around AMPRNet.
73s William VE0HAK
Mr. Waites...
Thank you for your viewpoint. This is also one that I share. My knowledge in Radios, was learned from the school of hard knocks, and a lot of trial and error. The more I learn, the more I realize that I do not know... <grin>
My limited knowledge of Networking was acquired when running an Explorer Post in Advanced Computer Programming in Tampa, FL. My employer and the local BSA council sponsored Explorer Post 227 for Guys and Gals aged 13-21, and dedicated a classroom for their exclusive use.
I was the "Executive Advisor", (an unpaid, volunteer title) and I and the rest of the Advisors also used the classroom as an advanced Computer Lab, where we and the post members constructed an environment that supported several Internet and Intranet Servers, a number of workstations, and several different types of ethernet networks. All hardware and software installation and maintenance was conducted by the Post members under the leadership of the advisors. For a long time we were connected directly to the corporate DMZ, but eventually were moved to a separate T1 connection as a corporate security precaution.
Because we were directly connected to one of the Corporate Internet "connection points", we did not experiment with many of the actual protocols used for Internet data transportation - Knowledge that I wish I had today, even though I am now retired, and the Post disappeared in 2007 when the Building that housed the Post's classroom was sold.
Never occurred to me at that time that the "44" net could have been used in that environment. I really like your idea of being able to use the "44" net as an educational facility just like Amateur Radio is used... The only question, is how do you allow experimentation without impacting the "44" net as a service?
/s/ Bill Turner, wb4alm
On 10/4/20 8:40 AM, William Waites via 44Net wrote:
AMPRNet / HamNet routing is quite complicated for a non-IT guy...
The advantage of using BGP even in this trivial case is...
I don't think the most important question is about selling BGP or any particular technology (I'm well versed in internetwork engineering and worked in that field professionally for many years; I'm in academia now).
I'm writing this because education was a specific question in the survey.
The reason that we have amateur radio is to enable experimentation with using the radio spectrum in a way that is otherwise not permitted or practical. With the Internet, there are certain things that are only possible to experiment with if you have your own addresses and other network numbers. AMPRNet is (perhaps that's too strong, and we could say, "can be") a way of enabling a kind of experimentation on the Internet similar to what we do with the radio spectrum.
Putting the Internet on a similar footing to radio as an object of experimentation is not what every amateur radio operator wants to do -- for some, their interests and even their identity is bound up in radio. The Internet is something else. For others, the interest is in ways of communicating and they have a very similar character.
Much of what we learn from radio is about point to point communication. A QSO doesn't classically involve an intermediary. But then we have repeaters and digipeters that pass traffic. Those are single intermediaries and are very simple to understand. Beyond that, sometimes we invent ways of networking repeaters (typically very differently from the way that packet-switched networks function). That starts to get into territory that is not well-understood by many of the users, or indeed some of the operators of the repeaters.
We also have an old practice of manual message passing. That process resembles packet-switched networking in some ways, though the rule for how to decide where to pass a message to is not always clearly defined.
There is, it seems to me, a cluster of topics that can be described as, "how can messages be passed across a network", that most amateur radio operators, and most users of the Internet are never exposed to. How does these actually work in practice? Where does this mysterious thing called BGP that people seem to think is so difficult fit in? How can messages be passed between all of these different networks to get to their intended destination? This is a gap that could usefully be addressed by developing material within or around AMPRNet.
73s William VE0HAK _________________________________________ 44Net mailing list 44Net@mailman.ampr.org https://mailman.ampr.org/mailman/listinfo/44net
Bill, et al: You can easily make your experimental "sandbox" not routable to the rest of 44Net if you want. Everything is configurable.
On 10/4/2020 9:44 AM, Bill Turner, WB4ALM via 44Net wrote:
Never occurred to me at that time that the "44" net could have been used in that environment. I really like your idea of being able to use the "44" net as an educational facility just like Amateur Radio is used... The only question, is how do you allow experimentation without impacting the "44" net as a service?
/s/ Bill Turner, wb4alm
Charles (/as Bill smacks his forehead with the palm of his hand/)
Of Course! Now why didn't I think of that.
After all, routes occur automagically, right? and they are always self-healing, right?
Yeah, sure.
Yes, we did teach our students about "magic smoke and mirrors"
...if you let the "magic smoke out, it won't work anymore...."
The classroom had plenty of power outlets. and when I asked to have a disconnect switch so I could insure that all of the work stations were powered off, they mounted a huge transformer at the top of the electrical closet with a relay across the primary.
We used it once.
and spent the next two weeks degaussing all of the CRTs whose internal mounting straps became severely magnetized by the "EMF" pulse that was created when the transformer was powered down.
I was told later, that the office space immediately above the classroom also had some problems with their Display CRTs that the Company's I.T. technicians couldn't figure out.
It seems that they all had severe color distortion in the lower right hand corner of the displays, and most also had a radiating circular rainbow effect....
I didn't say a thing!
/s/ Bill, wb4alm
On 10/4/20 10:01 AM, Charles J. Hargrove via 44Net wrote:
Bill, et al: You can easily make your experimental "sandbox" not routable to the rest of 44Net if you want. Everything is configurable.
On 10/4/2020 9:44 AM, Bill Turner, WB4ALM via 44Net wrote:
Never occurred to me at that time that the "44" net could have been used in that environment. I really like your idea of being able to use the "44" net as an educational facility just like Amateur Radio is used... The only question, is how do you allow experimentation without impacting the "44" net as a service?
/s/ Bill Turner, wb4alm
-
Bill Turner, WB4ALM -
Charles J. Hargrove -
Rob Janssen -
William Waites