|
|
| Author |
Message |
| 25 new of 118 responses total. |
cross
|
|
response 50 of 118:
| 
Feb 21 03:02 UTC 2007 |
Regarding #47; How much spam you get each day is irrelevant. The fact that
you like pine is irrelevant. It's great for you and all, but 99.999% of the
rest of the world would really rather do it the way that, well, 99.999% of
the world does it. And I've done timing tests on my home computer showing
that, in fact, gmail is faster than you dialing into grex and starting pine.
If you want to do your think, go right ahead; no one is stopping you or
telling you not to. But please, stop acting like it's `better' some how.
You know what I don't understand about you, Sindi? You really, truly, just
seem *incapable* of understanding that, while *you* don't like GUI's,
webmail, fast comuters, or any of the rest of it, other people do. Your
capacity for just not acknowledging that is fascinating and all.
The reality of the matter is this: grex email is not going to change very
much in the near term. Space isn't the issue, quotas aren't the issue, none
of the things you seem to have infinite time to post about are the issue.
The issue staff time to babysit email. So you might as well get used to the
idea that that's just not going to change any time soon: we don't have the
staff for it, and even if we did, honestly? No one wants to work on email.
It's a `done' thing and it's tedious and annoying to set up correctly. If
you really want to use pine for email, then I highly recommend you start
using SDF for it, because it's just not going to change on grex, no matter
how much you keep posting about it nor how much you complain about it in the
BBS.
Now let me say this: I'm sorry that this is the case. I really am. But the
resources just aren't there to fix it right now, and quite frankly, you're
just going to have to accept that because it's reality. You accept gravity;
this is the same. I've offered to do some of the work, but I need
administrative access to do that, and I don't know what's up with that. As
a board member, I won't strong arm the staff into giving it to *me*, and I
really don't know if they want me doing that kind of work or not: it's up to
them. But as of right now? No one has the time and inclination necessary
to make email on grex work.
If you really insist on using grex for email, then I suggest you be happy
that it works right now. Long term, the solution is to make email opt-in
for new users, enforce quotas, etc. I suspect that few enough users will
want to use grex for email that we can then run spamassassin et al for
everyone by default.
|
mcnally
|
|
response 51 of 118:
|
Feb 21 05:53 UTC 2007 |
Here's a suggestion for a very simple opt-in scheme for incoming
mail for new users.
1) modify the account creation program so that it creates a
.forward file that invokes vacation and a a .vacation.msg
indicating that mail to the account is not being delivered
until the recipient opts in to mail delivery. Initialize
the vacation db files before exiting. Frankly the vacation
part can be left out in favor of a .forward that delivers
to /dev/null if desired, but I kind of prefer a solution
that doesn't swallow mail silently, at least at first.
2) create a "mail-opt-in" program that compares the default
.vacation.msg file and .forward file to the ones in the
user's home directory and removes the user's files if both
match the default.
3) create a "mail-opt-out" program (if anyone thinks there
would be demand for it) that backs up any existing
.forward and .vacation.msg files for the user, then
creates new ones with the default opt-out configuration.
I haven't looked at newuser for a while but I think that'd be
very easy to implement and it would do a reasonable job of
doing what's desired.
As for Sindi, if she wants a way for people to opt out of mail
delivery on Grex, she can advise them to execute the following
commands in the shell:
[ -f ~/.forward ] && mv ~/.forward ~/forward.$$; echo "/dev/null" > ~/.forward
|
glenda
|
|
response 52 of 118:
|
Feb 21 11:58 UTC 2007 |
I'm sorry, but I am not going to spend my time looking into users' accounts
to see if they have a .forward or not. Not my business, or yours, what files
a user has or doesn't have as long as they aren't doing something to harm the
system. I will not be going into user accounts to check to see the last time
they read their email to tell whether they have accessed within a certain
amount of time or not. Again, I see that as an invasion of privacy tantamount
to looking into a persons physical mailbox to see if they are taking their
mail into their house or not, and how much of what appears in that box is junk
mail vs bills or personal letters. Not my business, staff or otherwise as
long as the owning user account is active.
|
cyklone
|
|
response 53 of 118:
|
Feb 21 13:26 UTC 2007 |
It occurs to me that if you really want to limit email accounts to
non-spamming real people, make it opt in AND make then go through bbs to
get instructions for setting up the account. That way sindi could keep her
recycling crew on grex by giving them computers AND instructions on how to
access bbs. Another layer of "discouragement" could be put in place by
requiring a person to actually post something in bbs before staff gives
them some random bit of info needed to set up an email account. Hey, for
people who want to see more participation in bbs, this will at least cause
people to browse it at least once or twice.
|
nharmon
|
|
response 54 of 118:
|
Feb 21 13:43 UTC 2007 |
Has anybody thought about doing a user study on what features the users
find most important? Like, if given the options of e-mail, conferencing,
text-based web browsing, programming libraries, etc. Which would they
choose as the most important? Least important? Maybe it is none of these
and the users have their own suggestions for features that they would
like to see us offer.
We could create a survey program and place a note in the motd asking for
the users' help.
|
slynne
|
|
response 55 of 118:
|
Feb 21 14:30 UTC 2007 |
I like the idea of a survey program but have no idea how to set such a
thing up.
|
cross
|
|
response 56 of 118:
|
Feb 21 14:32 UTC 2007 |
Plop it into newuser?
|
ric
|
|
response 57 of 118:
|
Feb 21 14:33 UTC 2007 |
Interestingly enough, we're discussing the same thing over on M-Net.. user
surveys. I'd actually consider a free account from SurveyMonkey.com to do
the actual surveys..
|
easlern
|
|
response 58 of 118:
|
Feb 21 14:48 UTC 2007 |
Re 44: we've been having visitors from Quebec in party the last couple weeks.
Their computer class teacher sent them here to learn some Unix.
|
keesan
|
|
response 59 of 118:
|
Feb 21 15:52 UTC 2007 |
I do use SDF for important mail. I do use spamd at grex. I don't have
broadband. I am happy for cross that his setup is fast enough that gmail
works for him faster than grex and pine do for me. I have never seen any
webmail (even fastmail.fm) that is anywhere near as fast as mail at a shell
account using pine, mutt, or mailx. You can tell when someone's mail account
was last accessed by typing 'ls -l /var/mail/keesan' for example. Presumably
someone could write a script to identify accounts that have not been accessed
for three months. I don't know if .forward files are publicly readable. How
hard would it be to put back 1MB quotas, at least for non-members?
|
cross
|
|
response 60 of 118:
|
Feb 21 16:25 UTC 2007 |
Regarding #59; No one is going to do it. Just accept that.
|
ric
|
|
response 61 of 118:
|
Feb 21 17:41 UTC 2007 |
re 59 - every time new incoming mail arrives, the mailbox you referred to is
"accessed"
gmail rocks.
So does gmail for domains.
|
mcnally
|
|
response 62 of 118:
|
Feb 21 17:45 UTC 2007 |
re #52: was that meant to be in response to #51? Because if so, I don't
think you understood what I was proposing.
|
cross
|
|
response 63 of 118:
|
Feb 21 18:03 UTC 2007 |
Regarding #62; No, I think #52 was in reponse to the previous posting by
Sindi.
|
glenda
|
|
response 64 of 118:
|
Feb 21 19:28 UTC 2007 |
And why should we be going into accounts to see when email was last accessed
by that account? Again, IT IS NOT OUR BUSINESS to watch whether or not a user
accesses his/her email account. Since we have the reap utility back in place
now, inactive accounts and their attached email accounts should be getting
taken care of on a regular basis. Yes, we have a problem when the reaping
wasn't being done, but that has been fixed now. We just reaped a huge number
of accounts. Let's wait and see if var/mail filling up becomes a problem
again. You may not have an ethical problem with looking into everyone's
account to see when they last accessed email, but I do. I wouldn't look into
your USPS mailbox either. In fact I think that the government frowns upon
that type of behavior. Why should email be treated any differently?
|
keesan
|
|
response 65 of 118:
|
Feb 21 21:49 UTC 2007 |
You DO NOT HAVE TO go into an account, just look at the date on it.
Spamassassin would work a lot faster on 1000 than 6000 accounts.
If you put a 1MB quota on accounts and let people ask for larger ones, would
that be a way to bounce spam from those accounts before using spamassassin
on mail to other accounts?
|
cross
|
|
response 66 of 118:
|
Feb 21 22:07 UTC 2007 |
Regarding #65; What Glenda is asking is what data that would give you that
would be remotely interesting.
|
keesan
|
|
response 67 of 118:
|
Feb 21 22:48 UTC 2007 |
You could figure out the best way to filter spam on accuonts which ARE being
accessed.
|
cross
|
|
response 68 of 118:
|
Feb 21 22:54 UTC 2007 |
You could do that by looking at logs. Please, Sindi, don't try to solve
technical problems that you have absolutely no knowledge of. I know you're
trying to help, but relentlessly pushing a solution to a problem that really
isn't a solution is just irritating.
|
keesan
|
|
response 69 of 118:
|
Feb 21 23:07 UTC 2007 |
So what solution do you propose for reducing the number of unused mail
accounts to the point where spamassassin is practical for the rest?
|
cross
|
|
response 70 of 118:
|
Feb 22 00:20 UTC 2007 |
I've already made my proposals. Go look them up.
|
keesan
|
|
response 71 of 118:
|
Feb 22 00:28 UTC 2007 |
You seem to be proposing that people stop using grex for mail.
|
krokus
|
|
response 72 of 118:
|
Feb 22 00:55 UTC 2007 |
re 69
They aren't looking for an option of that, right now, from what I'm
reading. It also seems like this won't be happening anytime in the
foreseeable future.
You are in an extremly small minority, probably less than 20 people
on the system that are trying to use it as a primary email location.
This just doesn't justify the staff's time, which is very limited to
begin with.
If you really like using pine for email, how about running it on your
computer, tied to Yahoo via SMTP and POP3? (This does require paying
Yahoo a small annual fee.) Yahoo has pretty good spam filtering, and
I get very little into any of my accounts with them.
I just thought of another possible solution, get a copy of UUPC, and
assuming that uucp is still available, bring the mail across onto your
computer. Then you can process it however you see fit.
|
cross
|
|
response 73 of 118:
|
Feb 22 01:07 UTC 2007 |
Regarding #71; Wow. That just now sank in?
Yes. I am proposing that people stop using grex for email unless they really
have no alternative.
|
glenda
|
|
response 74 of 118:
|
Feb 22 03:07 UTC 2007 |
We don't propose anything for reducing the number of unused mail accounts.
Currently mail accounts come with user accounts. If a user doesn't want to
use that account that is ok, it is their business not ours. We just got rid
of some 48,000 accounts. And looking at the dates in a person's account to
see when they last accessed the mail is STILL an invasion of privacy. Why
can't you see that. With 48,000 accounts gone the problem should be mostly
gone. We will be doing reaps of inactive accounts more often. I will do them
myself if need be. Lets see if this eases the problem. Give it a change for
crying out loud. You have been harping on this issue for so long, I think
that it has become a habit. Drop for a while and give things a chance before
asking staff to do something drastic and invasive.
|