On Wed, 15 May 2019 18:47:01 +0300
Henrik K wrote:
> On Wed, May 15, 2019 at 04:14:30PM +0100, RW wrote:
> >
> > > I think the concept of scoresets is pointless these days anyway.
> > > Does someone actually run legit mailserver without bayes and
> > > network tests?
> >
> > But if you do tha
I've already moved over to using my personal gmail.
Sadly I don't have control over my corporate email account and the
signature keep reappearing.
On Wed, 15 May 2019 at 21:28, @lbutlr wrote:
> This garbafge is inappropriate for a mailing list. It is also enforceable
> BS (I use to post ever em
This garbafge is inappropriate for a mailing list. It is also enforceable BS (I
use to post ever email with this kind of garbage to a public website)
On 15 May 2019, at 10:55, Paul Stead wrote:
> This message is private and confidential. If you have received this message
> in error, please noti
On 15/05/2019, 16:19, "RW" wrote:
That's my point. It leaves little incentive to distinguish between
network and non-network runs.
Good point... I don't know the QA scripts well enough to be able to comment
more.
It does look like the net contributions during the week from jarif are
On Wed, May 15, 2019 at 04:14:30PM +0100, RW wrote:
>
> > I think the concept of scoresets is pointless these days anyway. Does
> > someone actually run legit mailserver without bayes and network tests?
>
> But if you do that you are running a score set that has been optimized
> for only network
On Wed, 15 May 2019 14:59:18 +
Paul Stead wrote:
> On 15/05/2019, 15:45, "RW" wrote:
>
> >
> >Network rules are only run every saturday:
> >https://ruleqa.spamassassin.org/20190511-r1859108-n
>
> Why is that necessary when network results should be reused? Most
> of them
On Wed, 15 May 2019 17:51:54 +0300
Henrik K wrote:
> > That not the point. Without taking account of Bayes, the other rules
> > get tuned differently. Bayes has a substantial effect on the score
> > of almost everything scanned.
>
> I think the concept of scoresets is pointless these days anyw
On 15/05/2019, 15:45, "RW" wrote:
>
>Network rules are only run every saturday:
>https://ruleqa.spamassassin.org/20190511-r1859108-n
Why is that necessary when network results should be reused? Most of
them are meaningless if retested after several days.
That's the reaso
On Wed, May 15, 2019 at 03:45:22PM +0100, RW wrote:
> On Wed, 15 May 2019 16:41:00 +0300
> Henrik K wrote:
>
> > On Wed, May 15, 2019 at 02:15:19PM +0100, RW wrote:
> > >
> > > Why are there no QA statistics for BAYES_* rules?
> >
> > How do you propose to generate such statistics, when all co
On Wed, 15 May 2019 16:41:00 +0300
Henrik K wrote:
> On Wed, May 15, 2019 at 02:15:19PM +0100, RW wrote:
> >
> > Why are there no QA statistics for BAYES_* rules?
>
> How do you propose to generate such statistics, when all contributors
> already are supposed to have fully sorted ham/spam corp
I noticed the jarif ruleset contributing net scores during nightlies a few
weeks ago - I've asked jarif about this but couldn't see an immediate
problem/solution.
I've also raised a potential issue on the ruleqa RE some potential problems.
On 15/05/2019, 14:16, "RW" wrote:
Also why do al
On 15/05/2019, 14:41, "Henrik K" wrote:
jarif has some flags wrong if doing it every day..
https://lists.apache.org/thread.html/ff734261cb1d8ec9dea9df42f314a60ec20c1919b8bd21c71b38553f@%3Cruleqa.spamassassin.apache.org%3E
--
Paul Stead
Senior Engineer
Zen Internet
Direct: 01706 902018
On Wed, May 15, 2019 at 02:15:19PM +0100, RW wrote:
>
> Why are there no QA statistics for BAYES_* rules?
How do you propose to generate such statistics, when all contributors
already are supposed to have fully sorted ham/spam corpuses? Seems kind of
redundant as all spam would hit BAYES_99 etc.
Why are there no QA statistics for BAYES_* rules?
Also why do all the network rule statistics come from a single
contributor labelled 'jarif'? A corpus with only 484 ham in it.
If this is genuinely what is being contributed, how is it possible to
generate all four score sets?
14 matches
Mail list logo
