Question Example of AOL Mail filter?

[ckollars]

I'm trying to construct intelligent distribution of incoming emails into different folders using AOL Mail filter rules ...and it's not going well.

The "new" AOL email seems to have filtering only of individual addresses, not filtering of entire domains - spam/notspam settings done that way are so poor at handling unanticipated new addresses that they're not very useful. That's why I'm trying to construct my own filter rules.

The available documentation I've been able to find is woefully incomplete (and I further suspect their implementation is just plain buggy). [An example unanswered question of mine is which characters in an email address are word separators: in foo@bar.baz can bar be a target of 'contains'? what about bar.baz ? or @bar ? etc?]

It seems the best way to get past all my hangups is to mimic a working example. So can anybody point me at a working set of AOL Mail filter rules?

 

[Ralston18]

No specific set (full disclosure) of working filter/rules to present but there are a number of links that can help.

Perhaps:

https://www.gmass.co/blog/aol-mail-settings/

https://help.aol.com/articles/Control-excessive-spam-email

https://help.aol.com/articles/aol-mail-mail-settings

https://help.aol.com/articles/use-filters-to-sort-and-organize-messages-in-aol-mail

The last link indicates that up to 500 filters (rules) can be created in AOL.

I suggest taking another look at AOL's email configuration settings and filtering. Read the above links while exploring the applicable email/filtering screens.

Create individual simple rules versus some large complicated rule.

Try to start with a couple of filters that will eliminate or sort out the most common spammers etc..

Keep in mind that the rules/filters are processed in some sequential manner (top to bottom) and that there may be some unexpected side effects as a result.

Also: do not use any "Unsubscribe" rules . All that may do is to tell the sender that your email address exists and is live. Spammers will not honor "unsubscribe" any way. Legitmate (?) emailers may honor "unsubscribe" but only for X amount of time. Then they come back again.

Personally (Outlook user) I use whitelisting. Everything else goes to a folder that I can scan through and cull as necessary.

 

[ckollars]

...

Thanks for trying, but I really need a working example.

AOL has changed their UI very recently, and "Options" (and most of the settings it used to give access to) no longer exists. And all the AOL "help" pages are woefully incomplete.

Unlike any other email I'm familiar with, AOL Webmail scanning/whitelist/etc. occurs TWICE: once internally, and a second time with the user-defined filters. And there seems to be very little commonality between the two passes.

And the current AOL Mail seems to just plain NOT work exactly the way it's documented. For example I've found cases where the seemingly simple principle of rules being processed _in _order until one of them matches is not actually the case.

I've produced complex sets of email filter rules for other programs --going clear back to `procmail`-- that do things that once in a while are actually better than generally available tools like `spamassassin`. But the current AOL Mail has me stumped.

After a couple weeks of experimenting and web prowling, I'm beginning to suspect there are so many flaws that it's just plain not possible to produce usable handling of email with AOL "filters" - that's one of the reasons why I want to see a WORKING EXAMPLE.

 

[Ralston18]

Hopefully there will be some aol email user able to respond.

In the meantime, consider posting a couple of examples of what did not work.

Some aol emails that got paste the filters.

Show the filter(s) and show what got through.

Perhaps a couple of simple filters/rules and then a couple of more complex (logic) filters and rules.

Just redact personally identifying information or use a pseudonym if names are included in the filter.

Not trying to support/justify aol.

Simply curious about the nature and scope of the filtering problems.

 

[ckollars]

This is NOT (currently) about matching or not matching. It's about the undocumented and inconsistent behavior of filtering rules.

Just finding out for sure which characters are "word separators" in email addresses would help considerably (i.e. in foo@blarney.baz, is blarney a matchable word? - and how is it different in foo@bar.blarney.baz?). But basic things like this are not documented anywhere in the AOL Help or anywhere else that I've been able to find. I can figure it out by slowly testing different permutations - but that seems a silly waste of time if someone already has the answer.

I could provide several different problem illustrations (including occasional apparent flouting of the "topmost first" meta-rule). But first I'd need to vet those illustrations carefully to be sure they aren't marred by typos or confusions. In the meantime I'm happy to solve my problems myself rather than have someone else solve them for me. But in order to do that I need a WORKING EXAMPLE.

 

[ckollars]

Here for your amusement is one simple illustration of puzzling behavior that contradicts the documentation:

on AOL webmail at http://mail.aol.com...
Create two new folders: 'badword' and 'main'
Get into the list of filter rules (Click gear icon then click more settings at the right, then select filter at the left)
Enter a single rule: %Badword - Deliver to "Badword" if Subject contains "badword"
send a test email from another account with subject 'Aaa badword bbb'
verify that the received email is indeed correctly moved into the 'badword' folder
now add a second rule: ^default - Deliver to "Main" if To/CC ends with "aol.com"
then send a second test email, again from another account, with similar but identifiable subject 'Aaa badword ccc'
rules nearer the top always take precedence, so the additional rule at the bottom won't affect processing of the test email at all, right?
so -as before- look in the 'badword' folder for the second test email ...AND IT'S NOT THERE
now look in the 'main' folder ...AND IT'S THERE INSTEAD --even though the new rule contains the only reference to that folder, and the new rule is at the bottom so it should have no effect, as the received email has already matched an earlier rule