Προσοχή στις απάτες! Δεν θα σας ζητήσουμε ποτέ να καλέσετε ή να στείλετε μήνυμα σε κάποιον αριθμό τηλεφώνου ή να μοιραστείτε προσωπικά δεδομένα. Αναφέρετε τυχόν ύποπτη δραστηριότητα μέσω της επιλογής «Αναφορά κατάχρησης».

Μάθετε περισσότερα

Community Input on Spam

  • 3 απαντήσεις
  • 1 έχει αυτό το πρόβλημα
  • 3 προβολές
  • Τελευταία απάντηση από Matt

Just curious whether or not Mozilla takes into consideration any community input with regard to what it flags as possible spam. In other words, when I mark something as "Junk", does Thunderbird communicate that to some central computer so that, when enough users do likewise, it will start flagging that type of message as possible spam? Is there a setting that allows me to participate in that kind of input? If not, who decides what is flagged as possible spam? How is the decision made by the software?

Just curious whether or not Mozilla takes into consideration any community input with regard to what it flags as possible spam. In other words, when I mark something as "Junk", does Thunderbird communicate that to some central computer so that, when enough users do likewise, it will start flagging that type of message as possible spam? Is there a setting that allows me to participate in that kind of input? If not, who decides what is flagged as possible spam? How is the decision made by the software?

Όλες οι απαντήσεις (3)

jwf said

Just curious whether or not Mozilla takes into consideration any community input with regard to what it flags as possible spam. In other words, when I mark something as "Junk", does Thunderbird communicate that to some central computer so that, when enough users do likewise, it will start flagging that type of message as possible spam?

No.

Is there a setting that allows me to participate in that kind of input?

No because it does not exist as a process. Your definition of spam and mine are probably very very different. I define spam as unsolicited commercial email. Others define it as any unwanted email and have trouble with spam filters because they generally do not use that definition.

If not, who decides what is flagged as possible spam? How is the decision made by the software?

You teach it by marking messages as spam and others as not spam. Initially it "guesses" until the training file has some substance (like 50 is spam decisions by you and 50 not spam decision by you for false positives.) See https://support.mozilla.org/en-US/kb/thunderbird-and-junk-spam-messages

Thunderbird uses your address books as a "whitelist" and never regards mail from folk in the selected address books as junk. That is the only reference to the email address in junk filtering. Everything else in making the decision is based on message content. True spam rarely uses a valid from address (unless it is someone elses) and is not considered.

Thank you, Matt -- yes, all that is pretty clear and makes good sense . . . BUT, there is still a lot of email that comes to me from advertisers and political bloggers that I'm quite sure I did not sign up for. Also, some of it keeps coming even after I request to be unsubscribes (which I only do if I am pretty sure the sender is just rude and not malicious). And some of it keeps coming even after I have flagged it, myself, several times. SO, I have often wished that there was a function in Thunderbird to report all those irritating messages to a central processor so that when a certain threshold is reached-- i.e. when a SUFFICIENT number of users complain --Thunderbird would automatically flag those types of messages (which have a recognisable form, if not a valid return address).

NOTE: To be fair, I've used this email address for 25 years or so and have shared it pretty indiscriminately on the internet. A new email that I have been using for only a couple of years-- and which I do not use in discussion groups or online accounts --has, so far, been spam free.

I get political garbage, here in Australia our government deliberately exempted themselves and the vote scrounging from the anti spam laws. Just like they can ring anyone and drive them nuts even after you register on their own do not ring database.

The solution to this sort of thing is a filter. The article is getting dated, but the information is still valid. https://support.mozilla.org/en-US/kb/blocking-sender This has seen more updates and is a little more informative about filters in general https://support.mozilla.org/en-US/kb/organize-your-messages-using-filters