Path: ...!news.misty.com!weretis.net!feeder9.news.weretis.net!news.quux.org!eternal-september.org!feeder3.eternal-september.org!news.eternal-september.org!eternal-september.org!.POSTED!not-for-mail
From: bp@www.zefox.net
Newsgroups: news.software.nntp
Subject: Re: ad-hoc wifi news transport
Date: Tue, 8 Apr 2025 19:12:13 -0000 (UTC)
Organization: A noiseless patient Spider
Lines: 98
Message-ID: <vt3sed$2rs7q$1@dont-email.me>
References: <20250320204547.0000274b@dne3.net> <874izak212.fsf@gmail.com> <vshbnj$3nh1s$1@dont-email.me> <x87plhsax2t.fsf@somewhere.edu> <vspreu$ouq9$1@dont-email.me> <vt14tg$dj1$1@news.dne3.net> <vt18nn$graf$1@dont-email.me> <20250408144842.00007267@dne3.net>
Injection-Date: Tue, 08 Apr 2025 21:12:14 +0200 (CEST)
Injection-Info: dont-email.me; posting-host="9f649473fe0e1dd9109421a4e1ceebf8";
	logging-data="3010810"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX1/ydf15yYBMW46eVjONZ+P0pyUdQQ/s7fw="
User-Agent: tin/2.6.4-20241224 ("Helmsdale") (FreeBSD/14.2-STABLE (arm64))
Cancel-Lock: sha1:QKeuE/y0Sh+unqiyPetJAb8nlHA=
Bytes: 5444

Toaster <toaster@dne3.net> wrote:
> On Mon, 7 Apr 2025 19:23:35 -0000 (UTC)
> bp@www.zefox.net wrote:
> 
>> Toaster <news@dne3.net> wrote:
>> > On 4/4/2025 7:54 PM, bp@www.zefox.net wrote:
>> > snip
>> >> I don't think it required unique hostnames, though they were
>> >> preferred, since the bang paths generally were different for
>> >> different hosts even if the had the same name. In a sense the bang
>> >> path was the identity of any given host.
>> >> 
>> >> UUCP died out because DNS and ICANN made TCP/IP much easier to
>> >> administer and was incomparably faster. With with some thought it
>> >> might be useful again.
>> >> 
>> >> Thanks for reading,
>> >> 
>> >> bob prohaska
>> >>   
>> > I'd like to experiment around with uucp and nncp. looking at nncp 
>> > protocol currently.
>> 
>> Are you thinking of using WiFi LANs as the transport layer? 
>> 
>> My router can see about a dozen WAPs, each of them can likely see 
>> more than a dozen each. If I could peer with just a few of them and
>> we all agreed to share some fraction of our bandwidth back to our
>> ISP and other peers it would make for a rather dense mesh.
>> 
>> bob prohaska
>> 
> 
> Yes, that's the idea. I'm also looking into yggdrasil network. Zero
> configuration, end to end encrypted. The idea would be everyone peers
> together via wifi, directional links or omni-directional. Directional
> is more scalable. Then you share resources with people you know
> (internet access, files, etc) via public key. I.e. my friend and I
> trust each other, we both share our internet access to each other and
> allow default routing out.
> 
> If enough people do this then it would be a good auxilliary to standard
> internet access, good for low to mid bandwidth localized traffic. Right
> now most traffic on the internet goes to a few centralized sites, but
> does that have to be the case?

In principle, no. In practice, it simplifies matters enormously. There's
a reason central authorities tend to outcompete egalitarians, at least
initially 8-)

> 
> In a peer to peer network, you just install an application that
> communicates directly with the person you are trying to reach, no
> server needed. All it would take is a small shift in which applications
> are used.
> 
> Another idea is to ditch IP communication altogether and just use the
> mesh as a news network. Articles bounce around the mesh like they do on
> usenet. That's a tough sell, can't do much other than post like we do
> here on usenet.
> 
> Zero infrastructure cost, run by volunteers = free communications. 

What!!!, who do you think pays for all the infrastructure needed for
this scheme? We all do. It isn't free, it's shared, hopefully in an
equitable way. 


> I could see using it for out-of-band access to home servers, private sms,
> zero infrastructure chat applications. I'd put a news server and irc on
> it, make a phone app and dress it up nice and pretty for my friends
> that aren't technical. Could even support voip to your landline/hotspot.
>
One of the key points about uucp is that it wasn't, and could not be, 
"predictable delay" (there's no such thing as "real time"). At least
some of those features (voip) need short, predictable delays to work.
 
> The dream is the ability to connect your phone to this free net, and
> talk to your friends in your community directly instead of trusting
> some far off thing. Battery powered nodes would work when nothing else
> would (i.e. tornado took out most of everything, enough nodes remain
> and the network persists). All this needs is enough buy in from people
> to make a minimal network. I'd even pay for the units because im weird
> like that.

Again, nothing is free. All users have to share each other's costs. High
bandwidth, low-latency traffic likely won't work well, if at all. Don't
underestimate the routing overhead that such a system entails. For uucp
that fell on administrators to sort out manually, and they deserted uucp
in droves as soon as they got a T1 link. Perhaps some of that can be automated.

These comments aren't meant to be _very_ discouraging, just  realistic. 
Some may find them very discouraging anyway.

Thanks for writing,

bob prohaska