|
Grex > Coop6 > #10: Should Grex be running a POP server? | |
|
| Author |
Message |
| 25 new of 131 responses total. |
carson
|
|
response 25 of 131:
| 
Sep 27 05:27 UTC 1994 |
is using POP really more "convenient" than logging in to get mail?
|
rcurl
|
|
response 26 of 131:
|
Sep 27 05:40 UTC 1994 |
It would be like running other clients for other purposes: lynx, to
reach the grex home pages; Fetch, to ftp; an Archie client to search
directories; etc. There are no commands to enter on the host - everything
is done in hypertext or in menus. The result is less "hanging around".
The @sura archie site was going to go entirely to needing clients, and
excluding telnet, because of the greater client "efficiency". How
convenient it is, depends on the quality of the client interface. We've
been urged to use Eurdora on CAEN (UM engineering net), because of the
lower demands on the host for e-mail use.
|
srw
|
|
response 27 of 131:
|
Sep 27 05:45 UTC 1994 |
Yup Indeed. Also the Eudora (and I'd imagine Chameleon, too) mail interface
is a lot more friendly than either elm or pine with your favorite unix editor.
This would be perceived as a big convenience.
|
carson
|
|
response 28 of 131:
|
Sep 27 06:17 UTC 1994 |
This response has been erased.
|
carson
|
|
response 29 of 131:
|
Sep 27 06:19 UTC 1994 |
Rane wrote:
lower demands on the host for e-mail use.
that in itself sounds like a potential benefit for Grex, if that's what a
POP would translate to.
|
chelsea
|
|
response 30 of 131:
|
Sep 27 10:43 UTC 1994 |
Pop should get a trial run and then be evaluated based on
what we find out.
|
steve
|
|
response 31 of 131:
|
Sep 27 12:46 UTC 1994 |
We can John, but I'm not sure how we evaluate it. If there are
logs that it keeps of who did what when, then we can measure it.
|
remmers
|
|
response 32 of 131:
|
Sep 27 17:19 UTC 1994 |
That was Mary, not John.
|
davel
|
|
response 33 of 131:
|
Sep 27 21:41 UTC 1994 |
Um. I'm somewhat against offering services on a trial-run basis without
some kind of consensus about what the criteria for evaluation are. It's
not really fair to people who start to use them if we just cut them off,
and the sense of this unfairness is likely to militate against withdrawing
a service even if it really loads everything down, IMO
|
chelsea
|
|
response 34 of 131:
|
Sep 27 23:32 UTC 1994 |
Okay, so we are now setting criteria that Grex services will need to serve
the whole of the community in order to be worth offering at all. Is this
policy going to be retroactive? Will we be removing those resource
intensive services which mostly only benefit the individual user? Or
maybe only those services which benefit individual users who don't pay
dues? Or maybe we just don't worry much about those who request services
that aren't in demand or is it more like we can't consider those which are
in demand? And isn't it amazing when you remember the whole of Grex was
put online without concern for determining official criteria for
evaluation, feasiblity studies, etc. and blah, blah, blah?
Red flag here. We are becoming what we fled.
|
popcorn
|
|
response 35 of 131:
|
Sep 28 00:25 UTC 1994 |
Re #34:
I disagree. When we founded Grex, we knew there would be heated policy
debates in the coop conference. We knew they wouldn't always be fun,
but it was a necessary part of the way a system with a large number of
owners (the members) would have to be run. The whole point is that
everybody has input into setting policies. That's what's going on
here. When I look at this discussion, I see a group of people batting
around ideas (quite civilly, at that!) about how a new program would
fit into Grex. I don't see what's wrong. I don't know what you're
looking for, Mary.
|
steve
|
|
response 36 of 131:
|
Sep 28 03:37 UTC 1994 |
Mary, will you restate your concerns again? I don't follow
#34.
My feelings are that Grex should help the greatest number of people
that it can, given limited resources. If we can help a lot of
people with a POP server, then we should do it. BUt if we think
that the 'community' of Grex isn't helped, but others already on
the Internet are, then we shouldn't.
|
rcurl
|
|
response 37 of 131:
|
Sep 28 05:27 UTC 1994 |
The "whole community" can be well served by services that very few
users use. An analogy is the fire department, which is a community
service, which everone tries to *avoid* using.
But still: Mary, thank you very much for your strong support of
services that may not be in great demand, such as closed conferences.
|
chelsea
|
|
response 38 of 131:
|
Sep 28 09:45 UTC 1994 |
My feeling about how we are starting to replay what happened
with other systems is not about the quantity or even the tone
of the discussion it's over the multitude of excuses used by
those who already have theirs to protect theirs. There seems
to be a wide range of opinion on whether this program will
be used just a wee bit or a whole lot, whether it will be a
resource drain or maybe end up sparing resources. So, because
we can't prove ahead of time how it may impact the system
we'll just tell those requesting the service that it's not
right for Grex's. I guess I'd rather we try to accommodate
people's requests on a trial basis and only remove the
option if it turns out to cause problems. And yes, I think
our users could handle the disappointment quite well, as even
our youngest users are adults capable of understanding the
concept of a trial period. Sheesh.
There is a difference between the community and the Grex community.
If you make it necessary that any proposed service must enhance
the Grex community then you are going to be saying, up front,
something like we are a club and our activities are going to
be directed toward making it a fun place for our club members to
hang out. And if we take this position we should be consistent.
And again, Valerie, I have no problem with this being discussed.
Mostly, I have a problem with creeping policy, and there is
one working its way though this thread, that something must
benefit the Grex community as a whole to be considered worthy
of being offered.
Rane, this has nothing whatsoever to do with my objections to
closed conferences. This proposed service doesn't even involve
conferencing. And no matter how carefully you reshuffle the
words closed conferences are closed conferences and mail is mail.
|
chelsea
|
|
response 39 of 131:
|
Sep 28 10:13 UTC 1994 |
But I suppose someone could make the suggestion we offer
POP only to 501(c)3 organizations which are registered in
Michigan and fit a political agenda Grex wouldn't find
embarrassing. Go for it.
|
davel
|
|
response 40 of 131:
|
Sep 28 10:22 UTC 1994 |
Mary, how easily do you really think "our users" will accommodate it if
"our users" turn out to include: people who have us as their email address
& suddenly find out they can't get to their mail, and people who suddenly
find that every time they try to access mail they get file-system-full
errors because there are a lot of people with 2MB mail files they can't
get to? There are plenty of services that can be done on a trial basis &
stopped; but once you start advertising an email address to people you're
likely to be really annoyed if it suddenly stops letting you read your
mail; and will every single POP user log in, empty mail box, & set up
.forward file so that it doesn't fill up again?
And I'm with Rane, I'm afraid. Your attitude on closed conferences seems
to me totally inconsistent with the arguments you use on things like this.
Yes, mail is mail, finger is finger, conferences are conferences, and so
on; but others are at least using some kind of recognizable general criteria
for evaluating them all, such as "will it take too many scarce resources?"
You seem to be for or against trying things on the basis of nothing more
than a sense of mission to not do anything M-net does.
|
steve
|
|
response 41 of 131:
|
Sep 28 12:12 UTC 1994 |
Mary, if we have a limited resource like the Internet link,
how else should we manage it?
If I understand you correctly, you are concerned with the fact that
some of us are concerned about providing a service that anyone on
the Internet could use, could drain the link even more, yet not do
anything "for" Grex other than perhaps 1) give the already existing
mail users of Grex a different way of getting mail, 2) make the link
even slower and possibly repel more people because of the even slower
response time.
An alternative that might make you scream Mary, would be to allow POP
service for members only.
I'm going to sounds like a broken record again, but I'd be
willing to see POP service for everyone on the net, if we had
the bandwidth. But since we have 1/53 of a T1 link, I'm afraid we
have to think of the services we offer and their impact on Grex.
|
remmers
|
|
response 42 of 131:
|
Sep 28 13:29 UTC 1994 |
I don't buy that, STeve. This concern that we're supposed to feel over
our small bandwidth didn't stop us from opening up incoming telnet to
the whole net, a move that has had far more impact on our load than a
POP server is likely to have.
If we had had the kind of "what if" debate over the impact of opening
up incoming telnet, that we're having now over a POP server, the
decision might well have been not to open up Grex to the net. If I'd
known beforehand what the impact on Grex's resources would be of
opening up telnet, such as high system load and demands on staff, I
probably would have voted against it myself. But I would have been
wrong. Despite the hassles, I think it's one of the best moves Grex
ever made.
If our concern over the load on the link should really be our governing
priority, then we should be looking for ways to cut down the impact of
incoming telnet, by shutting down newuser, limiting drastically the
number of pty's, etc. But I don't hear anyone suggesting that we
should do that.
Let's not forget that the internet connection was also put up initially
for a trial period of 90 days. That didn't seem to bother anybody.
It worked well. What's wrong with doing the same for POP?
Nobody seems to know for sure what the impact of POP would be, we'll
only find out by trying, my gut feeling (could be wrong) is the impact
wouldn't be huge, so my inclination would be to try it out for a trial
run of a specified length of time.
|
carson
|
|
response 43 of 131:
|
Sep 28 13:37 UTC 1994 |
strangely enough, I don't think a POP server would have a huge impact
on load or bandwidth either. On the other hand, I haven't read much
about what would be involved in setting up such a server, esp. re:
staff time. If it's going to be an obvious timesuck, I wouldn't want
to see it a priority.
|
chelsea
|
|
response 44 of 131:
|
Sep 28 14:22 UTC 1994 |
(Psst... Someone said Mnet is offering POP) <-- clue time
My criteria in evaluating all services is that they be
available to all, equally, minimizing any perks for
special interest groups, paying customers, etc. And
if offering the service to all threatens to overload
our resources then the service isn't right for Grex.
I think I've been very clear and consistent in this area.
Not popular, but consistent (and persistent).
|
carson
|
|
response 45 of 131:
|
Sep 28 14:42 UTC 1994 |
I guess the question is: what resources would installing and
maintaining a POP server affect? I see it affecting staff time,
bandwidth, and diskspace. After that question, then ask: how
would these resources be affected?
|
robh
|
|
response 46 of 131:
|
Sep 28 15:15 UTC 1994 |
Good point, chelsea - if M-Net allows POP, then we shouldn't
allow it under any circumstances. Thanks for the input!
|
steve
|
|
response 47 of 131:
|
Sep 28 15:47 UTC 1994 |
John, opening up incomming telnet was a way to increase the
activity on Grex, to bring in new folks to the conferences, and
to add life into Grex. Its has benefitted the community of Grex
a lot, I'd say.
But when we look at the number of new people around here in
the conferences compared to the number of new people in general,
we find that only a small handful of the 8,400 created since
March 14th have *ever* gotten into a conference. Ever. So if
we continue to think of conferencing as the "heart & soul" of Grex,
we have an interesting problem, in that many many MANY people
haven't the slightest idea of Grex's goals, and are interested
in things like email, talking, IRC, and (amazingly to a lesser
degree these days) news reading.
SO whats a Grex for? I can't say myself, but I know of the range
of activities I'd like to see here. I'd definately like for there
to be a sense of community here.
OPening up incomming telnet was the right thing to do. Its
been interesting, watching to see how people react to Grex, ask
questions, etc. OVerall, I think its gone very well.
But quite possibly in contrast to that, the idea of a POP
service to all is the opposite in terms of a sense of community.
People could (and will, if we set this up) get an account on
Grex, and will never be seen again. But they'll suck up disk
space, and bandwidth, and occupy space on the link that might
be used by someone conferencing, *but they won't be a part of
Grex in any sense*.
With people having to log into Grex for mail as it stands,
they have the slight chance of becomming a part of us. I've
been able to get several people into the conferences because
of write sessions when they had questions. I won't be able to do
that via POP.
IN the end, its a questions of what we want Grex for, really.
|
alb
|
|
response 48 of 131:
|
Sep 28 16:13 UTC 1994 |
Why not just ask M-Net how having a POP server has worked out
for them?
|
remmers
|
|
response 49 of 131:
|
Sep 28 17:10 UTC 1994 |
Re #47: Lots of assumptions about what "might" happen there,
but do you know for sure that's how it'd work out? Nothing
you've said really argues against trying it out for a while,
announcing plainly and publicly that it's only a trial run,
then re-evaluating after a reasonable period has elapsed.
So if there's someone who's willing to do the work of setting
it up and maintaining it, why not?
Re #48: That sounds very sensible.
|