On Sun, 7 Jun 2009 12:05:05 -0500, Aaron T. [email protected]
wrote in
[email protected]:
[re: requiring comp.lang.ruby users to register with ruby-talk list to
have their posts gatewayed to the list. I said it was a small
barrier, Frank Senault said it was a big one …]
[snip]
Simply put, if you don’t find my proposed solution workable, then
please provide a better one which has the impact of stopping the spam
(keeping the headers while nice, won’t accomplish that).
Aaron, how much spam are you seeing? I’m seeing only a handful a day
on my news server. I ask because if newsfeed that is used for the
gateway propagates a lot of spam, maybe we could find a cleaner
newsfeed instead.
From: “Aaron T.” [email protected]
Another option, is do what the svn-users list does- have moderators
approve messages from email addresses which are not subscribed. Yes,
this means if you’re not subscribed, then your post gets delayed.
However, if you have enough mods across different timezones then it’s
usually an acceptable delay. If not, you can always subscribe as I
mentioned above. I’ll take it one step further and offer up to be a
moderator for a minimum of 1 year.
I suspect classification could be automated to the extent
that moderators should be called upon very infrequently.
If we trained a classifier (like maybe
http://classifier.rubyforge.org/ ) with ham and spam from
comp.lang.ruby, I suspect posts that registered a very
high score as ham could be passed along unchallenged by
the gateway. (I’m presuming we aren’t dealing with
spammers who are sophisticated enough in their targeting
of the newsgroup to actually append some pseudo ruby-
related content to their posts?)
So the gateway could continue to pass along high scoring
ham unchallenged, and only route posts to a moderator
queue if they didn’t register as having obvious ruby-
related content.
(I suppose we could get slightly fancier and even auto
whitelist a poster’s email address once a high scoring
ham post had been received from them by the gateway,
after which, being on the whitelist, their posts would
be given the benefit of the doubt.)
Anyway… seems like a problem that could be mostly
automated. I suspect most of the work would be in
implementing the moderator queue mechanism.
Regards,
Bill
On Mon, Jun 8, 2009 at 9:10 AM, Charles C.[email protected] wrote:
Simply put, if you don’t find my proposed solution workable, then
please provide a better one which has the impact of stopping the spam
(keeping the headers while nice, won’t accomplish that).
Aaron, how much spam are you seeing? I’m seeing only a handful a day
on my news server. I ask because if newsfeed that is used for the
gateway propagates a lot of spam, maybe we could find a cleaner
newsfeed instead.
As I said, not enough spam to get me to unsubscribe. Used to be about
1 a day, now it seems to be like 3-5. Am I making a bigger deal out
of this then it probably warrants at this time? Probably. Like many,
I just have a very low threshold for spam nowadays and I’m concerned
that the trend seems to be moving in the wrong direction.
Le Sun, 7 Jun 2009 11:27:02 -0500, James G. a écrit :
So, request received. It’s a little work though and my summer is
pretty insanely busy. Please be patient with me if it takes me a bit
to get to it.
Sure. BTW, I think we could really use a sandbox environment for that
kind
of things. Tell me when you have a bit of time to spare for that
project,
and I’ll try to come with something.
Fred
On Sun, 7 Jun 2009 09:51:08 +0200, “F. Senault” [email protected]
wrote in [email protected]:
Le 7 juin 2009 à 03:48, Charles C. a écrit :
This would take some work, but it is possible.
With the sad state of most usenet servers world wide, a status change
will not be propagated correctly everywhere.
That’s more of a problem in the alt hierarchy. comp.lang.ruby is a
big-8 group, and changes are propagated much better in the big-8. Not
perfectly, mind you, but pretty well. Google is a notable exception.
Le Mon, 08 Jun 2009 12:11:54 -0400, Charles C. a écrit :
That’s more of a problem in the alt hierarchy.
I have some experience from the regional side of things (fr.*), and it’s
pretty bleak. We have a hell of a time to get the local ISPs to take
into
account the changes, even with signed messages using a 10 years old key.
comp.lang.ruby is a
big-8 group, and changes are propagated much better in the big-8. Not
perfectly, mind you, but pretty well. Google is a notable exception.
Mh. Google alone is enough to be very carful in this situation, IMHO.
Fred
On Mon, 8 Jun 2009 11:51:29 -0500, Aaron T. [email protected]
wrote in
[email protected]:
1 a day, now it seems to be like 3-5. Am I making a bigger deal out
of this then it probably warrants at this time? Probably. Like many,
I just have a very low threshold for spam nowadays and I’m concerned
that the trend seems to be moving in the wrong direction.
I’ve noticed an increase in spam from one or a few sources in
technical groups on Usenet recently. After enough complaints, the
perpetrator should be shut down and go away. Then another will
surface.
“Jörg W Mittag” [email protected] schrieb im Newsbeitrag
news:[email protected]…
One of the nice properties of NNTP (and one of the many things that
makes Usenet infinitely more useful than E-Mail for discussion groups)
is that you can delete articles. IOW: as long as you are not the first
person to download that article, you shouldn’t actually see it.
Im pretty sure it is impossible to delete third party articles on the
servers.
It should be even impossible to cancel an own article once somebody has
answered to.
If things would be that easy, solution would be (most probably, in a
non-governmental environment) already there.
Michael Bruschkewitz
Charles C. wrote:
I’ve noticed an increase in spam from one or a few sources in
technical groups on Usenet recently. After enough complaints, the
perpetrator should be shut down and go away. Then another will
surface.
One of the nice properties of NNTP (and one of the many things that
makes Usenet infinitely more useful than E-Mail for discussion groups)
is that you can delete articles. IOW: as long as you are not the first
person to download that article, you shouldn’t actually see it.
However, I must admit that I never do this myself on comp.lang.ruby,
because I am simply not sure what the correct rules are in the US
Usenet. (I’m sort of familiar with the German language Usenet, i.e.
the de.!alt hierarchy, but the rules are very different there.)
jwm
On Fri, 12 Jun 2009 09:11:40 +0200, “Michael Bruschkewitz”
[email protected] wrote in
[email protected]:
“Jörg W Mittag” [email protected] schrieb im Newsbeitrag
news:[email protected]…
One of the nice properties of NNTP (and one of the many things that
makes Usenet infinitely more useful than E-Mail for discussion groups)
is that you can delete articles. IOW: as long as you are not the first
person to download that article, you shouldn’t actually see it.
Im pretty sure it is impossible to delete third party articles on the
servers.
You’re correct. If you delete a post in your news client, it is
simply deleted from the local cache of headers. It still exists on
the server and others will still see it.
It should be even impossible to cancel an own article once somebody has
answered to.
This is not the case. Any post can be canceled at any time, provided
that the server in question honors cancellations. Most do not, due to
past abuse.
“Charles C.” [email protected] schrieb im Newsbeitrag
news:[email protected]…
On Fri, 12 Jun 2009 09:11:40 +0200, “Michael Bruschkewitz”
[email protected] wrote in
It should be even impossible to cancel an own article once somebody has
answered to.
This is not the case. Any post can be canceled at any time, provided
that the server in question honors cancellations. Most do not, due to
past abuse.
Thanks for clarification.