Grex Town Hall -- How do we move forward? - Fall 2015/Winter 2016

Dec 31 16:45 UTC 2015

This item is for capturing opinions and ideas about how Grex can move
forward, what Grex can do to encourage more people to use the system,
and what we want Grex to provide in terms of services or applications.

response 2 of 40:

Jan 5 22:34 UTC 2016

I find myself with a little time, for the first time in years. So where should
I begin? I have never studied authentication models, but that seems to be
where effort is needed. Any suggestions for background reading?

response 4 of 40:

Jan 13 18:30 UTC 2016

I know cross redid the grex password hash a few years ago. Is that what you
mean? 

response 6 of 40:

Jan 13 23:29 UTC 2016

So what's needed with PAM?

response 8 of 40:

Jan 15 15:15 UTC 2016

You mean on FreeBSD?

The issue is that FreeBSD stores encrypted passwords in a file
that's owned by root and only readable by root. Since PAM is a
shared library that's linked into every program that wants to use
it for authentication, PAM modules are executed with the permissions
of the program that's using them. For the web server, we want this
to NOT be root. So the question becomes, how does the web server
authenticate users for things like backtalk and RT? Most large-scale
systems don't use the traditional password database, deferring
instead to some kind of network directory service that contains
authentication data (e.g. LDAP).

On OpenBSD, this isn't an issue since OpenBSD's 'auth' library
invokes a setuid program to check passwords (and talks to it over
a pipe or socket; I forget which).

response 10 of 40:

Oct 25 05:54 UTC 2016

I have no idea if my comments and ideas are welcome here, most of what is
being said appears to be internal discussion. Still, this is a semi-public
board, so I'll throw in what I can.

First of all, it's difficult to tell what Grex has to offer to the average
user. A shell account, sure, but I have to assume that many *nix users that
might want to participate have their own shell accounts on their own systems.
I for one have several, in various places, and don't necessarily need another.
I came here for the bbs, and to connect with other potentially like-minded
*nix users.

My second observation is that Grex feels very Michigan-ish. That's not a
horrible thing in and of itself, but if you're aiming for a user base that
is larger than what MI has to offer, I would suggest that you diversify your
discussions. Which brings me to...

Number 3: the bbs. The main reason I came here. I'm a vetran, used to program
and run BBSes back in the POTS days. I've been out of it for years, but have
recently re-connected with this type of crowd. It is interesting that they
live on, and beautiful in a way. I was excited and willing, but I have to say,
what you have going here is a beast. The crowd that would enjoy agora is a
small one indeed, and some other solution might receive far more
participation. I hope I'm not completely stomping on someone's baby by saying
so, but Agora is more of a pain in the neck than it is worth the content.

So to sum up, if you truly want feedback, and if you truly want a larger user
base, I'd suggest testing the waters with a different bbs, and making your
offerings (especially multi-user-centric offerings, and perhaps inclusive
online social offerings) more clear. 

Thanks for listening, and I look forward to being set on the right track if
I'm off topic or off course in this post!

response 12 of 40:

Oct 27 17:26 UTC 2016

I'm not sure "better" is the right goal, but more friendly perhaps, keeping
in mind that not every user in the potential target market will have a lot
of experiences with text-based social interactions. The entire purpose of a
BBS is to create interaction; a system is worthless if it doesn't have users
filling it with content, and interacting.

Since Grex is an OpenBSD system, I have no idea what systems would even be
available. Perhaps a custom-built one? I have no BSD experience, only Linux
sadly, so I can't recommend something, but I do stand by my previous post.
Make the offerings more understandable, and improve the social (bbs) aspects.

I'll throw this out, though I don't necessarily recommend it. It's written
in PHP, which I don't see here on grex. I'd be willing to re-write it in perl
as a custom version for grex if anyone was interested. It's a clone-of-sorts
of the bboard system that is found on SDF (which is quite active/popular even
with relatively novice users): 

https://sourceforge.net/projects/slightbbs/

It runs through inetd, but can be accessed at the command line just as easily
(at least on my system.) Again, I don't necessarily recommend it, but it is
an option. I'd love to know if there were better options available on the open
market.

response 14 of 40:

Oct 27 17:48 UTC 2016

didn't work for me

response 16 of 40:

Oct 29 11:55 UTC 2016

php 5.2 and 5.3 are installed on Grex.  See /usr/local/bin.
 
The main roadblock to getting anything going like this is staff time and
willingness to implement/install.  What we have for a bbs now is one
that was written by Grex staff a long time ago now.  It is Perl-based
which makes it relatively portable.  It might be possible to add a new
look or new functionality to it, but someone would need to spend the
time to figure out how it all works.

I think a new look and new functionality would be good and might attract
more users to grex.  However, you'll find there will be complaints
about things like users being able to edit a response, moderation of
responses, and such, that they are not used to or have philosophical
objections to.

At some point we just need to try something new.  We could run two bbses
in parallel, for example.  Then people who don't want to try new things
can not respond to the new system as well as the old system.  

phpBB I think was another php-based system that has been discussed in
the past.  In the OpenBSD ports files on Grex there is information on
fluxBB. There are a number of them available for Unix-based systems
in general.  Installation hassle, security and staff maintenance time
while in operation would be considerations for any such system.  We are
perennially short of staff time to do things.  

Keep making suggestions!  Maybe something will spark some interest in a
staff member to install a new bbs.  Or maybe someone will step up and
help staff with this.

response 18 of 40:

Oct 31 14:01 UTC 2016

I say go for it. If you can do it in your account, more's the better;
if it grows, we can make it system-wide.

It seems like you're talking about something like Mystic or Synchronet
(or MBSE or EleBBS or something) as a BBS? That may be interesting,
but my sense is that most of those BBS's just exist to pass around
the same echonet traffic that all the others do (e.g., FidoNet etc).

Personally, I'm not so interested in the BBSing aspects of Grex
anymore. I'd rather see us become a playground for programmers and
makers of varous stripes. Consider, for example, the maker communities
that have sprung up around designing retro single board computers
using Z80's and 68k's. Those folks are using web BBSes for most of
their collaboration. Why not a real Unix system like Grex?

But hey! If you've got an idea, time and motivation, run with it and
make it your own!