> "claw" == <[EMAIL PROTECTED]> writes:
claw> Ultimately I'd like to see separate moderate queues for the
claw> different hold reasons (at least at the UI level). Thus,
claw> for instance I'd like to be able to view just the messages
claw> held for moderator approval, just
On Mon, 25 Jun 2001 23:46:02 -0700
Chuq Von Rospach <[EMAIL PROTECTED]> wrote:
> On Monday, June 25, 2001, at 10:43 PM, J C Lawrence wrote:
>> I have a long standing rule on how I reward such people: I
>> unsubscribe them, immediately if not sooner.
> I'm not that hard-@ss yet...
Hehn. And w
On Monday, June 25, 2001, at 10:43 PM, J C Lawrence wrote:
>
> I have a long standing rule on how I reward such people: I
> unsubscribe them, immediately if not sooner.
>
I'm not that hard-@ss yet, but it's one reason why I have (and enforce)
a "don't try to be a list mom" rule. My rules are
On Mon, 25 Jun 2001 20:57:38 -0700
Chuq Von Rospach <[EMAIL PROTECTED]> wrote:
> (we could, I guess, get into my position that e-mail is a horrible
> way to do this sort of stuff... After all, on a typical discussion
> list, what's usually the first thing that happens after a bunch of
> people g
On Mon, 25 Jun 2001 23:39:32 -0400
bob > wrote:
> Hmm, can't say I've seen the long turnaround, but you could still
> play with the numbers, based on the fact that the mailbot will
> generate one reply for every message it receives. In other words,
> you could make a rule something like:
>
On Mon, 25 Jun 2001 23:21:54 -0400
bob > wrote:
>> If you can find something the mailbot did that (a) would allow
>> mailman to recognize it as such, and (b) wouldn't cause false
>> positives by blocking legitimate messages, let us know. We'll
>> look at improving mailman to trap it.
> Relativ
On Mon, 25 Jun 2001 21:21:20 -0700
Chuq Von Rospach <[EMAIL PROTECTED]> wrote:
> My problem is the false positive problem. You can tune it to catch
> 90% of the mail loops, but mail loops are generally pretty
> infrequent anyway. But can you tune it to do that AND not cause
> false positives wit
On Monday, June 25, 2001, at 09:41 PM, Bob Puff@NLE wrote:
> I agree that false positives are bad. But let's take a look at the
> past few messages people have posted,
true -- but it's no excuse to fix it badly. Or simply create other
problems.
> I doubt even the most prolific poster is go
On Monday, June 25, 2001, at 09:19 PM, Barry A. Warsaw wrote:
> Nope, and I agree with everything you've said. What I was thinking of
> was flagging situations where you see 10 or 50 or 100 posts from the
> same address in the span of a couple of minutes.
But you won't see that, Barry. Think a
I agree that false positives are bad. But let's take a look at the past few messages
people have posted, saying, "I've got a massive autoresponder problem. How do I stop
it?!" We're talking about hundreds, maybe thousands of posts per day. I doubt even
the most prolific poster is going to b
> "CVR" == Chuq Von Rospach <[EMAIL PROTECTED]> writes:
CVR> Could the new queueing system be set up with a timed-backoff
CVR> delay? you'd have to keep a fair amount of state, but the
CVR> minimum time between postings for a given user is, say, 30
CVR> minutes, and every ti
On Monday, June 25, 2001, at 09:13 PM, Bob Puff@NLE wrote:
> That was one wild autoresponder!
yeah. Made ME much more aware of why these things need to behave
properly. I try, although sometimes I fall short with my stuff.
> But I still maintain that having the two variables I mentioned, x
>
Hi Chuq,
That was one wild autoresponder!
But I still maintain that having the two variables I mentioned, x number of messages
in y minutes, this could be "tuned" per list to be 90% effective. On a high-volume
list (like this one), obviously you set the max number of messages higher than on a
On Monday, June 25, 2001, at 08:47 PM, Barry A. Warsaw wrote:
> One approachable suggestion: put a governor on the number of messages
> any one address is allowed to post to a list per time period.
>
Could the new queueing system be set up with a timed-backoff delay?
you'd have to keep a fair
On Monday, June 25, 2001, at 08:39 PM, Bob Puff@NLE wrote:
> Hmm, can't say I've seen the long turnaround, but you could still play
> with the numbers, based on the fact that the mailbot will generate one
> reply for every message it receives.
My worst case (and I swear this is true). I found
> "CVR" == Chuq Von Rospach <[EMAIL PROTECTED]> writes:
>> It would be nice if Mailman was more resistant to these mail
>> loops [2].
CVR> Suggestions on how?
One approachable suggestion: put a governor on the number of messages
any one address is allowed to post to a list per
> "CVR" == Chuq Von Rospach <[EMAIL PROTECTED]> writes:
CVR> One reason why I never use a vacation message.
Same here. Plus the fact that I realized such a message really makes
no difference whatsoever. I'm still as unresponsive as ever to the
vast avalanche of email that continues un
Chuq Von Rospach wrote:
>
> On Monday, June 25, 2001, at 08:21 PM, Bob Puff@NLE wrote:
>
> > Relatively easy. Set a limit on how many messages a user can post in
> > say a 10 minute period. Make it user editable. If you get >8, you've
> > got a loop!
>
> that slows it down, but doesn't catch
On Monday, June 25, 2001, at 08:21 PM, Bob Puff@NLE wrote:
> Relatively easy. Set a limit on how many messages a user can post in
> say a 10 minute period. Make it user editable. If you get >8, you've
> got a loop!
that slows it down, but doesn't catch any but the worst loops. It's just
a
> If you can find something the mailbot did that (a) would allow mailman
> to recognize it as such, and (b) wouldn't cause false positives by
> blocking legitimate messages, let us know. We'll look at improving
> mailman to trap it.
Relatively easy. Set a limit on how many messages a user ca
On Monday, June 25, 2001, at 05:19 PM, John W Baxter wrote:
> I often wonder why people who would *not* hang
>"Owner away--burgle here"
> signs on their houses on these trips send out "burgle here"
One reason why I never use a vacation message. The other is, for better
or worse, that I'm n
I often wonder why people who would *not* hang
"Owner away--burgle here"
signs on their houses on these trips send out "burgle here" invitations in
email.
Ah, well. I guess it's a sliding scale of some sort...I do hear "we're
away for the weekend" on more answering machines than I should (at
Chuq Von Rospach wrote:
> > Filter that.
>
> I have this mental image, which includes some very motivated finger
> gesturing. But that's probably not what you meant...
::snickering uncontrollably and wondering how she got caught::
Maaayybe. Maybe not. I certainly had plenty of other rude thing
On Monday, June 25, 2001, at 04:22 PM, Amanda wrote:
> I now
> have a filter that reads something like this: "If this text appears in
> the
> subject or body of the message, send it to the trash mailbox: out of the
> office; [..etc...]
And once you've done that, add in german, french, russia
Chuq Von Rospach wrote:
> Did the guy who set up the mailbot give mailman any way of telling this
> was a mailbot, or a loop? Did they include any keywords, like "out of
> the office" or "vacation" or "I am an idiot who's mailbot is going to
> make you crazy now"?
> [snip]
> If you can find somet
On Monday, June 25, 2001, at 11:02 AM, Gunnar Evermann wrote:
> Some guy send a request to list-request and left for holidays
> activating some stupid vacation feature in his Mailer (Outlook). When
> Mailman replied to the request message Outlook sent back not one but
> two vacation messages [1]
I am running Mailman 2.0 and today had a slight problem with the
installation
AFAICT what happened is:
Some guy send a request to list-request and left for holidays
activating some stupid vacation feature in his Mailer (Outlook). When
Mailman replied to the request message Outlook sent back not
27 matches
Mail list logo
