Gnus development mailing list
 help / color / mirror / Atom feed
* Re: Better defaults for adaptive scores together with decays.
       [not found] <87achfhej1.fsf@futtelifut.dyndns.org>
@ 2005-10-14 13:16 ` Reiner Steib
  2005-10-14 15:18   ` Martin Geisler
  0 siblings, 1 reply; 2+ messages in thread
From: Reiner Steib @ 2005-10-14 13:16 UTC (permalink / raw)
  Cc: Ding List

On Wed, Oct 12 2005, Martin Geisler wrote:

> Gnus v5.11
> GNU Emacs 22.0.50.1 (i486-pc-linux-gnu, X toolkit, Xaw3d scroll bars, multi-tty)
[...]
> When using adaptive scoring together with decaying of scores, authors
> are never really adaptively scored: when reading an article the from
> address gets +3, but the default decay function snaps scores between
> -3 and +3 to zero... so unless you happen to read several articles by
> the same author on a given day, then the scores will always stay at
> zero.
>
> Increasing the default scores (to 10 or something like that) seems to
> work for me.

How about making it depend on `gnus-score-decay-constant'?  But maybe
that's overkill and we should only mention it in the doc string.
Opinions?

(defcustom gnus-default-adaptive-score-alist
  `((gnus-kill-file-mark)
    (gnus-unread-mark)
    (gnus-read-mark
     (from , (+ 2 gnus-score-decay-constant))
     (subject , (+ 27 gnus-score-decay-constant)))
    (gnus-catchup-mark
     (subject , (+ -7 (* -1 gnus-score-decay-constant))))
    (gnus-killed-mark
     (from , (- -1 gnus-score-decay-constant))
     (subject , (+ -17 (* -1 gnus-score-decay-constant))))
    (gnus-del-mark
     (from , (- -1 gnus-score-decay-constant))
     (subject , (+ -12 (* -1 gnus-score-decay-constant)))))
  "*Alist of marks and scores.
If you use score decays, you might want to set values higher than
`gnus-score-decay-constant'."
  ;; [...]
  )

Result with the default value of `gnus-score-decay-constant' (3):

ELISP> gnus-default-adaptive-score-alist
((gnus-kill-file-mark)
 (gnus-unread-mark)
 (gnus-read-mark
  (from 5)
  (subject 30))
 (gnus-catchup-mark
  (subject -10))
 (gnus-killed-mark
  (from -4)
  (subject -20))
 (gnus-del-mark
  (from -4)
  (subject -15)))

Current default value:

((gnus-kill-file-mark)
 (gnus-unread-mark)
 (gnus-read-mark
  (from 3)
  (subject 30))
 (gnus-catchup-mark
  (subject -10))
 (gnus-killed-mark
  (from -1)
  (subject -20))
 (gnus-del-mark
  (from -2)
  (subject -15)))

Bye, Reiner.
-- 
       ,,,
      (o o)
---ooO-(_)-Ooo---  |  PGP key available  |  http://rsteib.home.pages.de/



^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: Better defaults for adaptive scores together with decays.
  2005-10-14 13:16 ` Better defaults for adaptive scores together with decays Reiner Steib
@ 2005-10-14 15:18   ` Martin Geisler
  0 siblings, 0 replies; 2+ messages in thread
From: Martin Geisler @ 2005-10-14 15:18 UTC (permalink / raw)


[-- Attachment #1: Type: text/plain, Size: 1243 bytes --]

Reiner Steib <reinersteib+from-uce@imap.cc> writes:

> On Wed, Oct 12 2005, Martin Geisler wrote:
>
>> Gnus v5.11
>> GNU Emacs 22.0.50.1 (i486-pc-linux-gnu, X toolkit, Xaw3d scroll bars, multi-tty)
> [...]
>> When using adaptive scoring together with decaying of scores, authors
>> are never really adaptively scored: when reading an article the from
>> address gets +3, but the default decay function snaps scores between
>> -3 and +3 to zero... so unless you happen to read several articles by
>> the same author on a given day, then the scores will always stay at
>> zero.
>
> How about making it depend on `gnus-score-decay-constant'?  But
> maybe that's overkill and we should only mention it in the doc
> string.  Opinions?

I think your suggestion is very good!  The important thing is that
people see the wanted effect right "out of the box" --- afterwards
they can adjust it as always.

Thanks for looking into this!

-- 
Martin Geisler                                     GnuPG Key: 0x7E45DD38

PHP Exif Library      |  PHP Weather             |  PHP Shell
http://pel.sf.net/    |  http://phpweather.net/  |  http://mgeisler.net/
Read/write Exif data  |  Show current weather    |  A shell in a browser

[-- Attachment #2: Type: application/pgp-signature, Size: 188 bytes --]

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2005-10-14 15:18 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
     [not found] <87achfhej1.fsf@futtelifut.dyndns.org>
2005-10-14 13:16 ` Better defaults for adaptive scores together with decays Reiner Steib
2005-10-14 15:18   ` Martin Geisler

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).