Discuss

discuss@lists.surbl.org

2 participants
1876 discussions

Re: [SURBL-Discuss] Revised DMOZ data, got Wikipedia domains too
by Frank Ellermann 09 Oct '04

09 Oct '04

Jeff Chan wrote: > Would there be many FNs (missed spams) if we whitelisted > all of these? In other words are these all truly False > Positives? > If not, which ones do you feel are true spammers and why. If WS and SC agree on something (or even more of the SURBLs) they might have a point. I tried to check kit.net (found in your list), and got an "unknown host" error. So maybe that was a FN (= spammer). I also tried to check made.com (on your original DMOZ list, it was listed in SC and AB IIRC), and came to the conclusion that this might be a Korean hoster. The subdomain listed in DMOZ looked pretty ordinary (at least the non-Flash part), no offers to buy something. But that could be misleading, I didn't visit the site with enabled JS. And of course the reason for the SC listing could be another subdomain, not the one listed by DMOZ. But made.com could be a FP (?) Hard to judge. More than one bit in multi is a hint. Bye.

1 0

Revised DMOZ data, got Wikipedia domains too
by Jeff Chan 09 Oct '04

09 Oct '04

Following up on the earlier check of DMOZ domains against SURBL data, I applied some of Quinlan's suggestions and grabbed different revisions of the DMOZ data and joined (intersected) them in order to eliminate changes due to things like editor- removed spammer/abuser domains, etc. This also means that new additions are ignored, but the corpus is so large that the benefits of capturing editor removals is probably more important. Snapshots are the three most recent available from 9/9/04, 9/25/04 and 10/7/04 (file dates) are intersected, resulting in fewer records (only those that are constant across all three snapshots): http://rdf.dmoz.org/rdf/archive/ Here are the line, word and character counts: 2300851 2300851 38065969 dmoz.srt 1169 11690 123909 dmoz-blocklist.summed.txt 1141 11410 120860 dmoz-blocklist.ws 1169 1169 17977 dmoz-blocklist.txt 7394 7394 97011 dmoz-whitelist.txt The above are revised versions of joins on blocklists, with list info, with ws hits, with just domains, and against the whitelist. These are in the whitelists directory, though we are still *not* applying the dmoz domains as whitelists: http://spamcheck.freeapp.net/whitelists/ The previous dmoz (9/25 version only IIRC) and hits are archived as: 2326173 2326173 38494184 dmoz.srt1 1338 13380 141946 dmoz-blocklist1.summed.txt 1173 11730 124298 dmoz-blocklist1.ws 1338 1338 20533 dmoz-blocklist1.txt 7375 7375 96720 dmoz-whitelist1.txt __ I was also able to grab 4 snapshots of the wikipedia, all sections (all languages). Only the two most recent snapshots had comparable numbers of sections so I used only those two. Using only two sets may be ok, since these are much smaller corpora and there's probably more hand-editing of them. http://download.wikimedia.org/ Where dmoz has about 2.3 million domains, wikipedia has about 174k domains: 173828 173828 2633441 wikipedia.srt 188 1880 19631 wikipedia-blocklist.summed.txt 188 188 2713 wikipedia-blocklist.txt 2437 2437 29581 wikipedia-whitelist.txt __ I also took the intersection of the three dmoz snapshots and the two wikipedia snapshots to get a smaller list containing only the ~102k domains found in both wikipedia and dmoz: 101619 101619 1498653 wikipedia-dmoz.srt 116 1160 11928 wikipedia-dmoz-blocklist.summed.txt 116 116 1591 wikipedia-dmoz-blocklist.txt 2223 2223 26854 wikipedia-dmoz-whitelist.txt This intersection of dmoz and wikipedia domains probably represents the best hope for large whitelist additions so far. This is probably imperfect data, but at least it has some checks by human editors and techniques applied to reduce spammer domains, including comparing the snapshots over time and intersecting the two relatively unrelated sources. Can anyone think of any other hand-edited databases, directories, encyclopedias, etc of URIs of hopefully legitimate (non-spammer) domains that are publically available? Please think about it a little, and speak up! While 102k domains isn't nearly as large as the 2.3M in dmoz, it's certainly more than the 12k or so whitelist records we currently have. How does the intersected list look as a potential whitelist? http://spamcheck.freeapp.net/whitelists/wikipedia-dmoz.srt Please also take a look at these blocklist hits (potential FPs) and share what you think: http://spamcheck.freeapp.net/whitelists/wikipedia-dmoz-blocklist.summed.txt Would there be many FNs (missed spams) if we whitelisted all of these? In other words are these all truly False Positives? If not, which ones do you feel are true spammers and why. Jeff C. -- "If it appears in hams, then don't list it."

4 9

RE: Revised DMOZ data, got Wikipedia domains too
by Chris Santerre 09 Oct '04

09 Oct '04

First, great work Jeff. > >While 102k domains isn't nearly as large as the 2.3M in dmoz, >it's certainly more than the 12k or so whitelist records we >currently have. How does the intersected list look as a >potential whitelist? > > http://spamcheck.freeapp.net/whitelists/wikipedia-dmoz.srt > I think this is just plain nuts to whitelist all of these! Why? If we don't try to whitelist the most popular sites, then what the heck it the point? We could whitelist millions of legit domains forever. The popular ones are the most important. Here is one from the above list. Why would listing this help us? http://oigawa-railway.co.jp/ (looks like a real popular site huh!) >Please also take a look at these blocklist hits (potential FPs) >and share what you think: > > http://spamcheck.freeapp.net/whitelists/wikipedia-dmoz-blocklist.summed.txt I picked of few of these that may give us problems, and none of them met our current criteria to list. (sissy-world.com, good grief that had to be a man at one time!) With the ability to now see whitelisted domains in the crossref page, I don't see a problem with whitelisting all these on the list. Because if they do start spamming again, we can see they are whitelisted and remove them. so: -1 for adding all those intersected to WL +1 for whitelisting the blacklist hits. --Chris

3 4

servepath is pissing me off
by Chris Santerre 09 Oct '04

09 Oct '04

they host these idiots and a bunch more! theviralspiral.com I think servepath.com sould be listed. They host obvious spam houses. And they don't host many domains, so to say they didn't know this was a spammer is a complete joke. Chris Santerre System Admin and SARE Ninja http://www.rulesemporium.com http://www.surbl.org 'It is not the strongest of the species that survives, not the most intelligent, but the one most responsive to change.' Charles Darwin

3 2

RE: [SURBL-Discuss] RE: Revised DMOZ data, got Wikipedia domains too
by Chris Santerre 08 Oct '04

08 Oct '04

>-----Original Message----- >From: Bill Landry [mailto:billl@pointshare.com] >Sent: Friday, October 08, 2004 3:06 PM >To: SURBL Discussion list >Subject: Re: [SURBL-Discuss] RE: Revised DMOZ data, got Wikipedia >domains too > > >----- Original Message ----- >From: "Chris Santerre" <csanterre(a)merchantsoverseas.com> > >> >> I think this is just plain nuts to whitelist all of >these! Why? If we >> >> don't try to whitelist the most popular sites, then what >the heck it >> >> the point? We could whitelist millions of legit domains >forever. The >> >> popular ones are the most important. >> > >> >The points: >> > >> > - whitelisting legitimate domains limits the >effectiveness of joe job >> > attacks that result in FPs in various SURBL blacklists >> > - whitelisting could be used as negative points for MAIL FROM if >> > combined with SPF (and more domains is better) >> >> Yeah, but not everyone is using SPF yet. But if they were, sure! >> > >> >In addition: >> > >> > - I would only whitelist those domains (a) subject to editorial >> > removal (b) so long as their domain registration is old >enough and >> > (c) so long as they pass other criteria such as no SBL >listing for >> > NS->A. >> >> Yeah date seems to be key in more and more cases. Granted a >spammer could >> buy one of these older ones, but hasn't happened often enough. >> >> > - I would maintain the automated whitelist separately >from the human >> > edited whitelist and handle it differently. For >example, perhaps >> > automated whitelist entries can only remove a single >blacklist hit >> > (like SpamCop), but to remove two independent blacklist hits, it >> > requires a human decision. >> > >> >> Did you look at the example from the list I gave? It doesn't >even have a >web >> page! Just says testing. I'm all for whitelisting, but popular/useful >> domains only. >> >> >> so: >> >> -1 for adding all those intersected to WL >> >> +1 for whitelisting the blacklist hits. >> > >> >I think there are other options available due to the miracle of >> >programming. ;-) >> >> Well like they say around here, "You can't argue with success." >> >> But taking away spam points based on an autowhitelist still makes me >> nervous. But you might have a few tricks up your sleeve D.Q. >that I don't >> know about yet ;) > >If the whitelist is setup as a SURWL, for example, using it or >not would be >an option to each individual mail administrator. And the >negative weight >applied to reduce the score would be totally dependant on the >discretion of >the SURWL user. > True, and having the option is good. Having the work overhead to do that on top of doing the blacklists is not. And the logic stands if it ain't blacklisted, it must be good ;) --Chris

2 1

RE: [SURBL-Discuss] Revised DMOZ data, got Wikipedia domains too
by Chris Santerre 08 Oct '04

08 Oct '04

>-----Original Message----- >From: Bill Landry [mailto:billl@pointshare.com] >Sent: Friday, October 08, 2004 3:26 PM >To: SURBL Discussion list >Subject: Re: [SURBL-Discuss] Revised DMOZ data, got Wikipedia domains >too > > >----- Original Message ----- >From: "Ryan Thompson" <ryan(a)sasknow.com> > >> >> I wouldn't appreciate it being whitelisted as then if >there's abuse, >> >> and it does get blacklisted, there's no pressure on the >domain holder >> >> to clean up. >> > >> > I'm hoping UC list will help with that. Anyone who wishes to use it >> > can add a minor score to spam for UC hits. This would put at least >> > some sort of preasure for these grey hats to do something. >> >> Ezzactly. Even if it's worth 0.1 points in SA (or >something), the stigma >> of having one's domain listed in some sort of bad "list" >might be enough >> to promote action, provided the accuracy of UC is quite >good. (I.e., we >> definitely still don't want to list everything :-) "Quite >good" in that >> case probably means < 1% FP, but I think we can do even >better than that >> with a few clever tricks. >> >> Personally, I'm hoping UC will be worth more than 0.1 even with a >> somewhat higher FP rate, since there will be little or no >overlap with >> other SURBL tests, but that'll depend on some mass-checks. :-) > >I have been tracking my hits from UC, and although the hit >rate if fairly >low compared to most of the SURBL lists, it has been giving me >good results >and hitting things I would consider spam. And as such, I have the test >currently weighted at 0.5. > >Bill LOL, thats cause UC just ain't that big yet. I keep trying to keep up with all the domains Jeff keeps whitelisting, but he's like some sort of mad robot on overdrive ;) --Chris (Good grief I've been hungry all day!)

1 0

RE: [SURBL-Discuss] Revised DMOZ data, got Wikipedia domains too
by Chris Santerre 08 Oct '04

08 Oct '04

>-----Original Message----- >From: Alex Broens [mailto:surbl@alexb.ch] >Sent: Friday, October 08, 2004 12:51 PM >To: SURBL Discussion list >Subject: Re: [SURBL-Discuss] Revised DMOZ data, got Wikipedia domains >too > > >Bill Landry wrote: >> ----- Original Message ----- >> From: "Alex Broens" <surbl(a)alexb.ch> >> >>>Jeff Chan wrote: >>> >>>> http://spamcheck.freeapp.net/whitelists/wikipedia-dmoz.srt >>>> >>>>Please also take a look at these blocklist hits (potential FPs) >>>>and share what you think: >>>> >>>> >> >> >http://spamcheck.freeapp.net/whitelists/wikipedia-dmoz-blocklis >t.summed.txt >> >>>>Would there be many FNs (missed spams) if we whitelisted all >>>>of these? In other words are these all truly False Positives? >>>>If not, which ones do you feel are true spammers and why. >>> >>>probably not a new idea, but why not run a "wl.surbl.org" >with all the >>>whitelisted domains and ppl can choose to use it or not. >> >> >> I like this idea! Whitelist the most commonly used 1,000 or >so domains, and >> then create a wl.surbl.org for the rest of the >wikipedia-dmoz domains. > >WOW... Bill didn't bark at me this time. > >my point is the following: > >take for example "angelfire. com". This domain may have >legitimate users >but my user base would NEVER have contact with anybody hosting >a site or >anything there. If they support spam, list them, put pressure >on them to >stop supporting spammers, bla, bla, bla. >I wouldn't appreciate it being whitelisted as then if there's >abuse, and >it does get blacklisted, there's no pressure on the domain holder to >clean up. I'm hoping UC list will help with that. Anyone who wishes to use it can add a minor score to spam for UC hits. This would put at least some sort of preasure for these grey hats to do something. > >As I imagine we're fighting spam here, not just filtering, I have a >certain difficulty understanding why the world is crying for >whitelisting instead of putting pressure on so called whitehats who >support abuse for a lifetime. I completely agree, but when working with SURBL I can't use this mindset. I have to go by Jeff's vision. However with UC its a whole different ball game. > >as Chris said, you could make whitelisting a lifetime task. >I believe the better approach would be to decrease potential FP's by >increasing the reporting QUALITY !!!!!!!! > >Alex > >//Are we fighting Spam or working for Messagelabs & Co. for free? // OUCH!!! I hear ya.....but OUCH! ;) --Chris

3 2

RE: Revised DMOZ data, got Wikipedia domains too
by Chris Santerre 08 Oct '04

08 Oct '04

>-----Original Message----- >From: Daniel Quinlan [mailto:quinlan@pathname.com] >Sent: Friday, October 08, 2004 2:01 PM >To: Chris Santerre >Cc: 'Jeff Chan'; SURBL Discuss; SpamAssassin Developers >Subject: Re: Revised DMOZ data, got Wikipedia domains too > > >Chris Santerre <csanterre(a)MerchantsOverseas.com> writes: > >> I think this is just plain nuts to whitelist all of these! Why? If we >> don't try to whitelist the most popular sites, then what the heck it >> the point? We could whitelist millions of legit domains forever. The >> popular ones are the most important. > >The points: > > - whitelisting legitimate domains limits the effectiveness of joe job > attacks that result in FPs in various SURBL blacklists > - whitelisting could be used as negative points for MAIL FROM if > combined with SPF (and more domains is better) Yeah, but not everyone is using SPF yet. But if they were, sure! > >In addition: > > - I would only whitelist those domains (a) subject to editorial > removal (b) so long as their domain registration is old enough and > (c) so long as they pass other criteria such as no SBL listing for > NS->A. Yeah date seems to be key in more and more cases. Granted a spammer could buy one of these older ones, but hasn't happened often enough. > - I would maintain the automated whitelist separately from the human > edited whitelist and handle it differently. For example, perhaps > automated whitelist entries can only remove a single blacklist hit > (like SpamCop), but to remove two independent blacklist hits, it > requires a human decision. > Did you look at the example from the list I gave? It doesn't even have a web page! Just says testing. I'm all for whitelisting, but popular/useful domains only. >> so: >> -1 for adding all those intersected to WL >> +1 for whitelisting the blacklist hits. > >I think there are other options available due to the miracle of >programming. ;-) > Well like they say around here, "You can't argue with success." But taking away spam points based on an autowhitelist still makes me nervous. But you might have a few tricks up your sleeve D.Q. that I don't know about yet ;) --Chris

2 1

Re: Revised DMOZ data, got Wikipedia domains too
by Daniel Quinlan 08 Oct '04

08 Oct '04

Chris Santerre <csanterre(a)MerchantsOverseas.com> writes: > I think this is just plain nuts to whitelist all of these! Why? If we > don't try to whitelist the most popular sites, then what the heck it > the point? We could whitelist millions of legit domains forever. The > popular ones are the most important. The points: - whitelisting legitimate domains limits the effectiveness of joe job attacks that result in FPs in various SURBL blacklists - whitelisting could be used as negative points for MAIL FROM if combined with SPF (and more domains is better) In addition: - I would only whitelist those domains (a) subject to editorial removal (b) so long as their domain registration is old enough and (c) so long as they pass other criteria such as no SBL listing for NS->A. - I would maintain the automated whitelist separately from the human edited whitelist and handle it differently. For example, perhaps automated whitelist entries can only remove a single blacklist hit (like SpamCop), but to remove two independent blacklist hits, it requires a human decision. > so: > -1 for adding all those intersected to WL > +1 for whitelisting the blacklist hits. I think there are other options available due to the miracle of programming. ;-) Daniel -- Daniel Quinlan ApacheCon! 13-17 November (3 SpamAssassin http://www.pathname.com/~quinlan/ http://www.apachecon.com/ sessions & more)

1 0

RE: [SURBL-Discuss] Revised DMOZ data, got Wikipedia domains too
by Chris Santerre 08 Oct '04

08 Oct '04

>-----Original Message----- >From: Alex Broens [mailto:surbl@alexb.ch] >Sent: Friday, October 08, 2004 11:53 AM >To: Jeff Chan; SURBL Discussion list >Subject: Re: [SURBL-Discuss] Revised DMOZ data, got Wikipedia domains >too > > >Jeff Chan wrote: >> http://spamcheck.freeapp.net/whitelists/wikipedia-dmoz.srt >> >> Please also take a look at these blocklist hits (potential FPs) >> and share what you think: >> >> >http://spamcheck.freeapp.net/whitelists/wikipedia-dmoz-blocklis >t.summed.txt >> >> Would there be many FNs (missed spams) if we whitelisted all >> of these? In other words are these all truly False Positives? >> If not, which ones do you feel are true spammers and why. > >probably not a new idea, but why not run a "wl.surbl.org" with all the >whitelisted domains and ppl can choose to use it or not. > > IMHO More traffic and scan time. Better to just ignore having them in the list. Also things like Yahoo, AOL, ect... will most likely have SA code to not even bother to look those up. --Chris

1 0

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Discuss