Yeti brought up a thing in a separate topic.

From: yeti <yeti@tilde.institute>
Date: Wed, 26 Feb 2025 08:09:27 +0042
To: tilde.projects

Fediverse is the new trend? Running good old mail (and news) created
the original federation and we could run such on SBCs. There is not
even a need to be permanently online and no need to buy domains (use
I2P, Tinc, Tor or similar instead).

Considering I've kinda had an idea in similar spirits to this since
Arab Spring. Admittedly i"m not all that technically minded, so it's
never gotten out of the idea phase. Plus 'OK I want people to be able
to use their phones since that's where most people are' so I've never
been able to mentally square the circle on how to Network when there
are no proper addresses:

Example being me walking by the local market/diner/whatever and Jim's
there. How would my phone (though it could just as easily be my tablet
and Jim's laptop,) know to talk to each other?

Or, what if I connect to the store's node could the store then link to
Jim and Anne's content to feed to me?

Maybe I'm just going through this the wrong way since Usenet seemed to
work pretty well with only a few servers a LOT of people talk to and
the servers sharing data to keep from constant doubling up on who got
what.

"Peernix" instead of "Pubnix" will force you to be the admin and to
learn a lot and all sensible data can stay at home. Only stuff to be published would be globally visible or even replicated among the
peers.

I like the name. Mind if I use that?

As I said I've had the concept on the brain for awhile as a 'oh hey if
the internet as a monolith ever went away or became unusable.' The why
of these things is immaterial to me and irrelevant to the overall
discussion beyond whether the transport layer can be used or not. For preference sake I'd prefer a 'assume the internet is not there, but if
it is then make use of the network to serve data' approach. Both
because while it exists that extends reach, and if someone wants to
build their own network these server/nodes should be able to take
advantage of that rather than only work as stand alone islands.

We really should do that!

In full agreement. I just happen to recognize I'm the dummy in the room.

--- Synchronet 3.20a-Linux NewsLink 1.2

Andrew Singleton <singletona082@ctrl-c.club> wrote:

I've never been able to mentally square the circle on how to Network
when there are no proper addresses:

There are addresses, but no central authority.

"Peernix" instead of "Pubnix" will force you to be the admin and to
learn a lot and all sensible data can stay at home. Only stuff to be
published would be globally visible or even replicated among the
peers.

I like the name. Mind if I use that?

If you use it for the bigger view or idea, there should not be a
conflict. Like Pubnix it's just not meant as name of a product, system
or a concrete implementation.

We really should do that!

In full agreement. I just happen to recognize I'm the dummy in the
room.

Don't we all have such moments?


I'm just starting to read about "Reticulum Network Stack" (and "Rnode")
and this may be even more interesting than using I2P and Tor, but for a
solid opinion it definitely is too early.

This seems to be a good appetiser:

Reticulum Network Stack 0.9.3 beta documentation
Getting Started Fast
<https://markqvist.github.io/Reticulum/manual/gettingstartedfast.html>

I'll add to the Meshtastic thread if I have news about MT, even if it
will be only that I'll switch to RNS, so this Peernix thread shall only
be minimally polluted.

But RNS is a candidate to mesh up peers, so it isn't completely
unrelated here.
--
When dictatorship is a fact, revolution becomes a right. -- Victor Hugo
--- Synchronet 3.20a-Linux NewsLink 1.2

On Wed, 26 Feb 2025 22:12:56 +0042
yeti <yeti@tilde.institute> wrote:

There are addresses, but no central authority.

If you use it for the bigger view or idea, there should not be a
conflict. Like Pubnix it's just not meant as name of a product,
system or a concrete implementation.

I view it less as a 'product' because, put bluntly, even if I could
churn out widgets that just worked? I don't like the idea of commercial
for its own sake. I'd want to just... distribute the things to get an infrastructure going. I don't want to use buzzwords to describe the
concept but 'anti-commerical' fits in this instance. I'm not opposed to
money making and I do see the companies making mesh things as 'OK I can understand why you're doing the thing.' This isn't about Money.

In full agreement. I just happen to recognize I'm the dummy in the
room.

Don't we all have such moments?

I'm in the 'the older I get the dumber I feel' camp, or possibly 'The
older I get the more I realized I didn't know' camp. It often leaves me
in a state of aggravation.

I'm just starting to read about "Reticulum Network Stack" (and
"Rnode") and this may be even more interesting than using I2P and
Tor, but for a solid opinion it definitely is too early.

This seems to be a good appetiser:

Reticulum Network Stack 0.9.3 beta documentation
Getting Started Fast
<https://markqvist.github.io/Reticulum/manual/gettingstartedfast.html>

But RNS is a candidate to mesh up peers, so it isn't completely
unrelated here.

Reticulum is something I 'kinda' like, but I'd skipped it over pretty thoroughly because I think the whole nomadnet thing is, put bluntly,
Stupid. yes it's very slick in presentation, and for me that's the
problem. It's a commandline thing trying to look pretty, and so in the
process can't really be translated to a GUI application either on
desktop or mobile.

'But why not just use a terminal on your-'

1. Your eyes are a hell of a lot better than mine if you can do that.

2. The reticulum mobile app is.... so much less than nomadnet to the
point 'here it's a SMS thing.' Which, you're not going to sell anyone
on the idea of using it if the graphical side are third class citizens.

3. From an accessibility standpoint, nomadnet is terrible. Full, flat
out. I've seen worse, but turn on orca and try navigating it.

4. Looking at the 'how you make nomadnet content' leaves me going
'thanks... I hate it.' Compare writing something in gemeni to this. It
is literally how you code in room information for a MUD.

5. This is probably the most damning of all for adoption. The guys that
made this were too preoccupied with making cool nerdy thing that they
didn't stop to consider how normal people would interact with it either
as consumers of content or trying to MAKE content. Scoff at the GUI all
you want, mock phone use all you like, but we Need 'normal' people to
find this interesting or there isn't going to be enough coverage for it
to be more than a tinkertoy.


--- Synchronet 3.20a-Linux NewsLink 1.2

On Wed, 26 Feb 2025 22:12:56 +0042
yeti <yeti@tilde.institute> wrote:

If you use it for the bigger view or idea, there should not be a
conflict. Like Pubnix it's just not meant as name of a product,
system or a concrete implementation.

I view it less as a 'product' because, put bluntly, even if I could
churn out widgets that just worked? I don't like the idea of commercial
for its own sake. I'd want to just... distribute the things to get an infrastructure going. I don't want to use buzzwords to describe the
concept but 'anti-commerical' fits in this instance. I'm not opposed to
money making and I do see the companies making mesh things as 'OK I can understand why you're doing the thing.' This isn't about Money.

In full agreement. I just happen to recognize I'm the dummy in the
room.

Don't we all have such moments?

I'm in the 'the older I get the dumber I feel' camp, or possibly 'The
older I get the more I realized I didn't know' camp. It often leaves me
in a state of aggravation.

I'm just starting to read about "Reticulum Network Stack" (and
"Rnode") and this may be even more interesting than using I2P and
Tor, but for a solid opinion it definitely is too early.

This seems to be a good appetiser:

Reticulum Network Stack 0.9.3 beta documentation
Getting Started Fast
<https://markqvist.github.io/Reticulum/manual/gettingstartedfast.html>

But RNS is a candidate to mesh up peers, so it isn't completely
unrelated here.

Reticulum is something I 'kinda' like, but I'd skipped it over pretty thoroughly because I think the whole nomadnet thing is, put bluntly,
Stupid. yes it's very slick in presentation, and for me that's the
problem. It's a commandline thing trying to look pretty, and so in the
process can't really be translated to a GUI application either on
desktop or mobile.

'But why not just use a terminal on your-'

1. Your eyes are a hell of a lot better than mine if you can do that.

2. The reticulum mobile app is.... so much less than nomadnet to the
point 'here it's a SMS thing.' Which, you're not going to sell anyone
on the idea of using it if the graphical side are third class citizens.

3. From an accessibility standpoint, nomadnet is terrible. Full, flat
out. I've seen worse, but turn on orca and try navigating it.

4. Looking at the 'how you make nomadnet content' leaves me going
'thanks... I hate it.' Compare writing something in gemeni to this. It
is literally how you code in room information for a MUD.

5. This is probably the most damning of all for adoption. The guys that
made this were too preoccupied with making cool nerdy thing that they
didn't stop to consider how normal people would interact with it either
as consumers of content or trying to MAKE content. Scoff at the GUI all
you want, mock phone use all you like, but we Need 'normal' people to
find this interesting or there isn't going to be enough coverage for it
to be more than a tinkertoy.


--- Synchronet 3.20a-Linux NewsLink 1.2