So we have a shell server for our customers. There aren't
that many folks using it anymore.

The shell servers we've had at sonic.com have always been
near and dear to my heart. I built them, and they were
labors of love. But it's time to turn the service down.

We like to give customers a soft landing when turning
down services. I'm wondering if anyone has suggestions
for good shell providers, places like Panix. I'm a panix
user myself, and that seems like a good example of suggestions
I'm hoping for.

When discussing this with a colleague, he suggested that
users could just get a vps from AWS. That's a bit too
"bare-bones" for the users I'm thinking of. Some of
them might be able to transition to being a system
administrator, but a lot just want to run tin, pine,
mutt, irssi, tf, and so forth.

Any suggestions greatly appreciated.

--
-Scott Doty
https://www.sonic.com/about-us#founders

On 3/8/24 12:36 AM, vallor wrote:
> So we have a shell server for our customers. There aren't
> that many folks using it anymore.
>
> The shell servers we've had at sonic.com have always been
> near and dear to my heart. I built them, and they were
> labors of love. But it's time to turn the service down.
>
> We like to give customers a soft landing when turning
> down services. I'm wondering if anyone has suggestions
> for good shell providers, places like Panix. I'm a panix
> user myself, and that seems like a good example of suggestions
> I'm hoping for.
>
> When discussing this with a colleague, he suggested that
> users could just get a vps from AWS. That's a bit too
> "bare-bones" for the users I'm thinking of. Some of
> them might be able to transition to being a system
> administrator, but a lot just want to run tin, pine,
> mutt, irssi, tf, and so forth.
>
> Any suggestions greatly appreciated.
>

The biggest is sdf.org and possibly the most user-friendly. It does
have a tiered membership structure and one generally will need to pay a
one-time $5 validation for a decent shell.

The various tilde servers are also good but I think email via a local
SMTP might not be an option. Take a look at tilde.club for example.

Check out https://tildeverse.org/ they might be able to find something
there for free.

On Fri, 8 Mar 2024, vallor wrote:

> So we have a shell server for our customers. There aren't
> that many folks using it anymore.
>
> The shell servers we've had at sonic.com have always been
> near and dear to my heart. I built them, and they were
> labors of love. But it's time to turn the service down.
>
> We like to give customers a soft landing when turning
> down services. I'm wondering if anyone has suggestions
> for good shell providers, places like Panix. I'm a panix
> user myself, and that seems like a good example of suggestions
> I'm hoping for.
>
> When discussing this with a colleague, he suggested that
> users could just get a vps from AWS. That's a bit too
> "bare-bones" for the users I'm thinking of. Some of
> them might be able to transition to being a system
> administrator, but a lot just want to run tin, pine,
> mutt, irssi, tf, and so forth.
>
> Any suggestions greatly appreciated.
>
>

On 2024-03-08, vallor <vallor@cultnix.org> wrote:
> We like to give customers a soft landing when turning
> down services. I'm wondering if anyone has suggestions
> for good shell providers, places like Panix.

That concept mostly went away 30 years ago with free Unix-like operating
systems that run on low-spec consumer hardware.

> "bare-bones" for the users I'm thinking of. Some of
> them might be able to transition to being a system
> administrator, but a lot just want to run tin, pine,
> mutt, irssi, tf, and so forth.

To access some remote shell account you need a machine
that is internet connected and can run SSH. That machine
can just run a freeware OS with all the above packages.

--
TXR Programming Language: http://nongnu.org/txr
Cygnal: Cygwin Native Application Library: http://kylheku.com/cygnal
Mastodon: @Kazinator@mstdn.ca

Kaz Kylheku <433-929-6894@kylheku.com> writes:

> On 2024-03-08, vallor <vallor@cultnix.org> wrote:
>> We like to give customers a soft landing when turning
>> down services. I'm wondering if anyone has suggestions
>> for good shell providers, places like Panix.
>
> That concept mostly went away 30 years ago with free Unix-like operating
> systems that run on low-spec consumer hardware.
>
>> "bare-bones" for the users I'm thinking of. Some of
>> them might be able to transition to being a system
>> administrator, but a lot just want to run tin, pine,
>> mutt, irssi, tf, and so forth.
>
> To access some remote shell account you need a machine
> that is internet connected and can run SSH. That machine
> can just run a freeware OS with all the above packages.

That's all true, but it seems that people like to share a system
together because people love to belong to a community. In the
beginning, I believe such communities were mostly made of local people
and the Internet has sort of destroyed that. I believe people still
long for these local communities again. BBSs connected local people.
The Internet seems to have done the opposite.

(*) On community building

There's also this conjecture that a person can't deal with more than N
people. The number I usually hear is 150. In other words, human
communities should be small (and local).

One principle I think about to achieve that is one begins a community
but others can only come through invitation. If someone misbehaves, we
know who invited that person---the tree of invitation being public.

Yes, we lose anonimity. Perhaps anonimity is overrated.

On Fri, 8 Mar 2024, Julieta Shem wrote:

> Kaz Kylheku <433-929-6894@kylheku.com> writes:
>
>> On 2024-03-08, vallor <vallor@cultnix.org> wrote:
>>> We like to give customers a soft landing when turning
>>> down services. I'm wondering if anyone has suggestions
>>> for good shell providers, places like Panix.
>>
>> That concept mostly went away 30 years ago with free Unix-like operating
>> systems that run on low-spec consumer hardware.
>>
>>> "bare-bones" for the users I'm thinking of. Some of
>>> them might be able to transition to being a system
>>> administrator, but a lot just want to run tin, pine,
>>> mutt, irssi, tf, and so forth.
>>
>> To access some remote shell account you need a machine
>> that is internet connected and can run SSH. That machine
>> can just run a freeware OS with all the above packages.
>
> That's all true, but it seems that people like to share a system
> together because people love to belong to a community. In the
> beginning, I believe such communities were mostly made of local people
> and the Internet has sort of destroyed that. I believe people still
> long for these local communities again. BBSs connected local people.
> The Internet seems to have done the opposite.
>
> (*) On community building
>
> There's also this conjecture that a person can't deal with more than N
> people. The number I usually hear is 150. In other words, human
> communities should be small (and local).
>
> One principle I think about to achieve that is one begins a community
> but others can only come through invitation. If someone misbehaves, we
> know who invited that person---the tree of invitation being public.
>
> Yes, we lose anonimity. Perhaps anonimity is overrated.
>

I run a community for IT-professionals and have been doing so for close to
10 years. I think what you say is correct and that there definitely is a
need for it.

If you don't have one, start one! =)

D <nospam@example.net> writes:

> On Fri, 8 Mar 2024, Julieta Shem wrote:
>
>> Kaz Kylheku <433-929-6894@kylheku.com> writes:
>>
>>> On 2024-03-08, vallor <vallor@cultnix.org> wrote:
>>>> We like to give customers a soft landing when turning
>>>> down services. I'm wondering if anyone has suggestions
>>>> for good shell providers, places like Panix.
>>>
>>> That concept mostly went away 30 years ago with free Unix-like operating
>>> systems that run on low-spec consumer hardware.
>>>
>>>> "bare-bones" for the users I'm thinking of. Some of
>>>> them might be able to transition to being a system
>>>> administrator, but a lot just want to run tin, pine,
>>>> mutt, irssi, tf, and so forth.
>>>
>>> To access some remote shell account you need a machine
>>> that is internet connected and can run SSH. That machine
>>> can just run a freeware OS with all the above packages.
>>
>> That's all true, but it seems that people like to share a system
>> together because people love to belong to a community. In the
>> beginning, I believe such communities were mostly made of local people
>> and the Internet has sort of destroyed that. I believe people still
>> long for these local communities again. BBSs connected local people.
>> The Internet seems to have done the opposite.
>>
>> (*) On community building
>>
>> There's also this conjecture that a person can't deal with more than N
>> people. The number I usually hear is 150. In other words, human
>> communities should be small (and local).
>>
>> One principle I think about to achieve that is one begins a community
>> but others can only come through invitation. If someone misbehaves, we
>> know who invited that person---the tree of invitation being public.
>>
>> Yes, we lose anonimity. Perhaps anonimity is overrated.
>>
>
> I run a community for IT-professionals and have been doing so for
> close to 10 years. I think what you say is correct and that there
> definitely is a need for it.
>
> If you don't have one, start one! =)

That's great to know because I really am building one. Mine is also a
community of IT-professionals, so I started it out by writing the very
software that runs the communication. It works as a playground: people
enjoy hacking the very software that they and their peers use.

Can I joing your community? I promise to behave. :)

On Fri, 8 Mar 2024 09:46:32 -0000 (UTC), Kaz Kylheku
<433-929-6894@kylheku.com> wrote in <20240308013928.226@kylheku.com>:

> On 2024-03-08, vallor <vallor@cultnix.org> wrote:
>> We like to give customers a soft landing when turning down services.
>> I'm wondering if anyone has suggestions for good shell providers,
>> places like Panix.
>
> That concept mostly went away 30 years ago with free Unix-like operating
> systems that run on low-spec consumer hardware.

That turns out not to be the case.

Some shell users don't want to be system administrators.

>
>> "bare-bones" for the users I'm thinking of. Some of them might be able
>> to transition to being a system administrator, but a lot just want to
>> run tin, pine, mutt, irssi, tf, and so forth.
>
> To access some remote shell account you need a machine that is internet
> connected and can run SSH. That machine can just run a freeware OS with
> all the above packages.

...with all the system administration that entails.

I suppose the formal name for my request is recommendations for "managed
shell services".

Thanks to those who responded, sdf.org and the tildeverse are great
suggestions.

--
-v

vallor <vallor@cultnix.org> writes:
> On Fri, 8 Mar 2024 09:46:32 -0000 (UTC), Kaz Kylheku
> <433-929-6894@kylheku.com> wrote in <20240308013928.226@kylheku.com>:
[...]
> To access some remote shell account you need a machine that is internet
>> connected and can run SSH. That machine can just run a freeware OS with
>> all the above packages.
>
> ...with all the system administration that entails.
[...]

Or you can run a Windows machine with an ssh client such as PuTTY or
TeraTerm, or an Android or ChromeOS device with an ssh client, or a Mac
if that's your thing.

--
Keith Thompson (The_Other_Keith) Keith.S.Thompson+u@gmail.com
Working, but not speaking, for Medtronic
void Void(void) { Void(); } /* The recursive call of the void */

On Fri, 8 Mar 2024, Julieta Shem wrote:

> D <nospam@example.net> writes:
>
>> On Fri, 8 Mar 2024, Julieta Shem wrote:
>>
>>> Kaz Kylheku <433-929-6894@kylheku.com> writes:
>>>
>>>> On 2024-03-08, vallor <vallor@cultnix.org> wrote:
>>>>> We like to give customers a soft landing when turning
>>>>> down services. I'm wondering if anyone has suggestions
>>>>> for good shell providers, places like Panix.
>>>>
>>>> That concept mostly went away 30 years ago with free Unix-like operating
>>>> systems that run on low-spec consumer hardware.
>>>>
>>>>> "bare-bones" for the users I'm thinking of. Some of
>>>>> them might be able to transition to being a system
>>>>> administrator, but a lot just want to run tin, pine,
>>>>> mutt, irssi, tf, and so forth.
>>>>
>>>> To access some remote shell account you need a machine
>>>> that is internet connected and can run SSH. That machine
>>>> can just run a freeware OS with all the above packages.
>>>
>>> That's all true, but it seems that people like to share a system
>>> together because people love to belong to a community. In the
>>> beginning, I believe such communities were mostly made of local people
>>> and the Internet has sort of destroyed that. I believe people still
>>> long for these local communities again. BBSs connected local people.
>>> The Internet seems to have done the opposite.
>>>
>>> (*) On community building
>>>
>>> There's also this conjecture that a person can't deal with more than N
>>> people. The number I usually hear is 150. In other words, human
>>> communities should be small (and local).
>>>
>>> One principle I think about to achieve that is one begins a community
>>> but others can only come through invitation. If someone misbehaves, we
>>> know who invited that person---the tree of invitation being public.
>>>
>>> Yes, we lose anonimity. Perhaps anonimity is overrated.
>>>
>>
>> I run a community for IT-professionals and have been doing so for
>> close to 10 years. I think what you say is correct and that there
>> definitely is a need for it.
>>
>> If you don't have one, start one! =)
>
> That's great to know because I really am building one. Mine is also a
> community of IT-professionals, so I started it out by writing the very
> software that runs the communication. It works as a playground: people
> enjoy hacking the very software that they and their peers use.
>
> Can I joing your community? I promise to behave. :)

Well, if you're ever in Stockholm, sure you can! =) Everyone is welcome.
But the community only exists in the real world, and no streaming from the
events is allowed. The format is 3 lectures, one sponsored and two
community ones, and afterwards there is always burgers and beer. This
concept has worked well for almost 10 years, with between 20 and 130
visitors per event.

On Fri, 8 Mar 2024, Julieta Shem wrote:

> D <nospam@example.net> writes:
>
>> On Fri, 8 Mar 2024, Julieta Shem wrote:
>>
>>> Kaz Kylheku <433-929-6894@kylheku.com> writes:
>>>
>>>> On 2024-03-08, vallor <vallor@cultnix.org> wrote:
>>>>> We like to give customers a soft landing when turning
>>>>> down services. I'm wondering if anyone has suggestions
>>>>> for good shell providers, places like Panix.
>>>>
>>>> That concept mostly went away 30 years ago with free Unix-like operating
>>>> systems that run on low-spec consumer hardware.
>>>>
>>>>> "bare-bones" for the users I'm thinking of. Some of
>>>>> them might be able to transition to being a system
>>>>> administrator, but a lot just want to run tin, pine,
>>>>> mutt, irssi, tf, and so forth.
>>>>
>>>> To access some remote shell account you need a machine
>>>> that is internet connected and can run SSH. That machine
>>>> can just run a freeware OS with all the above packages.
>>>
>>> That's all true, but it seems that people like to share a system
>>> together because people love to belong to a community. In the
>>> beginning, I believe such communities were mostly made of local people
>>> and the Internet has sort of destroyed that. I believe people still
>>> long for these local communities again. BBSs connected local people.
>>> The Internet seems to have done the opposite.
>>>
>>> (*) On community building
>>>
>>> There's also this conjecture that a person can't deal with more than N
>>> people. The number I usually hear is 150. In other words, human
>>> communities should be small (and local).
>>>
>>> One principle I think about to achieve that is one begins a community
>>> but others can only come through invitation. If someone misbehaves, we
>>> know who invited that person---the tree of invitation being public.
>>>
>>> Yes, we lose anonimity. Perhaps anonimity is overrated.
>>>
>>
>> I run a community for IT-professionals and have been doing so for
>> close to 10 years. I think what you say is correct and that there
>> definitely is a need for it.
>>
>> If you don't have one, start one! =)
>
> That's great to know because I really am building one. Mine is also a
> community of IT-professionals, so I started it out by writing the very
> software that runs the communication. It works as a playground: people
> enjoy hacking the very software that they and their peers use.
>
> Can I joing your community? I promise to behave. :)
>

Oh, and tell me more about your community! What have you learned? Is it
flourishing? Do you intend to keep it small or grow?

On 2024-03-08, Julieta Shem <jshem@yaxenu.org> wrote:
> Kaz Kylheku <433-929-6894@kylheku.com> writes:
>
>> On 2024-03-08, vallor <vallor@cultnix.org> wrote:
>>> We like to give customers a soft landing when turning
>>> down services. I'm wondering if anyone has suggestions
>>> for good shell providers, places like Panix.
>>
>> That concept mostly went away 30 years ago with free Unix-like operating
>> systems that run on low-spec consumer hardware.
>>
>>> "bare-bones" for the users I'm thinking of. Some of
>>> them might be able to transition to being a system
>>> administrator, but a lot just want to run tin, pine,
>>> mutt, irssi, tf, and so forth.
>>
>> To access some remote shell account you need a machine
>> that is internet connected and can run SSH. That machine
>> can just run a freeware OS with all the above packages.
>
> That's all true, but it seems that people like to share a system
> together because people love to belong to a community.

Obviously, if you're using tin, you have your community in Usenet; pine
and mutt in e-mail, etc), irssi in IRC; TinyFugue in MUDs.

Those don't go away if you have your own machine to run them on.

I had some shell accounts some thirty years ago. I didn't know the other
users logged in.

If any of them hack root, they can see any private files you may have
in your account.

The era of secure multi-user computing is behind us; we now know that
processors cannot actually be trusted to enforce their documented
protection mechanisms like user/supervisor separation, due to side
channel attacks.

--
TXR Programming Language: http://nongnu.org/txr
Cygnal: Cygwin Native Application Library: http://kylheku.com/cygnal
Mastodon: @Kazinator@mstdn.ca

On 2024-03-08, vallor <vallor@cultnix.org> wrote:
> On Fri, 8 Mar 2024 09:46:32 -0000 (UTC), Kaz Kylheku
><433-929-6894@kylheku.com> wrote in <20240308013928.226@kylheku.com>:
>
>> On 2024-03-08, vallor <vallor@cultnix.org> wrote:
>>> We like to give customers a soft landing when turning down services.
>>> I'm wondering if anyone has suggestions for good shell providers,
>>> places like Panix.
>>
>> That concept mostly went away 30 years ago with free Unix-like operating
>> systems that run on low-spec consumer hardware.
>
> That turns out not to be the case.
>
> Some shell users don't want to be system administrators.

Yes, luckily, in those 30 years, we also got to the point where anyone
can install the stuff with their eyes closed, or even buy a system with
it preinstalled.

There is next to nothing to administer on a machine that just runs
client applications like tin or irssi.

>>> "bare-bones" for the users I'm thinking of. Some of them might be able
>>> to transition to being a system administrator, but a lot just want to
>>> run tin, pine, mutt, irssi, tf, and so forth.
>>
>> To access some remote shell account you need a machine that is internet
>> connected and can run SSH. That machine can just run a freeware OS with
>> all the above packages.
>
> ...with all the system administration that entails.

Next to zero. "sudo apt install mutt" is strawman system
administration.

--
TXR Programming Language: http://nongnu.org/txr
Cygnal: Cygwin Native Application Library: http://kylheku.com/cygnal
Mastodon: @Kazinator@mstdn.ca

On 2024-03-08, Keith Thompson <Keith.S.Thompson+u@gmail.com> wrote:
> vallor <vallor@cultnix.org> writes:
>> On Fri, 8 Mar 2024 09:46:32 -0000 (UTC), Kaz Kylheku
>> <433-929-6894@kylheku.com> wrote in <20240308013928.226@kylheku.com>:
> [...]
>> To access some remote shell account you need a machine that is internet
>>> connected and can run SSH. That machine can just run a freeware OS with
>>> all the above packages.
>>
>> ...with all the system administration that entails.
> [...]
>
> Or you can run a Windows machine with an ssh client such as PuTTY or
> TeraTerm, or an Android or ChromeOS device with an ssh client, or a Mac
> if that's your thing.

We are muddying the argument now. The argument is that if you have
a machine which can run an SSH client, chance are that instead of that,
you can just run a freeware distro where you have mutt, pine, irssi, tf,
and whatnot.

(I disagree that this constitutes "system administration".)

Yes; if you have Windows you can connect to a "managed shell account";
that is neither here nor there.

--
TXR Programming Language: http://nongnu.org/txr
Cygnal: Cygwin Native Application Library: http://kylheku.com/cygnal
Mastodon: @Kazinator@mstdn.ca

>
> From: Kaz Kylheku <433-929-6894@kylheku.com>
> Reply to: Kaz Kylheku
> Date: Sat, 9 Mar 2024 01:48:41 -0000 (UTC)
> Organization: A noiseless patient Spider
> Newsgroups:
> comp.unix.shell
> Followup to: newsgroup
> References:
> <usef6b$1iemc$1@dont-email.me>
> <20240308013928.226@kylheku.com>

> Kaz Kylheku <433-929-6894@kylheku.com> writes:
> ..
> The era of secure multi-user computing is behind us..

Only for you. I've got several shell account on various sites and people
I've been conversing with for decades now. Most of the newer sites
are much smaller, informally run, and not open to the general public.
I view this as a positive; IMHO authentic community doesn't scale.

-AI

On 3/8/24 19:48, Kaz Kylheku wrote:
> The era of secure multi-user computing is behind us;

Nonsense.

The era of the shell account's heyday is definitely behind us.

But -- ostensibly -- secure multi-user computing is alive and quite
well. RDP / VDI is very much a thing and those are multi-user computing.

My day job is supporting a farm of Solaris servers that clients log into
interactively to run applications.

> we now know that processors cannot actually be trusted to enforce
> their documented protection mechanisms like user/supervisor separation,
> due to side channel attacks.

Not quite true.

The optimizations that have been introduced cause problems.

But disabling those optimizations significantly restores trust.

Also, that trust is largely an x86 specific issue. There are other
processor architectures that don't have the same issues.

--
Grant. . . .

On 3/8/24 19:53, Kaz Kylheku wrote:
> We are muddying the argument now. The argument is that if you have
> a machine which can run an SSH client, chance are that instead of that,
> you can just run a freeware distro where you have mutt, pine, irssi, tf,
> and whatnot.

The Chrome OS devices that I've used would like to disagree with you.
It's trivial to get an SSH client. Doing so is even supported by the
hardware and software vendor. Trying to get to a shell on the system or
replace to OS thereon is definitely not supported.

--
Grant. . . .

On Fri, 8 Mar 2024, Grant Taylor wrote:

> On 3/8/24 19:48, Kaz Kylheku wrote:
>> The era of secure multi-user computing is behind us;
>
> Nonsense.
>
> The era of the shell account's heyday is definitely behind us.
>
> But -- ostensibly -- secure multi-user computing is alive and quite well.
> RDP / VDI is very much a thing and those are multi-user computing.
>
> My day job is supporting a farm of Solaris servers that clients log into
> interactively to run applications.
>
>> we now know that processors cannot actually be trusted to enforce their
>> documented protection mechanisms like user/supervisor separation, due to
>> side channel attacks.
>
> Not quite true.
>
> The optimizations that have been introduced cause problems.
>
> But disabling those optimizations significantly restores trust.
>
> Also, that trust is largely an x86 specific issue. There are other processor
> architectures that don't have the same issues.

I'm with Grant on this one. I've seen countless systems which support
multi-user scenarios throughout my life. I've worked as a system
administrator, but if you never had that role I imagine that perhaps it
would be easier to reach the conclusion of the end of the multi-user.

D <nospam@example.net> writes:

[...]

>>>> (*) On community building
>>>>
>>>> There's also this conjecture that a person can't deal with more than N
>>>> people. The number I usually hear is 150. In other words, human
>>>> communities should be small (and local).
>>>>
>>>> One principle I think about to achieve that is one begins a community
>>>> but others can only come through invitation. If someone misbehaves, we
>>>> know who invited that person---the tree of invitation being public.
>>>>
>>>> Yes, we lose anonimity. Perhaps anonimity is overrated.
>>>>
>>>
>>> I run a community for IT-professionals and have been doing so for
>>> close to 10 years. I think what you say is correct and that there
>>> definitely is a need for it.
>>>
>>> If you don't have one, start one! =)
>>
>> That's great to know because I really am building one. Mine is also a
>> community of IT-professionals, so I started it out by writing the very
>> software that runs the communication. It works as a playground: people
>> enjoy hacking the very software that they and their peers use.
>>
>> Can I joing your community? I promise to behave. :)
>>
>
> Oh, and tell me more about your community! What have you learned? Is
> it flourishing? Do you intend to keep it small or grow?

Not quite flourishing, but it's been overall encouraging with some
elements of pessimism. I don't think I've learned anything yet. I'd
like to see it growing to the point I can't keep up with hanging out
with everyone at the same time, but it shouldn't grow too big so the
group loses its cohesion.

With some experience in programming and knowing everyone claims interest
in programming, I decided to do it my way. I wrote a prototype of an
NNTP server. An NNTP server is a world of opportunities for programming
and creativity and community experiments. Instead of showing them that
I wrote the system, I merely invited some of my friends to discuss
programming.

(*) The idea

A community should make the members feel they belong. For a group of
programmers, understanding and writing a piece of the system helps them
to feel like they belong. An NNTP server is something very simple, so
people can quickly learn how it works and can implement something of
their own ideas. Whatever programming is done affects the whole
community---bugs, say.

Users can only join through invitation.

--8<---------------cut here---------------start------------->8---
200 Welcome! Say ``help'' for a menu.
help
[...]
create-account my-friend
200 Okay, account MY-FRIEND created with password ``mbnxgf''.
--8<---------------cut here---------------end--------------->8---

Now you can give your friend an account. If your friend misbehaves, we
can all know who invited her. We'll have no spam problem, say. Every
user has a real-world connection to the community, even if a random
person on the Internet was invited. (The invitations make up a tree.)

--8<---------------cut here---------------start------------->8---
users
200 List of current users:
ROOT, last seen on Fri Mar 8 22:03:00 2024, invited (X)
X, last seen on Sat Mar 9 11:23:21 2024, invited (MY-FRIEND J)
J, never seen, invited nobody
MY-FRIEND, never seen, invited nobody
..
--8<---------------cut here---------------end--------------->8---

To feel like they belong, users can create their own groups. (Crazy,
huh?)

--8<---------------cut here---------------start------------->8---
create-group comp.my.favorite.topic
280 group comp.my.favorite.topic created
--8<---------------cut here---------------end--------------->8---

People can now subscribe to this group. The fact that a new group was
created is posted to a control group. So everyone can keep an eye on
what's going on in the community. This control group can receive the
usual posts from people, so that people can discuss the server event
right where it was announced.

Accounts can be disabled, except for ROOT. You can only disable an
account if you are a root of the tree of invitation. For example, X
invited J and MY-FRIEND, so X can disable both of these accounts. If J
invites Z, then J would be able to disable the Z account not X nor
MY-FRIEND or any other. (X would be able to disable Z as well.) So
users can invite their own subcommunities and be their sysadmin, say.
(If someone loses their password, anyone up in the tree of invitation
can recover it for them.)

After three months, if an account has not posted anything to the server,
that account is automatically disabled. The idea in this is to keep the
community tight: you can feel comfortable speaking out your ideas
because the community is fairly closed and you have an intuition of who
is reading you. You know that someone who is there only to watch would
actually not have an account (after a while). That's why everyone has
to post, so that you can have a sense of who the person is. (Those who
do not get interested in the groups and eventually forget about the
whole thing will just get disabled. If they regret losing the account,
they will need to ask someone to reenable them.)

That's an experiment. It's hard to come in (as you need to be close to
someone who is in) and, once you're in, you must commit: if you were
only curious to see what goes on, that's fine, but you'll get
automatically expelled eventually. I don't know how long should the
inactivity period be.

If not anything else, it's a programming playground. There are so many
ideas one can decide to implement. For instance, we could implement
anonymous groups where every message posted to it gets the FROM-header
replaced with a random name (and only the necessary headers go with the
message). We could have a group that's the opposite, where the FROM
header is replaced with the nntp-username of the poster. There are /so/
many other ideas one could try out.

D <nospam@example.net> writes:

[...]

>>> I run a community for IT-professionals and have been doing so for
>>> close to 10 years. I think what you say is correct and that there
>>> definitely is a need for it.
>>>
>>> If you don't have one, start one! =)
>>
>> That's great to know because I really am building one. Mine is also a
>> community of IT-professionals, so I started it out by writing the very
>> software that runs the communication. It works as a playground: people
>> enjoy hacking the very software that they and their peers use.
>>
>> Can I joing your community? I promise to behave. :)
>
> Well, if you're ever in Stockholm, sure you can! =)

So far away. :)

> Everyone is welcome. But the community only exists in the real world,
> and no streaming from the events is allowed. The format is 3 lectures,
> one sponsored and two community ones, and afterwards there is always
> burgers and beer. This concept has worked well for almost 10 years,
> with between 20 and 130 visitors per event.

A sponsored lecture. That's an amazing idea.

"Arti F. Idiot" nobody@invalid.invalid writes:

>>
>> From: Kaz Kylheku <433-929-6894@kylheku.com>
>> Reply to: Kaz Kylheku
>> Date: Sat, 9 Mar 2024 01:48:41 -0000 (UTC)
>> Organization: A noiseless patient Spider
>> Newsgroups:
>> comp.unix.shell
>> Followup to: newsgroup
>> References:
>> <usef6b$1iemc$1@dont-email.me>
>> <20240308013928.226@kylheku.com>

Hey. I loved your quote-introduction here! You must be running Gnus?
Can you share your code so that I can do the same? (I've been lately
looking around which variable I should set so that I can change the
``writes:''-string when I follow up, but did not even find that in the
Gnus documentation. I know it's there somewhere.)

>> Kaz Kylheku <433-929-6894@kylheku.com> writes:
>> ..
>> The era of secure multi-user computing is behind us..
>
> Only for you. I've got several shell account on various sites and people
> I've been conversing with for decades now. Most of the newer sites
> are much smaller, informally run, and not open to the general public.
> I view this as a positive; IMHO authentic community doesn't scale.

Precisely. An authentic community doesn't scale. I only came to this
conclusion very recently.

Grant Taylor <gtaylor@tnetconsulting.net> writes:
> On 3/8/24 19:48, Kaz Kylheku wrote:
>> The era of secure multi-user computing is behind us;

It was never all that secure even considering the software alone l-)

>> we now know that processors cannot actually be trusted to enforce
>> their documented protection mechanisms like user/supervisor
>> separation, due to side channel attacks.
>
> Not quite true.
>
> The optimizations that have been introduced cause problems.
>
> But disabling those optimizations significantly restores trust.
>
> Also, that trust is largely an x86 specific issue. There are other
> processor architectures that don't have the same issues.

Which did you have in mind? Arm, Power and (I think) MIPS have all had
microarchitectural security flaws too. The mitigations are mostly more
complex than just disabling the optimizations, which would result in
unusably slow computers.

--
https://www.greenend.org.uk/rjk/

On Sat, 9 Mar 2024, Julieta Shem wrote:

> D <nospam@example.net> writes:
>
> [...]
>
>>>>> (*) On community building
>>>>>
>>>>> There's also this conjecture that a person can't deal with more than N
>>>>> people. The number I usually hear is 150. In other words, human
>>>>> communities should be small (and local).
>>>>>
>>>>> One principle I think about to achieve that is one begins a community
>>>>> but others can only come through invitation. If someone misbehaves, we
>>>>> know who invited that person---the tree of invitation being public.
>>>>>
>>>>> Yes, we lose anonimity. Perhaps anonimity is overrated.
>>>>>
>>>>
>>>> I run a community for IT-professionals and have been doing so for
>>>> close to 10 years. I think what you say is correct and that there
>>>> definitely is a need for it.
>>>>
>>>> If you don't have one, start one! =)
>>>
>>> That's great to know because I really am building one. Mine is also a
>>> community of IT-professionals, so I started it out by writing the very
>>> software that runs the communication. It works as a playground: people
>>> enjoy hacking the very software that they and their peers use.
>>>
>>> Can I joing your community? I promise to behave. :)
>>>
>>
>> Oh, and tell me more about your community! What have you learned? Is
>> it flourishing? Do you intend to keep it small or grow?
>
> Not quite flourishing, but it's been overall encouraging with some
> elements of pessimism. I don't think I've learned anything yet. I'd
> like to see it growing to the point I can't keep up with hanging out
> with everyone at the same time, but it shouldn't grow too big so the
> group loses its cohesion.

This is in line with my experience. If it grows too big it becomes too
anonymous, and also, it becomes too much of a chore to do on a non-profit
basis. I try to limit my event to max 50 people. The one that reach 130
was like a non-paid job, so unless the sponsor helps out I don't do those
anymore.

Sounds like a great project! =)

I'm not a programmer, but my latest tinkering success was integrating the
fetchnews part of leafnode directly with alpine (my mail and news reader)
_without_ having to run a lokal nntp server.

This speeded up my news reading enormously since I just move files around
from the news spool right into alpine. =)

Best regards,
Daniel

>
> With some experience in programming and knowing everyone claims interest
> in programming, I decided to do it my way. I wrote a prototype of an
> NNTP server. An NNTP server is a world of opportunities for programming
> and creativity and community experiments. Instead of showing them that
> I wrote the system, I merely invited some of my friends to discuss
> programming.
>
> (*) The idea
>
> A community should make the members feel they belong. For a group of
> programmers, understanding and writing a piece of the system helps them
> to feel like they belong. An NNTP server is something very simple, so
> people can quickly learn how it works and can implement something of
> their own ideas. Whatever programming is done affects the whole
> community---bugs, say.
>
> Users can only join through invitation.
>
> --8<---------------cut here---------------start------------->8---
> 200 Welcome! Say ``help'' for a menu.
> help
> [...]
> create-account my-friend
> 200 Okay, account MY-FRIEND created with password ``mbnxgf''.
> --8<---------------cut here---------------end--------------->8---
>
> Now you can give your friend an account. If your friend misbehaves, we
> can all know who invited her. We'll have no spam problem, say. Every
> user has a real-world connection to the community, even if a random
> person on the Internet was invited. (The invitations make up a tree.)
>
> --8<---------------cut here---------------start------------->8---
> users
> 200 List of current users:
> ROOT, last seen on Fri Mar 8 22:03:00 2024, invited (X)
> X, last seen on Sat Mar 9 11:23:21 2024, invited (MY-FRIEND J)
> J, never seen, invited nobody
> MY-FRIEND, never seen, invited nobody
> .
> --8<---------------cut here---------------end--------------->8---
>
> To feel like they belong, users can create their own groups. (Crazy,
> huh?)
>
> --8<---------------cut here---------------start------------->8---
> create-group comp.my.favorite.topic
> 280 group comp.my.favorite.topic created
> --8<---------------cut here---------------end--------------->8---
>
> People can now subscribe to this group. The fact that a new group was
> created is posted to a control group. So everyone can keep an eye on
> what's going on in the community. This control group can receive the
> usual posts from people, so that people can discuss the server event
> right where it was announced.
>
> Accounts can be disabled, except for ROOT. You can only disable an
> account if you are a root of the tree of invitation. For example, X
> invited J and MY-FRIEND, so X can disable both of these accounts. If J
> invites Z, then J would be able to disable the Z account not X nor
> MY-FRIEND or any other. (X would be able to disable Z as well.) So
> users can invite their own subcommunities and be their sysadmin, say.
> (If someone loses their password, anyone up in the tree of invitation
> can recover it for them.)
>
> After three months, if an account has not posted anything to the server,
> that account is automatically disabled. The idea in this is to keep the
> community tight: you can feel comfortable speaking out your ideas
> because the community is fairly closed and you have an intuition of who
> is reading you. You know that someone who is there only to watch would
> actually not have an account (after a while). That's why everyone has
> to post, so that you can have a sense of who the person is. (Those who
> do not get interested in the groups and eventually forget about the
> whole thing will just get disabled. If they regret losing the account,
> they will need to ask someone to reenable them.)
>
> That's an experiment. It's hard to come in (as you need to be close to
> someone who is in) and, once you're in, you must commit: if you were
> only curious to see what goes on, that's fine, but you'll get
> automatically expelled eventually. I don't know how long should the
> inactivity period be.
>
> If not anything else, it's a programming playground. There are so many
> ideas one can decide to implement. For instance, we could implement
> anonymous groups where every message posted to it gets the FROM-header
> replaced with a random name (and only the necessary headers go with the
> message). We could have a group that's the opposite, where the FROM
> header is replaced with the nntp-username of the poster. There are /so/
> many other ideas one could try out.
>

On Sat, 9 Mar 2024, Julieta Shem wrote:

> D <nospam@example.net> writes:
>
> [...]
>
>>>> I run a community for IT-professionals and have been doing so for
>>>> close to 10 years. I think what you say is correct and that there
>>>> definitely is a need for it.
>>>>
>>>> If you don't have one, start one! =)
>>>
>>> That's great to know because I really am building one. Mine is also a
>>> community of IT-professionals, so I started it out by writing the very
>>> software that runs the communication. It works as a playground: people
>>> enjoy hacking the very software that they and their peers use.
>>>
>>> Can I joing your community? I promise to behave. :)
>>
>> Well, if you're ever in Stockholm, sure you can! =)
>
> So far away. :)

Where are you based? The US?

>> Everyone is welcome. But the community only exists in the real world,
>> and no streaming from the events is allowed. The format is 3 lectures,
>> one sponsored and two community ones, and afterwards there is always
>> burgers and beer. This concept has worked well for almost 10 years,
>> with between 20 and 130 visitors per event.
>
> A sponsored lecture. That's an amazing idea.

If you have good technologists, companies will _love_ to sponsor in the
hope of either

1. Being able to hire visitors.
or
2. Being able to advertise their service.

So usually I ask them to sponsor with venue (being in their office if it's
a good one) and/or buy burger + beer for everyone, which is quite cheap.
Where I live it's about 25 USD per person and if you just buy the alcohol
from the closest store, it's even cheaper than that.

At my last event we had a food truck park outside the event to serve the
burgers, and bought the beer from the super market close by. Cheap, warm
burgers and highly appreciated!

In comp.unix.shell, Kaz Kylheku <433-929-6894@kylheku.com> wrote:
> Yes, luckily, in those 30 years, we also got to the point where anyone
> can install the stuff with their eyes closed, or even buy a system
> with it preinstalled.
>
> There is next to nothing to administer on a machine that just runs
> client applications like tin or irssi.

With Panix I am not just getting a system with some client programs
installed. I have maintained software for hosting a website. I have mail
service available as webmail or as command line reader. I have a local
newserver with Usenet and private hierarchy groups. I have a .plan file
someone can "finger". I have a system where ssh answers on both port 22
and port 443 (https and ssh differ in who talks first, so the port can
be multiplexed with a slight delay). And where there's a web wrapper to
get to a shell from a browser window. I have maintained backups. I have
multiple versions of utilities where that makes sense (ruby, python,
perl, php, etc) and phased in replacements where "tool.new" is the new
version for a week for testing and only becomes "tool" if no one raises
issues.

I don't get that stuff on my self-managed virtual machines which I also
use.

Elijah
------
whose panix account is older than this upstart google thing