friedPotat0 / Spam-Scores

Thunderbird add-on to display spam scores according to mail headers. Supports spam/ham score headers of Rspamd and SpamAssassin. The add-on adds a column with the overall spam score to the mail list view and shows details of any matched spam/ham rule.
https://addons.thunderbird.net/de/thunderbird/addon/spam-scores/
Other
43 stars 6 forks source link

Questions regarding score ranges #50

Closed owariysy closed 3 months ago

owariysy commented 2 years ago

Dear Mr Christian Zaenker, creator of SPAM SCORE vers. 1.3,

Hello, would you please give me more information about your add-on called SPAM SCORE vers. 1.3, which I have installed on my Thunderbird 91.11.0 (32-bit) (= for win 10),

please, first and foremost, by being so kind as to answer the following 4 questions:

1) on what methods is it based? = I mean, how does it calculate the score?

2) does the score 1.6 mean that it is MORE likely to be a spam than score 0.00, as I can’t help understanding at first glance, //OR, instead, does it mean that it is LESS likely to be a spam than score 0.00?

3) does the score -1.6 mean that it is LESS likely to be a spam than score 0.00, as I can’t help understanding at first glance, //OR, instead, does it mean that it is MORE likely to be a spam than score 0.00?

4) what is the threshold score for an email to be considered a spam? e.g is an e-mail to be considered spam past if Spam Score’s spam score is higher than 2.00 (or less than 2.00, depending to the answer to points 2. and 3. above)?

Unfortunately, this part of the “Read it” that I found at https://github.com/friedPotat0/Spam-Scores#readme , which I am quoting below, does NOT answer my core doubt (my core doubt being “whether a positive: number means the e-mail message is more likely or less likely to be spam (/including phishing, scams, attachments containing viruses, and the like, and, conversely, whether a negative: number means […]”

“The total score of each mail with an existing spam header will be displayed along with a red, yellow or green icon depending on the score. The colours are by default calculated as follows:

Negative Score GREEN   = Score greater than 2
Neutral Score YELLOW= between -2 and 2 (both inclusive)  
Positive Score RED       = Score less than -2

   The icon score ranges can be changed in the [add-on options](https://github.com/friedPotat0/Spam-Scores#options).“

[End of quotation]

Would you like to know why it doesn't solve my doubt?

Because a) although I can “guess” that the GREEN colour means “OK”, and therefore “not-spam”,

and that that the RED colour means “OK”, and therefore “not-spam”,

yet, I cannot rely on guessing: I need to read it spelled out, typed in black and white by you, to be assured that I am getting it right.

b) because, unless I get point a) above solved , (= i.e. if you don’t solve my point a) above),

and, moreover, if you don’t give me any advice/guidance/ tip about the threshold past which one usually be confident that one has spotted a real spam (i.e. not a “false positive”)⟶

I cannot figure out the meaning of the sentence “Neutral Score YELLOW= between -2 and 2 (both inclusive) “ which is therefore not helpful to me, but just adds to my frustration.

……………………………………………………………... Thank you very much,

I will be looking forward to your answers in your kind reply.

Yours sincerely,

Owariysy

P.S. wait a minute: when I copied and pasted this whole letter of mine that I had written on a thundebird page, the part of your Read me " GREEN = Score greater than 2 YELLOW= between -2 and 2 (both inclusive)
Score RED = Score less than -2 " suddenly appeared with the tags negative Score, Neutral Score, Positive score, as follows:

Negative Score GREEN   = Score greater than 2
Neutral Score YELLOW= between -2 and 2 (both inclusive)  
Positive Score RED       = Score less than -2

So, it seems that we users cannot readthe aforesaid tags because thery are hidden, encrypted, so to speak, in the HTML code: is that so?

= THIS "glitch" IS SHOWN IN THE ATTACHED= UPLOADED FILE named "See for yourself here that those 'negative-positive-neutral' tags...

See for yourself here that those 'negative-positive-neutral' tags do not show on users' PCs - 2022-07-06 at 21hours28min  Italian time See for yourself here that those 'negative-positive-neutral' tags do not show on users' PCs - 2022-07-06 at 21hours28min  Italian time

owariysy commented 2 years ago

POST SCRIPTUM: Even once you will have made those "Positive score -Negative score -Neutral score tags" visible" on our PCs and browsers, I still will need to know what you (or you guys) mean by "positive" and "negative": does positive stand for "yes it probably is spam" or "No, it probably is not spam" ,

Because, you know, if one is told "No, it probably is not spam" one usually thinks "oh, what a positive (= nice") thing!" and therefore it is quite confusing, so would you be so kind as to clarify that to me (and, possibly, other users) in advance.

Thank you,

Owariysy, from Florence, Italy

friedPotat0 commented 2 years ago
  1. on what methods is it based? = I mean, how does it calculate the score?

The add-on itself does not calculate the score at all. It just parses the scores that already exist in the mail header. At least if the mail server adds such information to the header of the mails.

  1. does the score 1.6 mean that it is MORE likely to be a spam than score 0.00, as I can’t help understanding at first glance, //OR, instead, does it mean that it is LESS likely to be a spam than score 0.00?
  2. does the score -1.6 mean that it is LESS likely to be a spam than score 0.00, as I can’t help understanding at first glance, //OR, instead, does it mean that it is MORE likely to be a spam than score 0.00?

The score 1.6 is more likely to be spam than the score 0.0.

  1. what is the threshold score for an email to be considered a spam? e.g is an e-mail to be considered spam past if Spam Score’s spam score is higher than 2.00 (or less than 2.00, depending to the answer to points 2. and 3. above)?

By default mails with a score greater than 2 are displayed as spam (red icon). However, you can adjust this in the settings of the add-on.

So, it seems that we users cannot readthe aforesaid tags because thery are hidden, encrypted, so to speak, in the HTML code: is that so?

I'm trying to guess what you mean by your question, but the spam headers are located in the headers of the mails. So you can inspect them by opening a mail and clicking on 'More' -> 'View Source'. The add-on just simplifies things for the user by displaying the headers in an easily readable form.

I still will need to know what you (or you guys) mean by "positive" and "negative": does positive stand for "yes it probably is spam" or "No, it probably is not spam"

A positive score means that the email is more likely to be spam than a negative score. I am not responsible for this decision, as the add-on only takes these values from the headers of the mails. Software that classifies the mails on the server side, such as SpamAssassin or rspamd, have decided to mark spam with a positive and ham with a negative score.


Sorry for the confusion about the score ranges. I noticed a small error in the readme. The score ranges listed there were wrong. I've already updated the readme accordingly.

owariysy commented 2 years ago

Dear friedPotat0,

thank you very much indeed for your timely reply:

you really made my day, by answering my questions, and by telling me that my e-mail helped you spot and correct a wee error in your add-on's Readme.

For, to me it means that trying to express myself as clearly as I could, was worth the time, the effort, and my bearing being scolded by my wife for being late for dinner.

Tomorrow, I will be able to re-read your answers and study them more in depth.

Good night, at a quarter past midnight Italian summer time.

owariysy


From: friedPotat0 @.> Sent: Wednesday, 6 July 2022, 22:44 To: friedPotat0/Spam-Scores @.> Cc: owariysy @.>; Author @.> Subject: Re: [friedPotat0/Spam-Scores] Questions regarding score ranges (Issue #50)

  1. on what methods is it based? = I mean, how does it calculate the score?

The add-on itself does not calculate the score at all. It just parses the scores that already exist in the mail header. At least if the mail server adds such information to the header of the mails.

  1. does the score 1.6 mean that it is MORE likely to be a spam than score 0.00, as I can’t help understanding at first glance, //OR, instead, does it mean that it is LESS likely to be a spam than score 0.00?
  2. does the score -1.6 mean that it is LESS likely to be a spam than score 0.00, as I can’t help understanding at first glance, //OR, instead, does it mean that it is MORE likely to be a spam than score 0.00?

The score 1.6 is more likely to be spam than the score 0.0.

  1. what is the threshold score for an email to be considered a spam? e.g is an e-mail to be considered spam past if Spam Score’s spam score is higher than 2.00 (or less than 2.00, depending to the answer to points 2. and 3. above)?

By default mails with a score greater than 2 are displayed as spam (red icon). However, you can adjust this in the settings of the add-on.

So, it seems that we users cannot readthe aforesaid tags because thery are hidden, encrypted, so to speak, in the HTML code: is that so?

I'm trying to guess what you mean by your question, but the spam headers are located in the headers of the mails. So you can inspect them by opening a mail and clicking on 'More' -> 'View Source'. The add-on just simplifies things for the user by displaying the headers in an easily readable form.


Sorry for the confusion about the score ranges. I noticed a small error in the readme. The score ranges listed there were wrong. I've already updated the readme accordingly.

— Reply to this email directly, view it on GitHubhttps://github.com/friedPotat0/Spam-Scores/issues/50#issuecomment-1176713907, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AZ6HY3ES5WAHGNJCMEWW2V3VSXV2RANCNFSM5224E7VA. You are receiving this because you authored the thread.