|
In response to: I think it's important for people to indicate how they
would like to see such things handled...
I can't say I was really appreciative of how things
were handled.
I believe a base test failure should have at the
bare minimum had a declude.junkmail mailing list announcement. Especially when
the emergency workaround is as simple comment out one line in a config
file.
And I also didn't appreciate the comment that it is
a minor test. To Declude's credit, Declude is one of the most customizable
programs I have ever encountered and has a very active customizing community on
the declude.junkmail mailing list. While it may be a minor test in the default
global.cfg, apparently others are using it more
significantly.
I'd like to see a better tracking of bugs and when
they are fixed or possible workarounds. How about a webpage listing known
problems? Some of this information is in the
release notes web page. This useful page on the Declude website has always
been one of the most buried. I know (and can even respect) Scott's policy on
documenting interim's, but I always hoped that documentation was an area that
CPHZ would improve. On a plus note, I think the base Junkmail /
Virus Online Manuals have improved.
Looking back, I did prefer the interim days. I
routinely downloaded the latest interims and used them. The interim release were
extremely stable and I only recall one problem while I was living on the interim
edge. I had confidence in the interims. Interims and hence improvements came out
with some frequency.
I haven't had the confidence to move to the 2.0
beta even though there are some fixes/additions that I desire. Interestingly I
can't pinpoint why I lost the confidence, but I have.
Lastly a general impression. Being an Imail/Declude
user, I've been feeling slightly neglected. I understand that Declude needs to
work on other email servers to keep remain a viable business, but that also
means that the programming effort is being used to get the SmarterMail (and
maybe other servers) working, and not on improving the base product. I
think that there have been important upgrades to the Declude Virus, but I feel
that Declude Junkmail has been treading water, or even slipping behind other
SPAM products.
----- Original Message -----
Sent: Monday, January 03, 2005 7:25
PM
Subject: Re: [Declude.JunkMail] 2005
SpamHeaders Glitch?
Just an FYI, Barry did call me this afternoon, and while the
exact approach that they would take wasn't shared, it was clear that he
understood the general need. This has only become an issue for us
because of the change in how things are released as interims were formerly
plentiful, very quick to be published, and Scott had laid down the law
concerning his unwillingness to provide documentation for interim
releases. Not that long ago there was no such thing as a broadcast
announcement (if I recall correctly), and only major issues (bugs and fixes
for new important issues such as virus detection) were announced to the
list. Now CPHZ has the resources to do all of these things better
(broadcast notifications, better documentation, etc.), but this caught them at
a stage where they had not yet developed a process to deal with such things,
possibly not fully understood the implications, and of course the holiday
helped to compound the issue.
I was just trying to piece together my
opinions in order to influence the decision as to how to implement this.
Since they do look at feedback and at least Barry and Scott do monitor this
list, I think it's important for people to indicate how they would like to see
such things handled. If this was Ipswitch that we were talking about,
our expectations would have been to hear virtually nothing until a week or two
later when the hotfix came out, and if the issue was not so widespread,
expecting a fix would be assuming too much in many cases.
My
expectation is that the response to the need will be calculated, but obviously
not as timely as some of us might have expected. It's probably good that
people haven't given up on having high expectations for them like we have for
others :)
Matt
Bill Landry wrote:
I agree with your comments, Matt. The other thing that has frustrated me is
the fact that a bug will be fixed in an interim release and no mention of it
will be made on the list until someone else complains about the problem on
the list. Then there would come a response, "oh, that was fixed two months
ago in interim release x.xx".
When bugs are reported to Declude that affect how the product functions,
Declude should make it a point to report those issues to its user base, or
at least to the list. They should also announce immediately when a bug has
been fixed so that we don't sit around twiddling our thumbs waiting for a
fix that's been available for two week or two months, or struggling with a
problem that's been fixed.
I held my tongue on this one, but was quite astounded that Declude did not
send out a customer notification immediately after this bug was reported.
Especially since this is a test that is enabled by default in the basic
global.cfg. I would venture to guess that a lot of people have their tests
pretty tightly configured, so that even a small weight addition could
trigger hold, or worse, delete actions to be taken. User need to be
notified right away about bugs like this so that they can decide if they
need to make changes or not. Heck, we even had people thinking that there
were problems with JunkMail plug-ins like Sniffer. Would have saved
everyone time and frustration had a notification message been sent out
immediately to all customers.
The other thing that has bothered me about this particular situation is the
rationalizing/excuses that have been posted as to why action was not taken
sooner. I would feel much better if Declude would have just owned up to the
fact that they dropped the ball on this one and promised to do better next
time.
Oh well, just my unsolicited opinion (they're a dime a dozen, you know)...
Bill
----- Original Message -----
From: "Matt" <[EMAIL PROTECTED]>
To: <[email protected]>
Sent: Monday, January 03, 2005 3:38 PM
Subject: Re: [Declude.JunkMail] 2005 SpamHeaders Glitch?
R. Scott Perry wrote:
The main reason this wasn't done was because it wasn't clear that this
was going to be as big an issue for our customers as it turned out to
be. The thought was that since this is normally a relatively minor
test, anyone that it does affect adversely would just comment out the
test.
IMO, anything that has a measurable detrimental affect on all systems
and all E-mail is very well within the bounds of what needs to be
communicated from my perspective, even if it is only scored at one point
in a default config. Not even a second thought.
My issue was similar to Kami's where I was using the test is
combinations to add extra weight, and the bug had the effect of making a
false positive with a single test much worse. It would have taken me
hours to clean everything up if I had not known about it until this
morning.
Even regarding other far more minor bugs; I've spent many wasted hours
trying to diagnose what was going on with bugs that were already known
to Declude. If such information was available to me by list or by site
of known issues, I would certainly save myself time and also prevent
other issues from occurring that I wasn't aware of. Take for example
the Subject parsing bug that was discovered with the introduction of
Yahoo's Domain Keys. I had two other people report to me issues with my
GIBBERISHSUB filter because of this bug, and at first when presented
with it, I didn't realize that this was the bug that was reported on
this list until I looked at it for about 15 minutes and suddenly
remembered. So something as minor as the bug that was primarily
affecting only messages from Yahoo, and was mostly only causing issues
with a somewhat common custom filter, in fact had some effect. I'm
afraid that everyone running GIBBERISHSUB right now is scoring the
majority of messages from Yahoo because of this, a fact probably
completely overlooked at Declude when determining what to do with it.
I think what is best is to allow us to determine what information is
useful and what isn't, but naturally within a reasonable limit. I
consider having access to brief descriptions of all known bugs upon
discovery to be highly valuable, and a time saver for myself as well as
something that will help me improve my QOS. I would prefer this
information to be 'pushed' to me in E-mail, but I would be happy with it
any way that I could get it. If you do decide to push it, you might
want to include the option to join a list for this purpose as part of a
more generalized announcement or in the footer for the listserv. I'll
bet that if made aware of the option, a large number if not most Declude
admins would choose it.
Matt
--
=====================================================
MailPure custom filters for Declude JunkMail Pro.
http://www.mailpure.com/software/
=====================================================
---
[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/
=====================================================
|
