Update from the NANOG Communications Committee regarding recent off-topic posts
As a quick update, we've implemented some list settings last week to help to keep spam off the list. New subscribers are moderated until we're comfortable with their posts. We rejected the idea of keyword based message filtering since not only is a lot of work to maintain, it's trivial to get around it if you really want to post banned words. Comments and suggestions are welcome. Matt Griswold, on behalf of the NANOG Communications Committee
As a quick update, we've implemented some list settings last week to help to keep spam off the list. New subscribers are moderated until we're comfortable with their posts. We rejected the idea of keyword based message filtering since not only is a lot of work to maintain, it's trivial to get around it if you really want to post banned words.
Comments and suggestions are welcome.
Matt Griswold, on behalf of the NANOG Communications Committee
I dislike this change - how long are subscribers considered "new"? I believe (and I hope I'm wrong) that with this new rule the nanog maiing list will turn into another fulldisc (list activity greatly reduced) by this change. Before this change I had thought of nanog as the new fulldisc - I guess I will have to find yet ANOTHER mailing list for continued activity. And just for reference - have you guys SEEN the "Linux Kernel Mailing List"? - it gets frequent spam posts and yet is perfectly able to ignore the spam/irrelevant posts and continue on its remit.
On Mon, 30 Jul 2012 21:04:36 +0200, Panashe Flack said:
list for continued activity. And just for reference - have you guys SEEN the "Linux Kernel Mailing List"? - it gets frequent spam posts and yet is perfectly able to ignore the spam/irrelevant posts and continue on its remit.
For those who don't drink from the Linux-Kernel firehose, it averages 1 or 2 spams per day - and anywhere from 500 to 700 postings a day. As Linus Torvalds said, back when it was averaging 200 a day: "Note that nobody reads every post in linux-kernel. In fact, nobody who expects to have time left over to actually do any real kernel work will read even half. Except Alan Cox, but he's actually not human, but about a thousand gnomes working in under-ground caves in Swansea. None of the individual gnomes read all the postings either, they just work together really well." The list managers do an incredible job of stopping spam - but even if 50 or 75 a day got through, they'd just be lost in the noise. You're skipping several hundred messages a day, skipping a few more isn't any different.
I'm sorry Panashe is upset by this rule. Interestingly, "Your search - Panashe Flack nanog - did not match any documents." So my guess is that a post from that account has not happened before, meaning the post was moderated yet still made it through. Has anyone done a data mining experiment to see how many posts a month are from "new" members? My guess is it is a trivial percentage. -- TTFN, patrick On Jul 30, 2012, at 13:35 , valdis.kletnieks@vt.edu wrote:
On Mon, 30 Jul 2012 21:04:36 +0200, Panashe Flack said:
list for continued activity. And just for reference - have you guys SEEN the "Linux Kernel Mailing List"? - it gets frequent spam posts and yet is perfectly able to ignore the spam/irrelevant posts and continue on its remit.
For those who don't drink from the Linux-Kernel firehose, it averages 1 or 2 spams per day - and anywhere from 500 to 700 postings a day.
As Linus Torvalds said, back when it was averaging 200 a day:
"Note that nobody reads every post in linux-kernel. In fact, nobody who expects to have time left over to actually do any real kernel work will read even half. Except Alan Cox, but he's actually not human, but about a thousand gnomes working in under-ground caves in Swansea. None of the individual gnomes read all the postings either, they just work together really well."
The list managers do an incredible job of stopping spam - but even if 50 or 75 a day got through, they'd just be lost in the noise. You're skipping several hundred messages a day, skipping a few more isn't any different.
The fix for this issue is trivial. Every new signup should require a sponsor or a deposit of funds into a new member fund. Once a member has made a relevant post regarding a NANOG related item their funds are returned. If someone spams they forfeit the money and it is used to help defray the costs of attending NANOG for the 99%. If the poster has been sponsored by a current member, said member is flogged in public at the next meeting. ...runs Sent from my iPhone On Jul 30, 2012, at 10:42 AM, "Patrick W. Gilmore" <patrick@ianai.net> wrote:
I'm sorry Panashe is upset by this rule. Interestingly, "Your search - Panashe Flack nanog - did not match any documents." So my guess is that a post from that account has not happened before, meaning the post was moderated yet still made it through.
Has anyone done a data mining experiment to see how many posts a month are from "new" members? My guess is it is a trivial percentage.
-- TTFN, patrick
On Jul 30, 2012, at 13:35 , valdis.kletnieks@vt.edu wrote:
On Mon, 30 Jul 2012 21:04:36 +0200, Panashe Flack said:
list for continued activity. And just for reference - have you guys SEEN the "Linux Kernel Mailing List"? - it gets frequent spam posts and yet is perfectly able to ignore the spam/irrelevant posts and continue on its remit.
For those who don't drink from the Linux-Kernel firehose, it averages 1 or 2 spams per day - and anywhere from 500 to 700 postings a day.
As Linus Torvalds said, back when it was averaging 200 a day:
"Note that nobody reads every post in linux-kernel. In fact, nobody who expects to have time left over to actually do any real kernel work will read even half. Except Alan Cox, but he's actually not human, but about a thousand gnomes working in under-ground caves in Swansea. None of the individual gnomes read all the postings either, they just work together really well."
The list managers do an incredible job of stopping spam - but even if 50 or 75 a day got through, they'd just be lost in the noise. You're skipping several hundred messages a day, skipping a few more isn't any different.
On 7/30/12 10:57 AM, Steven Noble wrote:
The fix for this issue is trivial. Every new signup should require a sponsor or a deposit of funds into a new member fund. Once a member has made a relevant post regarding a NANOG related item their funds are returned.
If someone spams they forfeit the money and it is used to help defray the costs of attending NANOG for the 99%.
If the poster has been sponsored by a current member, said member is flogged in public at the next meeting. Most of the subscribers to the mailing list never post.
...runs
Sent from my iPhone
On Jul 30, 2012, at 10:42 AM, "Patrick W. Gilmore" <patrick@ianai.net> wrote:
I'm sorry Panashe is upset by this rule. Interestingly, "Your search - Panashe Flack nanog - did not match any documents." So my guess is that a post from that account has not happened before, meaning the post was moderated yet still made it through.
Has anyone done a data mining experiment to see how many posts a month are from "new" members? My guess is it is a trivial percentage.
-- TTFN, patrick
On Jul 30, 2012, at 13:35 , valdis.kletnieks@vt.edu wrote:
On Mon, 30 Jul 2012 21:04:36 +0200, Panashe Flack said:
list for continued activity. And just for reference - have you guys SEEN the "Linux Kernel Mailing List"? - it gets frequent spam posts and yet is perfectly able to ignore the spam/irrelevant posts and continue on its remit. For those who don't drink from the Linux-Kernel firehose, it averages 1 or 2 spams per day - and anywhere from 500 to 700 postings a day.
As Linus Torvalds said, back when it was averaging 200 a day:
"Note that nobody reads every post in linux-kernel. In fact, nobody who expects to have time left over to actually do any real kernel work will read even half. Except Alan Cox, but he's actually not human, but about a thousand gnomes working in under-ground caves in Swansea. None of the individual gnomes read all the postings either, they just work together really well."
The list managers do an incredible job of stopping spam - but even if 50 or 75 a day got through, they'd just be lost in the noise. You're skipping several hundred messages a day, skipping a few more isn't any different.
On Jul 30, 2012, at 15:04, joel jaeggli <joelja@bogus.com> wrote:
On 7/30/12 10:57 AM, Steven Noble wrote:
The fix for this issue is trivial. Every new signup ...
Most of the subscribers to the mailing list never post.
+1 (from an inveterate but VERY appreciative lurker) ..Allen
On 07/30/2012 09:23 PM, Allen McKinley Kitchen (gmail) wrote:
On Jul 30, 2012, at 15:04, joel jaeggli <joelja@bogus.com> wrote:
On 7/30/12 10:57 AM, Steven Noble wrote:
The fix for this issue is trivial. Every new signup ... Most of the subscribers to the mailing list never post.
+1 (from an inveterate but VERY appreciative lurker)
..Allen
I run a tiny network, no AS number. I try to build interesting features into my local hackerspace's network from what I find here. I don't post because I don't have useful experience to the size/scale of what is posted here. I don't know what your organization is really nor where you meet or who any of you are. But even in my small network, I have picked up 10x more operational knowledge here than what I learned from courses and classes, which always seem to push you to use X just because it exists or because its from a specific company. I guess I mean to say thanks, for the knowledge and the moderation. If most are like me, this will make it nicer to read. (except those people whos email client breaks Thunderbird's threading system. no kudos for them.)
On 07/30/2012 10:57 AM, Steven Noble wrote:
The fix for this issue is trivial. Every new signup should require a sponsor or a deposit of funds into a new member fund. Once a member has made a relevant post regarding a NANOG related item their funds are returned.
If someone spams they forfeit the money and it is used to help defray the costs of attending NANOG for the 99%.
If the poster has been sponsored by a current member, said member is flogged in public at the next meeting.
...runs
Sent from my iPhone
On Jul 30, 2012, at 10:42 AM, "Patrick W. Gilmore"<patrick@ianai.net> wrote:
I'm sorry Panashe is upset by this rule. Interestingly, "Your search - Panashe Flack nanog - did not match any documents." So my guess is that a post from that account has not happened before, meaning the post was moderated yet still made it through.
Has anyone done a data mining experiment to see how many posts a month are from "new" members? My guess is it is a trivial percentage.
-- TTFN, patrick
On Jul 30, 2012, at 13:35 , valdis.kletnieks@vt.edu wrote:
On Mon, 30 Jul 2012 21:04:36 +0200, Panashe Flack said:
list for continued activity. And just for reference - have you guys SEEN the "Linux Kernel Mailing List"? - it gets frequent spam posts and yet is perfectly able to ignore the spam/irrelevant posts and continue on its remit. For those who don't drink from the Linux-Kernel firehose, it averages 1 or 2 spams per day - and anywhere from 500 to 700 postings a day.
As Linus Torvalds said, back when it was averaging 200 a day:
"Note that nobody reads every post in linux-kernel. In fact, nobody who expects to have time left over to actually do any real kernel work will read even half. Except Alan Cox, but he's actually not human, but about a thousand gnomes working in under-ground caves in Swansea. None of the individual gnomes read all the postings either, they just work together really well."
The list managers do an incredible job of stopping spam - but even if 50 or 75 a day got through, they'd just be lost in the noise. You're skipping several hundred messages a day, skipping a few more isn't any different.
Would be an iPhone user to suggest such an idea. Thanks for not implementing this so us peons can learn a thing or two, too.
-- () ascii ribbon campaign - against html e-mail /\ www.asciiribbon.org - against proprietary attachments
On 7/30/2012 1:42 PM, Patrick W. Gilmore wrote:
I'm sorry Panashe is upset by this rule. Interestingly, "Your search - Panashe Flack nanog - did not match any documents." So my guess is that a post from that account has not happened before, meaning the post was moderated yet still made it through.
Has anyone done a data mining experiment to see how many posts a month are from "new" members? My guess is it is a trivial percentage.
Ignoring many harder to determine things like "who has changed their email address" and reducing it to simple shell commands, I got this: for i in `cat ../nanog_archive_index.html | grep txt | cut -f2 -d\"` ; do wget http://mailman.nanog.org/pipermail/nanog/$i; done du -sh=41M (uncompressed=100M). That seems small for all the mail since random 2007 but I'd rather use an official archive so people can duplicate results and refine things. grep -h "^From: " * | sort | uniq -c | sort -nr First of all I will say Owen is winning by a fair margin: 1562 From: owen at delong.com (Owen DeLong) 929 From: randy at psg.com (Randy Bush) 775 From: Valdis.Kletnieks at vt.edu (Valdis.Kletnieks at vt.edu) 688 From: morrowc.lists at gmail.com (Christopher Morrow) 621 From: jbates at brightok.net (Jack Bates) 558 From: jra at baylink.com (Jay Ashworth) 480 From: gbonser at seven.com (George Bonser) 450 From: patrick at ianai.net (Patrick W. Gilmore) 446 From: cidr-report at potaroo.net (cidr-report at potaroo.net) Total count: grep -h "^From: " * | wc -l 54166 # Totals for < 10 contributors for i in 1 2 3 4 5 6 7 8 9; do grep -h "^From: " * | sort | uniq -c | sort -nr | grep " $i" | wc -l; done 3129 1111 552 319 208 157 131 103 94 Total for less than 10 posts contributors: 5804 Percentages: 5804/54166=1% of posts from low contributors. # shows the number of people who've contributed that number of times. grep -h "^From: " * | sort | uniq -c | sort -nr | awk '{print $1}' | uniq -c | sort -nr # another interesting thing to look at is posts by month per user (dropping the -h from grep): grep "^From: " * | sort | uniq -c | sort -nr # not the most efficient, but tells you who posted the most in a month: for i in *; do grep "^From: " * | sort | uniq -c | sort -nr | grep $i | head -n 1; done # Per month, how many single post contributions happen/total. The numbers can be higher here since people who posted in a different month may still be counted as a new contributor for i in *; do echo -n "$i "; grep "^From: " $i | sort | uniq -c | sort -nr | grep " 1 " | wc -l | tr '\n' '/'; grep "^From: " $i | wc -l ; done
What's an order of magnitude between friends? Very occasionally yours, -- Jamie Bowden (jamie@photon.com) Sr. Sys. Admin. (703) 243-6613 x3848 Photon Research Associates, Inc. 1616 Fort Myer Drive, Suite 1000 Arlington, VA 22209
-----Original Message----- From: valdis.kletnieks@vt.edu [mailto:valdis.kletnieks@vt.edu] Sent: Thursday, August 02, 2012 4:56 PM To: Robert Drake Cc: nanog@nanog.org Subject: Re: Update from the NANOG Communications Committee regarding recent off-topic posts
On Thu, 02 Aug 2012 16:25:56 -0400, Robert Drake said:
Percentages: 5804/54166=1% of posts from low contributors.
I suspect you fat-fingered something - I get 10.7%, not 1%, for that calculation...
Friends don't let friends binary shift. On Thu, Aug 2, 2012 at 2:47 PM, Jamie Bowden <jamie@photon.com> wrote:
What's an order of magnitude between friends?
Very occasionally yours,
-- Jamie Bowden (jamie@photon.com) Sr. Sys. Admin. (703) 243-6613 x3848 Photon Research Associates, Inc. 1616 Fort Myer Drive, Suite 1000 Arlington, VA 22209
-----Original Message----- From: valdis.kletnieks@vt.edu [mailto:valdis.kletnieks@vt.edu] Sent: Thursday, August 02, 2012 4:56 PM To: Robert Drake Cc: nanog@nanog.org Subject: Re: Update from the NANOG Communications Committee regarding recent off-topic posts
On Thu, 02 Aug 2012 16:25:56 -0400, Robert Drake said:
Percentages: 5804/54166=1% of posts from low contributors.
I suspect you fat-fingered something - I get 10.7%, not 1%, for that calculation...
-- -george william herbert george.herbert@gmail.com
On 7/30/2012 12:04 PM, Panashe Flack wrote:
As a quick update, we've implemented some list settings last week to help to keep spam off the list. New subscribers are moderated until we're comfortable with their posts...
I dislike this change - how long are subscribers considered "new"?
I applaud this change. If I still traveled, I'd show up to the next NANOG, and buy the committee a beer. Instead, I send them my thanks. I run a couple of mailing lists, and every once in a while, someone will subscribe and set off my cynicism meter. I hit the moderate button on the new account, and sad to say, I've only been wrong to do so once, out of the last ten times I did it. Thanks again. -- "My name is Ozymandias, king of kings: Look on my works, ye Mighty, and despair!" Nothing beside remains. Round the decay Of that colossal wreck, boundless and bare The lone and level sands stretch far away.
----- Original Message -----
From: "Randy Bush" <randy@psg.com>
thanks MLC or whatever it calls itself this week
C'mon, Randy; It's been called that since it kicked me off 7 years ago. :-) Cheers, -- jra -- Jay R. Ashworth Baylink jra@baylink.com Designer The Things I Think RFC 2100 Ashworth & Associates http://baylink.pitas.com 2000 Land Rover DII St Petersburg FL USA +1 727 647 1274
On Jul 30, 2012, at 16:35 , Jay Ashworth <jra@baylink.com> wrote:
thanks MLC or whatever it calls itself this week
C'mon, Randy; It's been called that since it kicked me off 7 years ago. :-)
Except, of course, it has been called the Communications Committee for a while now. (The change was made because the committee took responsibility for more than just the mailing list.) But 1 change in 7 years made years ago does not, IMHO, merit a "whatever it calls itself this week" snark. -- TTFN, patrick
----- Original Message -----
From: "Patrick W. Gilmore" <patrick@ianai.net>
Except, of course, it has been called the Communications Committee for a while now. (The change was made because the committee took responsibility for more than just the mailing list.)
My turn for "silly me".
But 1 change in 7 years made years ago does not, IMHO, merit a "whatever it calls itself this week" snark.
No... not, it doesn't. Maybe it's been less time in Japan? Cheers, -- jra -- Jay R. Ashworth Baylink jra@baylink.com Designer The Things I Think RFC 2100 Ashworth & Associates http://baylink.pitas.com 2000 Land Rover DII St Petersburg FL USA #natog +1 727 647 1274
participants (15)
-
Allen McKinley Kitchen (gmail) -
Etaoin Shrdlu -
George Herbert -
Jamie Bowden -
Jay Ashworth -
joel jaeggli -
Matt Griswold -
Panashe Flack -
Patrick W. Gilmore -
Randy Bush -
Rhys Rhaven -
Robert Drake -
Scott Noel-Hemming -
Steven Noble -
valdis.kletnieks＠vt.edu