Scam mail?
-
Since some weeks ago, Thunderbird has been marking the CodeProject Insider as scam! So, I know it's not, but I was wondering which part of the Insider could be the one triggering the rule. :confused:
-
Since some weeks ago, Thunderbird has been marking the CodeProject Insider as scam! So, I know it's not, but I was wondering which part of the Insider could be the one triggering the rule. :confused:
For what its worth TB isn't doing that for CP emails to me. I believe Thunderbird learns spam by the items marked as junk already. So perhaps one of the CP emails has been marked as junk automatically and you didn't notice. Subsequent emails will then be marked as junk because that rule was not corrected. Check there are no CP emails marked as junk. Otherwise maybe you need to get TB to reset its training knowledgebase (not recommended!) In my experience the spam detection method employed by TB is becoming less and less useful. I'm hoping the next version addresses this problem.
-- The Obliterator
-
For what its worth TB isn't doing that for CP emails to me. I believe Thunderbird learns spam by the items marked as junk already. So perhaps one of the CP emails has been marked as junk automatically and you didn't notice. Subsequent emails will then be marked as junk because that rule was not corrected. Check there are no CP emails marked as junk. Otherwise maybe you need to get TB to reset its training knowledgebase (not recommended!) In my experience the spam detection method employed by TB is becoming less and less useful. I'm hoping the next version addresses this problem.
-- The Obliterator
Obliterator wrote:
In my experience the spam detection method employed by TB is becoming less and less useful. I'm hoping the next version addresses this problem.
I wouldn't hold my breath. The same enhanced OCR needed to defeat image spam would also blow the captcha system out of the water as well, and OCR of generalized hard to read text has been a holygrail for decades.
-- You have to explain to them [VB coders] what you mean by "typed". their first response is likely to be something like, "Of course my code is typed. Do you think i magically project it onto the screen with the power of my mind?" --- John Simmons / outlaw programmer
-
For what its worth TB isn't doing that for CP emails to me. I believe Thunderbird learns spam by the items marked as junk already. So perhaps one of the CP emails has been marked as junk automatically and you didn't notice. Subsequent emails will then be marked as junk because that rule was not corrected. Check there are no CP emails marked as junk. Otherwise maybe you need to get TB to reset its training knowledgebase (not recommended!) In my experience the spam detection method employed by TB is becoming less and less useful. I'm hoping the next version addresses this problem.
-- The Obliterator
Ok, there's a confusion here. I meant SCAM (as in phishing), not SPAM (as in junk mail). Through it marked the first Insider I received as Spam, I unmarked it and it never marked it as junk again. But the anti-phishing engine seems to be a separate one. Thanks for the advice, through. :)
-
Since some weeks ago, Thunderbird has been marking the CodeProject Insider as scam! So, I know it's not, but I was wondering which part of the Insider could be the one triggering the rule. :confused:
-
Ok, there's a confusion here. I meant SCAM (as in phishing), not SPAM (as in junk mail). Through it marked the first Insider I received as Spam, I unmarked it and it never marked it as junk again. But the anti-phishing engine seems to be a separate one. Thanks for the advice, through. :)
Sorry I missed the point you said scam rather than spam - that'll teach me to skim posts in the lounge! Anyhow, yes TB anti-phishing engine frequently generates false positives like this for me. Presumably it errs on the side of the caution. I believe the detection method is rather basic. I don't know for sure but in my experience one of its rules for determining scam mail appears to be to compare the originating domain (i.e. the from address) with any embedded links in the email. If an image or link in the email references a domain other than that specified in the from address it is instantly deemed to be a scam email. The same goes if the email references a fixed IP. It is quite likely CP's emails trip this mechanism - particularly if they link directly to a sponsor for example.
-- The Obliterator
-
Obliterator wrote:
In my experience the spam detection method employed by TB is becoming less and less useful. I'm hoping the next version addresses this problem.
I wouldn't hold my breath. The same enhanced OCR needed to defeat image spam would also blow the captcha system out of the water as well, and OCR of generalized hard to read text has been a holygrail for decades.
-- You have to explain to them [VB coders] what you mean by "typed". their first response is likely to be something like, "Of course my code is typed. Do you think i magically project it onto the screen with the power of my mind?" --- John Simmons / outlaw programmer
Agreed. But at the moment TB misses lots of simple spam (plain text, no images). The image spam as another war front entirely. In truth the spammers will always be one step ahead. I doubt he battle can never be won!
-- The Obliterator
-
Agreed. But at the moment TB misses lots of simple spam (plain text, no images). The image spam as another war front entirely. In truth the spammers will always be one step ahead. I doubt he battle can never be won!
-- The Obliterator
antispam is fundamentally reactive, although the current imagespam blight is a result of their taking an if you can't beat em, join em; approach to captchas. Sadly any future tech we develop to defeat forum spambots can be turned into a spamfilter buster just as easily. Killing spam permanently would probably require a ground up rearchitecting of the internet with strong authentication at every step. Given the resultant ability of totalitarianists to even more throughly oppress their slaves I have to wonder if the cure would end up being worse than the disease.
-- You have to explain to them [VB coders] what you mean by "typed". their first response is likely to be something like, "Of course my code is typed. Do you think i magically project it onto the screen with the power of my mind?" --- John Simmons / outlaw programmer
-
Sorry I missed the point you said scam rather than spam - that'll teach me to skim posts in the lounge! Anyhow, yes TB anti-phishing engine frequently generates false positives like this for me. Presumably it errs on the side of the caution. I believe the detection method is rather basic. I don't know for sure but in my experience one of its rules for determining scam mail appears to be to compare the originating domain (i.e. the from address) with any embedded links in the email. If an image or link in the email references a domain other than that specified in the from address it is instantly deemed to be a scam email. The same goes if the email references a fixed IP. It is quite likely CP's emails trip this mechanism - particularly if they link directly to a sponsor for example.
-- The Obliterator
I don't think the engine would be that easy. If TB marked as spam any mail that has urls to a different domain, then it would mark as spam emails from my friends that have links to other pages (since the email domain and the link are in different domains). I'm not sure about the fixed IP idea. In fact, TB gets just some of the messages as scam, and they all have links to other domains. So I guess it has something to do with some site they reference in those mails (I just compared some Insiders considered spam with some that aren't, but I couldn't figure out a particular site that's in all the ones marked and in none of the unmarked).
-
Agreed. But at the moment TB misses lots of simple spam (plain text, no images). The image spam as another war front entirely. In truth the spammers will always be one step ahead. I doubt he battle can never be won!
-- The Obliterator
Obliterator wrote:
spammers will always be one step ahead
The conman is always one-step ahead and hence can be chased by the cop right? :)
Vasudevan Deepak Kumar Personal Homepage Tech Gossips