[aprssig] Multi control station situational awareness - How's this supposed to work?

[Rob Warren]

Andrew,

Agreed. There is a lot of distribution and routing logic that was part of the rnews-rmail-uucp stack that is worth a second look. ...my mail/news node was running on a 8MB machine with a 2400bps modem.... Given the low-cost of storage, monitoring concurrent channels with sdr's and software tunable transmitters, I think there is potential for recycling this code base to do opportunistic routing for mobile stations. 

-R

> On Mar 31, 2020, at 12:00 PM, aprssig-request at lists.tapr.org wrote:
> 
> From: Andrew Pavlin <spam8mybrain at yahoo.com>
> Subject: Re: [aprssig] Multi control station situational awareness - How's this supposed to work?
> Date: March 30, 2020 at 10:49:05 PM EDT
> To: TAPR APRS Mailing List <aprssig at lists.tapr.org>, "Stephen H. Smith" <wa8lmf2 at aol.com>
> 
> 
> Hi, Steve.
> 
> Re-reading your email on this as I got restarted on my own effort in this area reminded me of some other ancient technology that we might want to resuscitate for this problem.
> 
> Does anybody remember Usenet (otherwise known as the InterNetNews)? It was a powerful means of implementing a distributed world-wide collection of thousands of bulletin boards of discussion threads, back before the World Wide Web, hosting service providers, and (nearly) ubiquitous broadband replaced Usenet with world-accessible single-server web forums and blogs. Like email in those days, Usenet only carried plain-text; like email, it could carry anything that could be bundled into a plain-text email message, such as binary files encoded by the useful uuencode and uudecode programs. It would automatically synchronize all the distributed copies of any given discussion group. And it could work over (by today's standards) ridiculously low-bandwidth links. In 1991, I was running a corporate Usenet news gateway over a leased-line Internet connection at a screaming 19.2 kilobaud. Yes, _that_ slow. Yes, we had dialup modems that went faster than that before broadband.
> 
> These days, the Usenet news server software is still available in most Linux distros (I just checked, and both Fedora Core and Raspian Buster still have it as an optional distro package). Many email clients still support NNTP (Network News Transport Protocol) as well as SMTP (Simple Mail Transport Protocol). And NNTP can transfer over any TCP/IP link (including TCPIP-over-AX.25 and HSMM, as well as the global Internet), and over batched low-level links (it used to use an old package called UUCP [Unix-to-Unix CoPy] to transfer updates over dialup links) at barely more infrastructure than the KISS protocol.
> 
> So, we could set up NNTP servers on Raspberry Pi computers (or anything else) and use any sorts of links to connect them together: Internet, HamWAN, AREDN, TARPN, heck maybe even fldigi file transfers (not much different than what UUCP did). Because NNTP uses a flood-fill algorithm to distribute messages over multiple paths, if one link goes down, the target at the other end of the failed link will eventually get it via several relays on other links as long as every news server has links to more than one other news server, and the topology doesn't have any Single Points Of Failure. No particular network topology is required; just like amateur radio, Usenet doesn't need a central control office (unlike cellphones). We can certainly get sufficient TCP/IP speeds over AX.25 packet with the 9600-baud TNCs (hardware and software) that are readily available now for a TARPN-style VHF network for areas where we can't do HamWAN/AREDN, but NNTP will still work over those networks as well. And, if we keep our Usenet separate from what's left of the old Internet Usenet, we don't have to worry (as much) about illegal content putting transmitting stations at risk or excessive traffic volume. After all, most public service events that use APRS put their event traffic on a different frequency than the national APRS frequency to avoid congestion.
> 
> So, if what is needed to solve the problem is a distributed bulletin board, Usenet solved it for us decades ago.
> 
> Just my $.03 (inflation, ya know).
> 
> Andrew, KA2DDO
> author of YAAC ("Yet Another APRS Client")
> 
> 
> On Wednesday, December 4, 2019, 12:11:44 AM EST, Stephen H. Smith via aprssig <aprssig at lists.tapr.org> wrote:
> 
> 
> On 12/3/2019 6:26 PM, chiefsfan2 at cox.net wrote:
> Since you had a analog landline phone still working that would be a reason to bring back some phone patches like we used to have. And now you can run a BBS on a rasp pi computer which makes for great portability and low power consumption
> Funny you should bring this up at this particular time.  Just last week, I was experimenting connecting an old Heathkit HD-1515 phone patch I found in my junk box to the 6-pin mini-DIN data port of a Yaesu FT-857D.   It worked perfectly both on FM for 2 meters and on SSB for HF.   I'm now going to add a 6-pin mini-DIN jack to the back panel of the patch, in parallel with the existing RCA RX and TX audio jacks.   I can then use a standard off-the-shelf  6-pin DIN to 6-pin DIN cable to connect the patch to any radio with a standard 6-pin data port.   Finally, I will add a double-throw center-off   locking-one-way  /  momentary-the-other-way toggle switch to the front panel to key the radio transmitter.   
> 
> I'm now thinking about getting one of those Bluetooth gizmos that links to a cellphone and and produces a couple of classic RJ-11 analog phone jacks. I could plug the patch into one and a classic desk phone set into the other.  This would allow phone patches either via  a "real" phone line, or via a cellphone connection if needed.
> 
> Another variation on this theme:  With a sound card interface setup normally as you would use for digimodes on a PC,  start up Skype instead of a soundcard digi-mode app. You can then run "phone patches" from radio users to users on Skype instead of a POTS line.  
> 
>