response 170 of 334:

Nov 26 20:56 UTC 2010

I think we should at least restart some of the older and cruftier conferences.
For instance, the "micros" conference; really, when was the last time someone
called a desktop computer a "microcomputer"?  Restarting the systems (nee
jellyware) conference resulted in a flurry of new activity for a few months,
though that's tappered off now; I sort of look at doing some selective
conference pruning as a way of repairing some of the more glaringly broken
windows on Grex.

resp:167 I think the thing that's likely to keep me interested is if Grex
remains an interesting place technically.  While the technology has gotten
kind of boring in the last few years, I think the idea of it remains
sufficiently compelling that I stick around to poke at things and see where
it can be taken.  The place definitely needs a facelift.

response 172 of 334:

Nov 26 21:05 UTC 2010

Sure.  But don't make that the main source for information about modern PCs.

response 174 of 334:

Nov 29 15:01 UTC 2010

For whatever reason, they don't seem to have the same problems we do 
with people attacking the server as soon as they get an account.  I 
don't know quite why that is, but I suspect that some of it is that 
it's just easier to push people's buttons on Grex, and the 
personalities that have been most disruptive over the past few years 
seem to get a kick out of doing that.

response 176 of 334:

Nov 29 17:26 UTC 2010

resp:175 The probability of such a thing is roughly zero.

We took action in the form of asking Chad to stop and locking account 
after account.  We talked about talking to law enforcement about him, 
but it didn't happen.  M-Net asked him to leave, and he did for a 
while, but then came back.

What sort of gets me is that the people who are saying we should do 
away with validation and so on are missing several points.

One is that Grex has always done some level of validation; from the 
time this place has been connected to the Internet, there's been an 
identity verification requirement to access more than just basic 
Internet services (e.g., do hostname lookups).  Not only is there 
precedent for this sort of thing, but member votes were taken back in 
the day and so on.  I don't think this is going to change any time 
soon.

The other is that people seem to have forgotten just what a mess Grex 
was for a couple of years prior to putting the current scheme into 
place.  Some people were howling for action (some who are now 
suggesting we get rid of validation).  Grex would be down or unusable 
for weeks or months at a time.  It was a sad situation....  What we 
got out of validation was some modicum of stability.  We put a speed 
bump in the road to getting to an unrestricted shell that set the bar 
high enough that the ankle biters got bored and stopped trying.  Now, 
we're talking about streamlining things further, but I don't think we 
should remove the speed bump: experience has shown us that without it, 
we get taken down.  And it wasn't just Chad: spammers were using Grex 
like it was going out of style and our ISP nearly shut us down several 
times as a result unless we took some action.  People regularly 
uploaded flooders and IRC bouncers and all sorts of stuff.  While the 
latter isn't particularly harmful, I don't think, the number of times 
I saw `udp.pl' in home directories was astounding.  Basically, the 
bottom line is that Grex was regularly being used as a platform to 
either attack, or attempt to attack, other systems elsewhere on the 
Internet.  We slowly narrowed things down further and further until 
brand new users got basically no access to the network at all, then 
they started attacking Grex itself.

Another issue is that of staff time and availability.  Grex's staff is 
much smaller than it used to be, and folks generally have way less 
time than they did previously.  It may be sad, but it's reality, and 
it begs the question, if we let people get on with no road blocks 
again, then who's going to clean up the resulting messes?  The 
spammers and crackers aren't going to pony up a couple of bucks to get 
to the shell when they can pay the same amount of money to rent a 
botnet for a few minutes (and send out an order of magnitude more 
email and/or flood packets).  If it cuts down on the amount of time I 
*have* to spend to keep Grex up and running, that's a good thing and I 
can use the limited time I have to play with Grex to do more 
interesting stuff, like improve software or services.

Finally, I believe that some of the conclusions people have drawn 
about validation are flawed.  It's true that introduction of the 
restricted shell roughly corresponded to a general decline in usage of 
the BBS and party, but correlation is not causation and, as I've 
posted before, the restricted shell came as a result of abuse of Grex, 
the BBS and party: I claim it was largely this abuse that caused the 
decline, not the validation speed bump.  There is less activity on 
Grex now than there was years ago, it's true, but I'd say that the 
level of actual, real discussion hasn't changed that much since the 
restricted shell got put in.  It's easy to mistake turbulent and/or 
destructive activity as just "activity"; when it goes away, things 
seem to slow down a lot (because they do) but what's there instead is 
what people are actually actively engaged in, not just observations 
about damage to the system.

It's easy to point a finger at the restricted shell, but look at what 
else is going on.  I mean, for how many *years* did the web site say 
that Grex was running on a Sun after we had transitioned to the 
OpenBSD machine?  How many dead links are still there?

Arbornet may not have a restricted shell, but SDF does, yet they get 
new users all the time.  Most of what I'm proposing is modelled on 
what they do.  It works there; why can't it work here?  I think it can.

I think that Grex's problems are a combination of technical and 
social, but the validation issue is somewhat separate.  Here are some 
technical things that I think could help out, if people were so 
inclined:

a) Update the web site.  Hey, we need a webmaster!  Anyone want to 
volunteer?  What's there now isn't terrible (I went through and 
converted almost all of the web pages to XHTML and cleaned up a lot of 
cruft a few years ago, but it's actually quite a large job).  
Specifically, if someone were so inclined, they could check out the 
web site from the Subversion repository (svn co /var/svn/trunk/grex) 
and make updates to the XHTML and CSS and send a patch.  A consistent 
look and feel across all pages would be great.  Making the CSS more 
generic would also be great (e.g., using relative percentages for 
things like widths instead of hardcoded numbers of pixels).  Pruning 
and/or updating dead links.  Making more of the content correspond to 
reality (e.g., the FAQs and so on).  Just proof-read the pages and 
look for grammar and spelling mistakes.

b) There are a lot of scripts in grexsoft that date from the Suns.  
Update these and make them reflect reality.  A lot of shell scripts 
are legitimately better written as Perl scripts.  Remove SunOS-isms 
(no, really).  A lot of scripts can just be straight-up deleted.

c) I think most of the C programs are good to go at this point, but 
verify this and clean things up as needed.  Someone who could write 
regression and/or unit tests for some of this stuff would be awesome.

d) Grex runs a few non-trivial software packages that need serious 
cleanups.  At this point, the big three are backtalk, fronttalk and 
party.  The former two need some serious work: backtalk needs some 
serious updates to make it compliant with web standards and make sure 
the HTML it generates is valid, etc.  The user interface(s) probably 
needs an overhaul.  Fronttalk is written in a pretty clean style, but 
is basically Perl 4, uses a lot of global variables, etc.  Cleaning 
that up would probably make it faster and fix some latent bugs.  Party 
has lots of legacy cruft that could be stripped out, again making it 
cleaner and probably a little faster.

Anyone who wants to work on this stuff is welcome; pretty much 
everything is in the subversion repository.  Anyone who wants can 
check it out and start making changes.  Send me diffs.

response 180 of 334:

Nov 29 20:51 UTC 2010

But what I don't get is, what's the point of turning it off?  It's not 
like there are hordes of new users clamouring to get accounts on Grex; 
until Grex has something to offer those people, they're not going to 
come here, regardless of whether they have to go through validation.

You could argue that what they come for is the shell, but I haven't 
seen any evidence of that.

response 182 of 334:

Nov 29 22:16 UTC 2010

One way to entice new users is to upgrade Backtalk or move to other 
conferencing software.  These days people don't want to participate in 
conferencing/bbs's unless they can upload graphics and show pictures 
and video in their messages.  

response 184 of 334:

Nov 30 00:09 UTC 2010

Ha!  That's a neat idea.  Unfortunately, I'm told our last inventory
of products ended up being disposed of after traveling to Europe where
they wouldn't sell, either.  Maybe times are different and if we limited
ourselves to very few items we'd do okay?  But we probably don't want to
repeat history if things won't sell.  What do others think?

response 186 of 334:

Nov 30 01:55 UTC 2010

resp:180 1. from what I've seen on SDF, people are always asking for 
MySQL and CGI. One guy I spoke to suggested it might be for testing 
purposes since he was completely clueless. But he did pay for arpa or 
whatever..
2. People need a friendly push when it comes to such things.. like
sample  scripts tailor made for Grex (a sample home-page with some
random cute  girl/or guys, pic and a wall and crud like that). Then they
get ideas  like Oo let me try that.. a decent home page is something no
one offers..  which is weird because look at Facebook :) 3. We have so
far never tried getting things going in a well organised  manner.. ads,
and features together.

response 188 of 334:

Nov 30 10:07 UTC 2010

Oo! yup! good idea! Actually i was thinking a cool cgi form or a shell 
script.. and it could build his home-page and .plan simultaneously. I 
could template that so it looks very nice (the home-page). The home-page
 could also double as a resume so ppl looking for jobs. .. you know,
free- email.. resume on the web.. cool-domain-name.. some
work-experience with  cyberspace.. it would be pretty useful to ppl!

response 190 of 334:

Nov 30 15:16 UTC 2010

I've streamlined newuser and it only asks a few basic questions now; 
it no longer asks about things like backspace characters or terminal 
types.  It now only asks questions about the person itself.  We can 
reduce those further.  For instance, do we really need people's phone 
numbers and addresses?  Probably not.  I'm not sure that's a huge 
hurdle, though.

It strikes me that newuser has to ask some questions just to set up 
the account: desired login name, the user's real name, an email 
address.  I'm not sure what the set of reasonable questions is, 
however.

response 192 of 334:

Nov 30 17:22 UTC 2010

We don't have an account management page and we definitely should.

I'm not so sure about asking for phone number, though.  Email address, 
definitely, but I think it is extraordinarily unlikely that someone 
would get a phone call, at least from me.  That said, I don't think it 
hurts anything to ask for it, but I think realistically it's not going 
to be used.

response 194 of 334:

Dec 1 14:33 UTC 2010

Being something of a minimalist, I'd personally prefer a newuser that's
short and sweet.  I'll think about what I'd like to see newuser say/ask
and post it.

Maybe discussion of newuser deserves its own item, particularly if we're
going to have go-rounds about what it should say and ask, to avoid
swamping the Grex Town Hall.  I'm glad we actually have a new newuser
to discuss, that does some sensible things that bring it more into line
with current practices on the web - specifically the email verification
and the captcha thing.