moderated Consider days that have elapsed in the "removed for marking as spam" function #suggestion


Bruce Bowman
 

Mark -- I had a Hotmail subscriber recently trigger the FBL mechanism on a message that was 19 months old. Seems he was clearing out his mailbox and simply hit the wrong button on one of them.

It would be nice if the system logic considered this possibility and didn't cause an unsub if the "spam" message is older than [say] 25 days. This might also help mitigate the similar problem that we occasionally run into with those who never bother to check their spam folder.

Thanks for your consideration,
Bruce


Donald Hellen
 

I believe this might prevent some issues if this can be done. Does the FBL have
anything defining a period of time in which it has to work?

Donald






On Sun, 02 Oct 2022 10:22:37 -0700, "Bruce Bowman" <bowman46118@...>
wrote:

Mark -- I had a Hotmail subscriber recently trigger the FBL mechanism on a message that was 19 months old. Seems he was clearing out his mailbox and simply hit the wrong button on one of them.

It would be nice if the system logic considered this possibility and didn't cause an unsub if the "spam" message is older than [say] 25 days. This might also help mitigate the similar problem that we occasionally run into with those who never bother to check their spam folder.

Thanks for your consideration,
Bruce




----------------------------------------------------
Some ham radio groups you may be interested in:
https://groups.io/g/ICOM https://groups.io/g/Ham-Antennas
https://groups.io/g/HamRadioHelp https://groups.io/g/Baofeng
https://groups.io/g/CHIRP https://rf-amplifiers.groups.io/g/main


 

Hi,

On Sun, Oct 2, 2022 at 3:07 PM Bruce Bowman <bowman46118@...> wrote:

It would be nice if the system logic considered this possibility and didn't cause an unsub if the "spam" message is older than [say] 25 days. This might also help mitigate the similar problem that we occasionally run into with those who never bother to check their spam folder.

I've set it so that if we get an FBL report on a message older than 30 days, we will ignore it.

Cheers,
Mark 


KWKloeber
 

Bruce

Under what scenario would a gio member receive spam from the gio messages sent out?  I'm confused.  I have never. 


Duane
 

On Wed, Oct 19, 2022 at 01:42 PM, KWKloeber wrote:
Under what scenario would a gio member receive spam from the gio messages sent out?
They don't, as such.  It's the member or their email provider marking them as spam.

Duane


KWKloeber
 


Under what scenario would a gio member receive spam from the gio messages sent out?

They don't, as such.  It's the member or their email provider marking them as spam.

Duane

***
That's what I thought as well. 

I THINK, that I understand that mail from the gio turn-around can go to only addresses that are subscribed members?  Is that how it works? 
Can a subscribed member (or someone who hijacked a member's email) send out spam to the group membership and, I suppose, cc: a non-member to receive spam? 


KWKloeber
 

Mark

From my interpretation of the reply to my question, actual spam being sent thru the gio system is (nearly?) impossible.  I have never received Spam as such, except in a rare circumstance of a nefarious new member (not vetted) sending telemarketing or some garbage links out.  I don't know of other instances of real spam that occur w/ the GIO message-turn-arounds.

Certainly, I may be off base on that ass/u/umption, but presuming that it is true simply for this wishy-washy IMO post, the tossing of members for that reason causes problems - it doesn't solve any problem and doesn't address the issue very well.  Merely adding a "days in A/R" test to decide whether the spam report is intended vs being inadvertent -- is a band-aid on (IMO) a poor system that can create more work for mods and others.  

Do you have numbers about how many spam bumps involve actual spam (vs. an inadvertent click, or the member's birth year, or ignorance, or straight-out belligerence re: marking msgs as spam)?    I would imagine the number involving actual spam is low.  But, the underlying reason is irrelevant.

I have said, and still maintain, that the system is punishing the innocent for the wrong reason.  If a member marks a GIO message as spam, it by definition is not spam.  The member is a member and chooses to be a member.  If s/he doesn't want to receive group messages then they wouldn't be a member.  But the system ass/u/mes that they WANT to be a member -- hence the auto msg and link to reinstate the subscription.   So, the reality is that the system is disingenuous -- it removes members because "they don't want to receive mail" BUT it really doesn't believe that's the true reason. 

In trying to weed out inadvertent spamozolas, the system should means test such things as; 1) Is this the first report by a member (if so, put it in a pending queue), 2) Was there a second spamozola received?  3) Were emails received (how many?) in-between the two that the member did not mark as spamozola?  Such tests would help weed out actual spam reports vs. inadvertent clicks (which the system could then ignore.) 

However even that testing, too, seems like a larger band-aid.  Members who receive actual spam do not want to become non-members.  They want the spam to stop not quit their membership

The system doesn't address the real issue if, indeed, it was actually spam that was received -- it only creates a further problem for that innocent member and potentially for Mods and potentially you and GMF -- with generated Why?/How? questions.  Spam may possibly be a problem that cannot be solved but providing a system that doesn't address the real problem doesn't make it go away.

At a minimum, the system should means-test the spamozolas that it receives (as above.) but as well it should auto-send a msg that "You reported..... bla bla".  and provide response choices, such as 1) The msg was not spam 2) Ignore all future spamozolas from me 3) I DO WANT to be removed from the              group. etc.

I possibly misunderstand the whole system/purpose/intention, but whatever it is, it doesn't work well.  Just MO of course.



 

Ken,

On Fri, Oct 21, 2022 at 2:22 PM KWKloeber via groups.io <KWKloeber=aol.com@groups.io> wrote:


From my interpretation of the reply to my question, actual spam being sent thru the gio system is (nearly?) impossible.  I have never received Spam as such, except in a rare circumstance of a nefarious new member (not vetted) sending telemarketing or some garbage links out.  I don't know of other instances of real spam that occur w/ the GIO message-turn-arounds.

Please remember that the FBL system is mandated by the big email providers; if we ignore the reports, we risk being banned. Also please remember that the FBL reports are indications that messages have ended up in a user's spam folder. That is not the same as 'the user received spam'.


Mark


Donald Hellen
 

Ken . . .

On Thu, 20 Oct 2022 21:36:58 -0700, "KWKloeber via groups.io"
<KWKloeber@...> wrote:

Spam may possibly be a problem that cannot be solved but providing a system that doesn't address the real problem doesn't make it go away.
It was hard to read your message with all text in one long paragraph (with no
paragraph breaks), but if I understand what you're stating, it appears you
misunderstand what the FBL (feedback loop) is all about.

In almost all cases, it has nothing to do with the member receiving actual spam.
A message in the member's inbox of the email provider was marked as spam. That
can be by a poorly trained spam filter that puts a "good" message in the spam
folder. Or the email provider marked it incorrectly as spam and it ended up in
the member's spam folder.

When this happens, if the member reads the message and deletes it while it's in
the spam folder, the member is telling the email provider this is spam, because
he did not move it out from the spam folder before it was deleted. Had the
member moved the message to the inbox, then deleted it, then the message would
not be counted as spam and no FBL occurs. So it gets marked as spam and started
the FBL to unsubscribe the member from the originating group.

Say the member doesn't see this good message incorrectly routed to the spam
folder. Some email providers delete all messages left in the spam folder
periodically, like after a message has been ignored there for 30 days. The email
provider deletes the message, which triggers the FBL, which unsubscribes the
member. All with the member doing nothing other than not understanding his
responsibility to check the spam folder regularly, dragging good messages from
it to train the spam filter. Email providers don't seem to go out of their way
to explain how to responsibly manage their email, so they are often to blame for
this happening without the member taking any action, but rather by inaction on
the member's part.

Next on the common problem list with this is the member uses an email client
like the one for Yahoo Mail, on a smartphone. Members have reported to me that
the delete as spam button is so close to the delete button that members can all
too easily tap the wrong button. This happens. A lot. And some people seem to do
it repeatedly every few days or weeks and don't learn from this mistake.

There is also the unlikely scenario where the member drags the good message INTO
the spam folder, then it gets deleted by the email provider or member. This is
unlikely, of course, unless someone spammed the GIO group, and the member, not
understanding the repercussions of his action, drags it into the spam folder,
where it eventually gets deleted and triggers the FBL.

I might have missed some ways this can happen, but this just shows that the
email providers did not think this feedback loop out very well, and it's subject
to false or accidental triggers. They didn't ask their email account holders how
best to make this work, and we have a broken system that doesn't do much at all
to handle or prevent spam.

The apparent intent was that if someone didn't care to read the instructions
sent to them by the groups they join on how to unsubscribe, they could just mark
a message from a group they wanted to leave as spam and let the FBL take care of
their unsubscription. I don't have statistics, but I'd be willing to bet that
much less than 1 out of 100 instances of a member being unsubscribed from a
group happens because of marking a message as spam actually was an intended act
by a member to leave the group. This is very unlikely. More often, if a member
wants to unsubscribe and don't know how, they'll post to the group asking to be
deleted from the membership. They don't realize that there is a link on every
message to do this (and members of the group are quick to point this out when
they ask to be removed from the group), and they don't realize that they could
trigger the FBL and just mark one message as spam to get unsubscribed through
that mechanism. When the FBL is triggered, it's almost never intentional (so as
to leave the group), so it doesn't work as the email providers hoped it would.

So we're stuck with a broken system and must live with it. You can try to
educate your members, but often it just starts a heated discussion on how this
isn't the right way to do things, that GIO is at fault, etc., when it's all the
actions of the email providers and if GIO doesn't unsubscribe a member when they
receive the request from the email provider to unsubscribe the member from the
involved group, then GIO can be blacklisted and cease to function properly,
possibly causing them to shut down completely. Explaining this often just
confuses members more as it doesn't make sense that this should happen. And for
good reason, because it shouldn't.

As for a message that was marked as spam after a certain length of time starting
with when the message was received, and GIO not following the FBL path, they may
still risk being blacklisted. Any change to how the FBL works would have to be
implemented by the email providers, and probably agreed upon by all the major
ones, which is probably not very likely to happen.

Donald



----------------------------------------------------
Some ham radio groups you may be interested in:
https://groups.io/g/ICOM https://groups.io/g/Ham-Antennas
https://groups.io/g/HamRadioHelp https://groups.io/g/Baofeng
https://groups.io/g/CHIRP https://rf-amplifiers.groups.io/g/main


KWKloeber
 

Mark

Donald and i have had off-line conversations bout this. 


Being the rules set by the big guys, how does the 30-day square with that?   Aren’t we potentially blacklisted anyway?  Or is there an acceptable window there for non blacklisting?


 

On Mon, Oct 24, 2022 at 2:13 PM KWKloeber via groups.io <KWKloeber=aol.com@groups.io> wrote:

Being the rules set by the big guys, how does the 30-day square with that?   Aren’t we potentially blacklisted anyway?  Or is there an acceptable window there for non blacklisting?


The short answer is I don't know. They won't tell us. I thought the 30 day window was a reasonable compromise, but I may be wrong. There's no way to know.

Mark