Suggestion: "Don't log for hidden" OR rejected [@]

[sg-since03]

As mentioned in another thread - spammers are creating sg addresses on their own, either by slightly altering user-created addresses, or creating entirely new ones. (Using user name + entirely new word.)

So OK, that got old, & I'm now using watchwords. And OK, it was fun for a while seeing spammers' creations DOA in the "Last 3 eaten message(s)" log, especially since they are unlinked. (DOA, after all.)

But now that's old too. I don't want to even see these _attempts_. Please make the "don't log for hidden addresses" option exclude those addresses from the list as well. Thanks.

Addendum: that list is useful for occasionally catching and following up on email that shouldn't have been deleted. Logging those phony addresses interferes with that function.

[SysKoll]

Hmmm, that's a tough one. What would be your criteria to decide whether or not an eaten message should be logged, then?

[sg-since03]

Hmmm, that's a tough one. What would be your criteria to decide whether or not an eaten message should be logged, then?

"Extra" messages sent to "expired" existing addresses only.

We'll call the address types "Viable" and "Non-Viable"
"Viable" = "all addresses which have forwarded at least one message." "Viable" by definition includes both "Hidden" and "Visible" [in address list. "Non-Hidden" ??]
"Non-Viable" = "all addresses which 'have not now, nor will they ever' forward any messages." Also defined as: addresses which do not/will not appear in a user's address list.

So "Don't log for Hidden" should also imply "Don't log for Non-Viable." No opinion on whether the default setting ("log for all") should include logging of "non-viable" addresses.

The situation is rare enough - what percentage of sg members use Watchwords and regularly encounter this problem? Users who set "Don't log for Hidden" wouldn't miss it, and the rest likely wouldn't notice or care. (i.e., probably would have no opinion if polled.)

[SysKoll]

Thank you for the write-up. I think I understand. Will give it a try.

[sg-since03]

Appreciate it, thank you.

[sg-since03]

Thank you for the write-up. I think I understand. Will give it a try.

Hi, this being December and all, it's probably a bad time... and I don't want to be a pest, especially since there are so many more important things going on - like your valiant efforts to keep this site up & running, for example... so I'll just ask:

Is this still on your "to-do" list for anytime in the next couple of months? I ask because my sg's "Log last three eaten" is useless - one particular variant is being hammered. The addresses are not being created, but they are hogging the log, preventing me from catching any accidental & "accidental"* deletions.

*Here's a great one: my online ink vendor. They send an email coupon every month. No thanks, 6,000 pages per cartridge lasts this home user much more than one month. So, I turn their address off. But whenever I get low on ink, I turn them on again. Seeing their address in the log reminds me to do this. There are others I'm forgetting; I could scan my loooong address list from time to time, but that would be impractical. <g>

Happy holidays, guys! And thanks for everything.

[SysKoll]

The quickest way to do this would be to change the code so that the ?don"t log for hidden address" now implies also "don't log attempts that fail because of watchwords".

Would that satisfy your requrements?

[sg-since03]

The quickest way to do this would be to change the code so that the don't log for hidden address" now implies also "don't log attempts that fail because of watchwords".

Would that satisfy your requirements?

YES! Yes, thank you! That was it, exactly.

[sg-since03]

Somewhere in the middle of the DOS storm... could you find the time to implement the change agreed-upon above?

Here's a sample from today's "Eaten Message Log": "From: "CastingDept@public.eastequal.com, To: ***phone-tg." Again, this address was never created due to watchwords, but it *is* still clogging up the log.

Aside from the weird format - asterisks are valid characters in email addresses?? - nothing in the "from" or "to" addresses resemble anything I've done online, certainly not with sg.

And so, my "Eaten Message Log" is useless. Help!

[SysKoll]

OK, I'm on it. I coded the feature -- really simple actually -- and now it's a matter of testing it.

You can thank my wife's new puppy who whined long enough to get me out of bed. And then I had to clean its poop. And then I needed some coffee to get me over the puke-inducing smell. And then I was wide awake while the rest of the household was asleep. Might as well code, right?

Code: Select all

        ,""",_
        ( };:_)
 /=*,  ,'~;,=~
!    ,/;';';)
`\,;';',:,','
 (;:',:',<;(
 !:,:)''' ",\
 );,/      "_)
<;<       /   \
 ",)     |     )


So I guess the despicable little bugger has some marginal usefulness after all.

[sg-since03]

Might as well code, right?

"Code we must"

OK, I'm on it. I coded the feature -- really simple actually -- and now it's a matter of testing it.

Hurrah! Thank you.

Now, to help you test... whoops. My test went thusly:
1. Cleared out all the spammers' junk with three emails to existing expired addresses.
2. Sent a test email to a new address which does not contain the watchword, i.e., an address which should be blocked.
3. Result - alas, it did show up in the log:
"date (UTC) 2008-03-30 17:25 from: <me> to: notgonnawork"

(I meant the address "wasn't gonna work"! Not your code! Didn't mean to "hex it"... darn.)

You can thank my wife's new puppy...
So I guess the despicable little bugger has some marginal usefulness after all.

Oh well, no doubt you'll have it working shortly, so give that dog a treat for me, OK? Some Pepto, or Milk of Magnesia...? <g>

[sg-since03]

PS - OT:
Some misguided soul has landed this board on ORDB's blacklist ("Origin blacklisted by ORDB"). Must have been either today or yesterday; that tag wasn't on my topic notification email two days ago.

EDIT: ...and now, half a day later the "blacklist" tag is gone. Did someone buy a clue for the ORDB? Hope so.

[SysKoll]

sg-since03,

Thanks for the test, but the new code isn't deployed yet! Hold on.

[sg-since03]

Thanks for the test, but the new code isn't deployed yet! Hold on.

{chides self:} "Oh, ye of little faith!"

Thanks, I can surely wait a bit longer, no problem.

[SysKoll]

sg-since03,

Can you retry your test now, please, and let us know the result?