On 12/12/2011 4:26 PM, Warren Togami Jr. wrote:
Respectfully, I am concerned by this "consensus" as it seems to have
been made without regard to the actual statistics.
The scores added can be viewed as placeholders to move the ticket
towards completion. If you have different scores in mind, please
suggest them.
However, these scores were not defined in a vacuum but were based on
systems not in masscheck.
I might note the
+1's were for inclusion, not for a particular manually set scores.
The +1's all followed after comments 7/10 which stated a particular set
of scores.
While I have strong confidence in the goodness of Mailspike based upon
its years of strong performance and good safety, these scores as set
in your commit are over-aggressive and not based on our own
statistics.
Feel free to recommend other scores. They are actually significantly
less aggressive than Joao's recommended scores and based on live usage.
* Was setting these scores manually your response to my concerns about
"reuse" and the difficulties we will face in GA rescoring? I might
even agree with this solution, although I believe it can be refined
with further discussion.
No. I made Comment 7 prior to your comments.
* Did you really intend for Mailspike to add a total of 4.1 or 4.5
points? (Given _BL is a composite rule that adds 1.0 on top of _L5 or
_ZBI.)
Yes, I did.
* Note: I never suggested Mailspike inclusion in pre-3.4.0 rules.
The scores are version encapsulated inline with the goal to not affect
pre 3.4.0. Pre 3.4.0 users should be able to define a score and use
them, however, and I would recommend that.
It was the concern over how to announce the inclusion of an RBL outside
of a new release that pushed the version block.
* Whitelists - yes, entirely different issue. Could we have a
separate discussion about our overall whitelist strategy? (Writing in
that separate bug...)
You never have to ask permission for that and I encourage you to do
so. However, if it was opened and languished for a few months to focus
on 3.4.0 from my perspective, I hope you'll understand.
regards,
KAM
