RE: [Declude.JunkMail] Feature request: COMBO tests
I for one am quite happy with the workaround for TESTSFAILED/END. I can't speak to which versions should support it, but with Matt's guidance and the permutation builder I posted here yesterday: http://www.subterrane.com/permgen.shtml I've found remarkable precision and dexterity. Just be sure to uncheck Fill empty strings Dan From: Markus Gufler [EMAIL PROTECTED] Reply-To: [EMAIL PROTECTED] Date: Wed, 19 May 2004 23:39:58 +0200 To: [EMAIL PROTECTED] Subject: SPAM: RE: [Declude.JunkMail] Feature request: COMBO tests Gotcha. Did not know of the standard ver limits Beside the limitation for pro users (who knows if future COMBO test - if they become true - will be available in the standard version?) I consider the TESTSFAILED/END solution a little bit inflexible and inefficient and so as Matt (who has discovered this possibility) said it's nothing else then a workaround Real AND/OR/NOT functionality for a new group of COMBO tests that are processed after all other tests should allow us to assign extra points for certain suspicious combinations of tests. Ass explained several times this would allow to set up a great set of filter files for bogus virus warnings comming from other dumb av filters. Markus --- [This E-mail was scanned for viruses by Declude Virus (http://www.declude.com)] --- This E-mail came from the Declude.JunkMail mailing list. To unsubscribe, just send an E-mail to [EMAIL PROTECTED], and type unsubscribe Declude.JunkMail. The archives can be found at http://www.mail-archive.com. --- [This E-mail was scanned for viruses by Declude Virus (http://www.declude.com)] --- This E-mail came from the Declude.JunkMail mailing list. To unsubscribe, just send an E-mail to [EMAIL PROTECTED], and type unsubscribe Declude.JunkMail. The archives can be found at http://www.mail-archive.com.
Re: [Declude.JunkMail] Feature request: COMBO tests
Dan, While it certainly works, it does require extra filters because the AND kludge only works when the tests appear in a certain order. It would be nice if that issue was solved, but for now, with a bit of extra work, it's certainly an improvement and brings important new capabilities to the table. Matt Dan Patnode wrote: I for one am quite happy with the "workaround" for TESTSFAILED/END. I can't speak to which versions should support it, but with Matt's guidance and the permutation builder I posted here yesterday: http://www.subterrane.com/permgen.shtml I've found remarkable precision and dexterity. Just be sure to uncheck "Fill empty strings" Dan From: "Markus Gufler" [EMAIL PROTECTED] Reply-To: [EMAIL PROTECTED] Date: Wed, 19 May 2004 23:39:58 +0200 To: [EMAIL PROTECTED] Subject: SPAM: RE: [Declude.JunkMail] Feature request: COMBO tests Gotcha. Did not know of the standard ver limits Beside the limitation for pro users (who knows if future COMBO test - if they become true - will be available in the standard version?) I consider the TESTSFAILED/END "solution" a little bit inflexible and inefficient and so as Matt (who has discovered this possibility) said it's nothing else then a "workaround" Real AND/OR/NOT functionality for a new group of COMBO tests that are processed after all other tests should allow us to assign extra points for certain suspicious combinations of tests. Ass explained several times this would allow to set up a great set of filter files for bogus virus warnings comming from other dumb av filters. Markus --- [This E-mail was scanned for viruses by Declude Virus (http://www.declude.com)] --- This E-mail came from the Declude.JunkMail mailing list. To unsubscribe, just send an E-mail to [EMAIL PROTECTED], and type "unsubscribe Declude.JunkMail". The archives can be found at http://www.mail-archive.com. --- [This E-mail was scanned for viruses by Declude Virus (http://www.declude.com)] --- This E-mail came from the Declude.JunkMail mailing list. To unsubscribe, just send an E-mail to [EMAIL PROTECTED], and type "unsubscribe Declude.JunkMail". The archives can be found at http://www.mail-archive.com. -- = MailPure custom filters for Declude JunkMail Pro. http://www.mailpure.com/software/ =
[Declude.JunkMail] Feature request: COMBO tests
I know Scott's out (hope he's enjoying it!), but wanted to post this one while I was thinking about it. I would like to be able to group tests together and give a weight to the group rather than the individual tests. That way if one or multiple tests fail, only one weight is added. This would give us a lot more flexibility in the weighting scheme and would address such issues as mailers listed on multiple DUL lists, or mailers that are blacklisted on multiple lists. We then start to structure weighting more around the class of test rather than the quality of individual tests, allowing lots of BLs to be used that may contain duplicates, but have a few listings that others don't, without pushing the weight up further than it should. I realize others have found a workaround for the multiple DUL issue by using custom filters, but I believe this would give much more overall flexibility. Note that it also implies AND and OR functionality: AND with weight aggregation, and OR with tests within a group. I realize this is a major change, but believe it would be well worth it. Thoughts? Darin.
Re: [Declude.JunkMail] Feature request: COMBO tests
My feeling is since it can be done to some extent using testsfailed filters, I'd rather see development on things we can't do. The NOTCONTAINS arguement has simplified these COMBO filters so most only take one filter to perform. Scott Fisher Director of IT Farm Progress Companies [EMAIL PROTECTED] 05/19/04 02:10PM I know Scott's out (hope he's enjoying it!), but wanted to post this one while I was thinking about it. I would like to be able to group tests together and give a weight to the group rather than the individual tests. That way if one or multiple tests fail, only one weight is added. This would give us a lot more flexibility in the weighting scheme and would address such issues as mailers listed on multiple DUL lists, or mailers that are blacklisted on multiple lists. We then start to structure weighting more around the class of test rather than the quality of individual tests, allowing lots of BLs to be used that may contain duplicates, but have a few listings that others don't, without pushing the weight up further than it should. I realize others have found a workaround for the multiple DUL issue by using custom filters, but I believe this would give much more overall flexibility. Note that it also implies AND and OR functionality: AND with weight aggregation, and OR with tests within a group. I realize this is a major change, but believe it would be well worth it. Thoughts? Darin. --- [This E-mail was scanned for viruses by Declude Virus (http://www.declude.com)] --- This E-mail came from the Declude.JunkMail mailing list. To unsubscribe, just send an E-mail to [EMAIL PROTECTED], and type unsubscribe Declude.JunkMail. The archives can be found at http://www.mail-archive.com.
Re: [Declude.JunkMail] Feature request: COMBO tests
Yes, but only for Pro licenses and custom filtering. Using weighting groups could allow Standard licenses to do this, as well as being much faster than text processing. Darin. - Original Message - From: Nick Hayer [EMAIL PROTECTED] To: Darin Cox [EMAIL PROTECTED]; [EMAIL PROTECTED] Sent: Wednesday, May 19, 2004 4:48 PM Subject: Re: [Declude.JunkMail] Feature request: COMBO tests Hi Darin - On 19 May 2004 at 15:10, Darin Cox wrote: I would like to be able to group tests together and give a weight to the group rather than the individual tests. That way if one or multiple tests fail, only one weight is added. The answer is: I realize others have found a workaround for the multiple DUL issue by using custom filters [In the archives you will find exactly this example provided by Matt] This works well - a feature request for NOTCONTAINS has been made and will improve the method -Nick Hayer --- [This E-mail was scanned for viruses by Declude Virus (http://www.declude.com)] --- This E-mail came from the Declude.JunkMail mailing list. To unsubscribe, just send an E-mail to [EMAIL PROTECTED], and type unsubscribe Declude.JunkMail. The archives can be found at http://www.mail-archive.com. --- [This E-mail was scanned for viruses by Declude Virus (http://www.declude.com)] --- This E-mail came from the Declude.JunkMail mailing list. To unsubscribe, just send an E-mail to [EMAIL PROTECTED], and type unsubscribe Declude.JunkMail. The archives can be found at http://www.mail-archive.com.
Re: [Declude.JunkMail] Feature request: COMBO tests
Hi Darin - On 19 May 2004 at 15:10, Darin Cox wrote: I would like to be able to group tests together and give a weight to the group rather than the individual tests. That way if one or multiple tests fail, only one weight is added. The answer is: I realize others have found a workaround for the multiple DUL issue by using custom filters [In the archives you will find exactly this example provided by Matt] This works well - a feature request for NOTCONTAINS has been made and will improve the method -Nick Hayer --- [This E-mail was scanned for viruses by Declude Virus (http://www.declude.com)] --- This E-mail came from the Declude.JunkMail mailing list. To unsubscribe, just send an E-mail to [EMAIL PROTECTED], and type unsubscribe Declude.JunkMail. The archives can be found at http://www.mail-archive.com.
Re: [Declude.JunkMail] Feature request: COMBO tests
True...but that only works for Pro licenses (for test filtering), which those on Standard can't use. Also, this sort of group weighting would be much more efficient than text matching. Darin. - Original Message - From: Scott Fisher [EMAIL PROTECTED] To: [EMAIL PROTECTED] Sent: Wednesday, May 19, 2004 3:33 PM Subject: Re: [Declude.JunkMail] Feature request: COMBO tests My feeling is since it can be done to some extent using testsfailed filters, I'd rather see development on things we can't do. The NOTCONTAINS arguement has simplified these COMBO filters so most only take one filter to perform. Scott Fisher Director of IT Farm Progress Companies [EMAIL PROTECTED] 05/19/04 02:10PM I know Scott's out (hope he's enjoying it!), but wanted to post this one while I was thinking about it. I would like to be able to group tests together and give a weight to the group rather than the individual tests. That way if one or multiple tests fail, only one weight is added. This would give us a lot more flexibility in the weighting scheme and would address such issues as mailers listed on multiple DUL lists, or mailers that are blacklisted on multiple lists. We then start to structure weighting more around the class of test rather than the quality of individual tests, allowing lots of BLs to be used that may contain duplicates, but have a few listings that others don't, without pushing the weight up further than it should. I realize others have found a workaround for the multiple DUL issue by using custom filters, but I believe this would give much more overall flexibility. Note that it also implies AND and OR functionality: AND with weight aggregation, and OR with tests within a group. I realize this is a major change, but believe it would be well worth it. Thoughts? Darin. --- [This E-mail was scanned for viruses by Declude Virus (http://www.declude.com)] --- This E-mail came from the Declude.JunkMail mailing list. To unsubscribe, just send an E-mail to [EMAIL PROTECTED], and type unsubscribe Declude.JunkMail. The archives can be found at http://www.mail-archive.com. --- [This E-mail was scanned for viruses by Declude Virus (http://www.declude.com)] --- This E-mail came from the Declude.JunkMail mailing list. To unsubscribe, just send an E-mail to [EMAIL PROTECTED], and type unsubscribe Declude.JunkMail. The archives can be found at http://www.mail-archive.com.
RE: [Declude.JunkMail] Feature request: COMBO tests
Gotcha. Did not know of the standard ver limits Beside the limitation for pro users (who knows if future COMBO test - if they become true - will be available in the standard version?) I consider the TESTSFAILED/END solution a little bit inflexible and inefficient and so as Matt (who has discovered this possibility) said it's nothing else then a workaround Real AND/OR/NOT functionality for a new group of COMBO tests that are processed after all other tests should allow us to assign extra points for certain suspicious combinations of tests. Ass explained several times this would allow to set up a great set of filter files for bogus virus warnings comming from other dumb av filters. Markus --- [This E-mail was scanned for viruses by Declude Virus (http://www.declude.com)] --- This E-mail came from the Declude.JunkMail mailing list. To unsubscribe, just send an E-mail to [EMAIL PROTECTED], and type unsubscribe Declude.JunkMail. The archives can be found at http://www.mail-archive.com.
Re: [Declude.JunkMail] Feature request: COMBO tests
On 19 May 2004 at 16:59, Darin Cox wrote: Yes, but only for Pro licenses and custom filtering. Using weighting groups could allow Standard licenses to do this, as well as being much faster than text processing. Darin. Gotcha. Did not know of the standard ver limits -Nick - Original Message - From: Nick Hayer [EMAIL PROTECTED] To: Darin Cox [EMAIL PROTECTED]; [EMAIL PROTECTED] Sent: Wednesday, May 19, 2004 4:48 PM Subject: Re: [Declude.JunkMail] Feature request: COMBO tests Hi Darin - On 19 May 2004 at 15:10, Darin Cox wrote: I would like to be able to group tests together and give a weight to the group rather than the individual tests. That way if one or multiple tests fail, only one weight is added. The answer is: I realize others have found a workaround for the multiple DUL issue by using custom filters [In the archives you will find exactly this example provided by Matt] This works well - a feature request for NOTCONTAINS has been made and will improve the method -Nick Hayer --- [This E-mail was scanned for viruses by Declude Virus (http://www.declude.com)] --- This E-mail came from the Declude.JunkMail mailing list. To unsubscribe, just send an E-mail to [EMAIL PROTECTED], and type unsubscribe Declude.JunkMail. The archives can be found at http://www.mail-archive.com. --- [This E-mail was scanned for viruses by Declude Virus (http://www.declude.com)] --- This E-mail came from the Declude.JunkMail mailing list. To unsubscribe, just send an E-mail to [EMAIL PROTECTED], and type unsubscribe Declude.JunkMail. The archives can be found at http://www.mail-archive.com. --- [This E-mail was scanned for viruses by Declude Virus (http://www.declude.com)] --- This E-mail came from the Declude.JunkMail mailing list. To unsubscribe, just send an E-mail to [EMAIL PROTECTED], and type unsubscribe Declude.JunkMail. The archives can be found at http://www.mail-archive.com.
