Re: GNUS vs. NNTP?

[Greg A. Woods]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> "Greg A. Woods" <woods@robohack.org> writes:
>
>> My guess is that those 55 articles were cancelled and expunged before I
>> tried to read them, but why did GNUS have to download what seems like
>> all of the headers for the entire set of old articles first, just to
>> then tell me there was no new news?
>
> It shouldn't have to.

It doesn't seem to always do it, but sometimes it's even worse.

I visited another group last night with 14 unread articles in it.  I had
not ever visited it before (after initially catching up all articles a
few hours previously), and doing so triggered a download of well over
200mb before causing emacs to bail from memory exhaustion.

Trying again in trn took a few moments, maybe 20 seconds, and showed me
the 14 articles.

> Have you done any Gnus customisations?

I have a rather extensive .emacs but it only sets up mail related stuff.

My .gnus file is here, in its entirety.  I'm guessing it's the setting
of gnus-fetch-old-headers, but I really didn't expect "some" to be
equivalent to "all" when threads seem to have unknown references.

I guess I'll try changing it to 20 or so and see if that improves things. 

;;; -*-lisp-*-

(setq gnus-fetch-old-headers 'some)

; add the description!
(setq gnus-group-line-format "%M%S%p%P%5y: %(%-45g%) %D\n")

;;; MIME (mm) stuff

(eval-after-load "mm-decode"
  '(progn
     (add-to-list 'mm-discouraged-alternatives "text/html")
     (add-to-list 'mm-discouraged-alternatives "text/richtext")))

(setq mm-automatic-display (remove "text/html" mm-automatic-display))

(setq mm-text-html-renderer 'links)     ; does this use browse-url?

-- 
                                                Greg A. Woods
                                                RoboHack <woods@robohack.ca>