Talk:Comparison of DNS blacklists
NPOV dispute - Observe Caution
The tone of this section is clearly not neutral. While observing caution is a commendable quality, I don't think it has any place in this comparison article. There is already a DNSBL#Criticisms section which deals with issues people may have with the way DNSBLs operate. Specifically, I take issue with statements such as "...blacklists were established by angry victims..." and "...take everything with a grain of salt". I believe the entire section could be removed, as the DNSBL page already deals with these issues appropriately. Samuli64 (talk) 06:03, 10 August 2012 (UTC)
The above is correct. The tone and wording is belligerent at best. Worse, the "references" are utter garbage.
- 13 is relevant to the claim, but is a forum posting.
- 14 and #15 are listed as proof of the author's claim that a defunct database that returns a fail to all requests is somehow 'damaging' to a server. Both links reference the company's statements about how/why the database is returning "blacklist all," but do not support the claim of damaging servers.
The section could stay, but should be a NPOV of (perhaps) suggested best-practices, rather than a weasel-worded diatribe.
Expanding the list
Does anyone have any specific suggestions for expanding this list? Doesn't look that bad to me, and I couldn't find any mention of it at the 'articles for expansion' page. Wesley 19:05, 23 April 2007 (UTC)
- I think the tag is OK. If someone knows a list (not his/her own, WP:COI) adding them to the list would be nice. Erik Warmelink 14:04, 3 June 2007 (UTC)
- The ISP homemail.com.au a.k.a grapevine.com.au apparently uses "the respected blackholes.wirehub.net RBL check server". I can find very few references to them on Google, and none on Wikipedia. Quite secretive, maybe they want to avoid Denial Of Service attacks by spammers? Anyone know more about them? —Preceding unsigned comment added by LimitedNews (talk • contribs) 01:19, 18 June 2008 (UTC)
The operator of that site seems to be quite forgiving for sloppy mailers (from How the data is compiled):
This include may include some amount of backscatter. As best I can, I have set up filtering rules to eliminate backscatter and outscatter. However, there are so many different kinds that it’s difficult to programmatically prevent them from slipping into the spam feed.
- In my opinion, backscatter is spam, it is unsolicited bulk e-mail.
Some mail is double opt-in (confirmed opt-in), and some is not.
- A mail source sending to N addresses which opted-in themselves and many more addresses which were "opted-in" by others (even if harvested by the sender), will still be considered ham.
Senders that misuse addresses are removed from this feed and lose their “ham” status. However, since I did legitimately give them an address, I don’t usually redirect them into the “spam” feed.
- After unsubscribing, it would be spam (especially when considering the previous point).
All that said, the worst part is that the site doesn't consider one-to-few e-mail at all. It only considers marketing lists and no one-to-few e-mail at all. Erik Warmelink 14:04, 3 June 2007 (UTC)
Only the ham data is suspect
It should be noted that the only issue with stats.dnsbl.com's data is the ham count. I think the site is still one of the most valuable ones out there because it very clearly and concisely displays a listing of spam-catching accuracy by percentage. Stats.dnsbl.com introduced me to psbl.surriel.com and for that I am extremely happy. Using psbl + spamhaus + spamcop has made me a very happy camper, even though psbl is the lightest-weight of the lot (I report to spamcop, so it picks up more for me than for read-only users). I only know of one other reference (SDSC) that has this kind of data, and it's not as presentable (no percentages, presented in three tables, and the total counts include ham). Adam KatzΔtalk 21:50, 6 February 2009 (UTC)
It looks like dsbl isn't going to come back so perhaps should be removed from the list - they state:
DSBL is GONE and highly unlikely to return. Please remove it from your mail server configuration.
- See above comment by Adam_Katz. I concur - though I just use spamhaus then PSBL 126.96.36.199 (talk) 01:39, 1 June 2010 (UTC)
This desperately needs a whitelist table as well.
People using these things may not be aware that they can use other DNS whitelists to help avoid some of the collateral-damage and other "mail loss side effects" by augmenting their protection with whitelists. 188.8.131.52 (talk) 13:36, 29 April 2012 (UTC)
ORBITrbl has been discontinued, so it should be removed from the list of RBLs. As of (approximatly) 2014-11-17 14:20 UTC all RBL queries to ORBITrbl return a negative result (indicating "service unavailable"). I believe this is so that anyone attempting to use ORBITrbl will turn it off (lest their incoming mail be effectively blocked).
Source: My mail logs and Blog post by Mark E. Jeftovic <markjr () easydns com>
Collateral Damage and SORBS
Noted that someone marked every zone in SORBS as collateral damage.. this would be deliberate disinformation based on other listings not being collateral damage.
The Proxy and Relay zones are and always were purely tested and confirmed - there was and is no collateral damage.
The SORBS spamzones with the exception of 'escalations' is purely spamtrap (whether the SORBS Admins' inboxes or spamtraps provided) driven.
The only SORBS zone that includes deliberate collateral damage is the 'escalations' zone and any aggregate zone that contains it.
If the consensus is any zone that that keeps listings until delist requests occur, then you *must* apply the same policy to all the other DNSbls - including Spamhaus otherwise you are very much pushing a POV and false/misleading information.
The above information has been correct for SORBS since late 2010, so there is no excuses for the deliberate misinformation. The SORBS 'express-delisting fine' was removed from the SORBS systems at the same time.