Bug 7017 - replace BAYES_999 with BAYES_100
Summary: replace BAYES_999 with BAYES_100
Status: RESOLVED WONTFIX
Alias: None
Product: Spamassassin
Classification: Unclassified
Component: Rules (show other bugs)
Version: unspecified
Hardware: All All
: P2 normal
Target Milestone: Undefined
Assignee: SpamAssassin Developer Mailing List
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2014-02-24 15:14 UTC by AXB
Modified: 2014-12-18 01:06 UTC (History)
2 users (show)


Attachment Type Modified Status Actions Submitter/CLA Status
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description AXB 2014-02-24 15:14:47 UTC 
Could we replace the BAYES_999 rule name with BAYES_100

describe BAYES_99	Bayes spam probability is 99 to 100%

describe BAYES_999      Bayes spam probability is 99.9 to 100%

999% ???

instead,
describe BAYES_100      Bayes spam probability is 99.9 to 100%

would make more sense than BAYES_99 on steroids

comments? votes?
Comment 1 Kevin A. McGrail 2014-02-24 15:23:30 UTC 
A rose by any other name... I see little point in changing the name.  There is no consistency to why the rules in bayes_23.cf are named but 999 so equal sense to 100.

And as we have made announcements including BAYES_999, re-making those announcements with BAYES_100 and making people change their rulesets yet again for zero technical gain is a -1 from me.
Comment 2 AXB 2014-02-25 10:34:29 UTC 
(In reply to Kevin A. McGrail from comment #1)
> A rose by any other name... I see little point in changing the name.  There
> is no consistency to why the rules in bayes_23.cf are named but 999 so equal
> sense to 100.
> 
> And as we have made announcements including BAYES_999, re-making those
> announcements with BAYES_100 and making people change their rulesets yet
> again for zero technical gain is a -1 from me.

BAYES_999 was added/published due to a user suggestion.

had the name been discussed, maybe we could have agreed into BAYES_100, also for cosmetic reasons.

pushing the BAYES_100 rulename and announcing the name change, in advance, wouldn't cause much more havoc than the initial release did .-)

As is now we're stuck a weird looking name - singing out of tune forever.
Comment 3 Kevin A. McGrail 2014-02-25 14:27:52 UTC 
The entire item didn't go as planned. I added it to the sandbox to review ruleqa.  The fact that perceptron couldn't handle bayes rules and would auto-promote with no score was unknown.

Everything after that point was reactive.  Other than the possibility of a different name, there is zero technical benefit.

Further, the convention of existing bayes rule names supports using the bottom of the range not the top.  Otherwise BAYES_99 would have been BAYES_100 to start with.

Or they had the same naming issue ;-)

Either way, it's still a -1 from me, sorry but if you get enough people to agree, I'll switch to a 0.
Comment 4 John Hardin 2014-02-25 17:09:46 UTC 
FWIW, I'm at +0.5 right now. I agree there are no technical merits, but it seems neater to me. And BAYES_50 falsifies the "named after the low end of the range" observation.
Comment 5 AXB 2014-02-25 22:08:02 UTC 
ARG! I hope BAYES_999 doesn't break ppl's stats as it broke mine

even changing to BAYES_100 looks silly in spamreports
two BAYES_* tags....
Comment 6 Mark Martinec 2014-12-18 01:06:29 UTC 
I guess it's too late now for a change in name
after nine months of existence of BAYES_999.

Tentatively closing as WONTFIX.
Please reopen if considering otherwise.