On Tue, 01 Dec 2009 15:50:58 -0500 "Mark T. B. Carroll" <[email protected]> wrote:
>> That's a feature, I'm not sure why it should be changed. Why is it >> harmful? MTBC> I copy and paste fragments of articles in checking them out in MTBC> other applications, taking notes, etc. But then the selection has MTBC> happened so when I get to replying, it just quotes a little bit of MTBC> the article, so I have to undo the selection first, whether by MTBC> going to some other article then back to that one, or using M-u to MTBC> mark it unread, quit the group altogether, reenter it and go back MTBC> to the article, or whatever. Maybe there's an easier way to get MTBC> rid of the selection, but a high fraction of my selecting has MTBC> nothing to do with what I might want to quote in reply, so having MTBC> to undo it just causes me extra work for no gain. On Wed, 02 Dec 2009 08:58:06 -0500 "Mark T. B. Carroll" <[email protected]> wrote: MTBC> Anders Wirzenius <[email protected]> writes: >> Try C-g before you reply. MTBC> Ah, cool, that works better and makes it, from my point of view, a MTBC> smaller bug! (-: I'm getting the impression that there's no variable MTBC> I've missed to just turn it off, then. MTBC> (Some other kind soul suggested to me by e-mail that `S W' to reply MTBC> might help, at least until this feature infects that too.) Sorry, I thought you knew how to clear the selection. I really don't think it needs to be turned off or that it's a bug from anyone's point of view. If you feel strongly about it, perhaps you or someone else can produce a patch to disable the feature on demand. It should be fairly easy. >> I think to correct that, the child process needs to disassociate itself >> from Emacs as its parent and that's pretty complicated (requires C calls >> that are not available in ELisp AFAIK). MTBC> Oh, that's a bit unfortunate! What happens is it starts up a web MTBC> browser, say, then I get reminded of or linked to other things I want to MTBC> check out, and in the same session after a while I'm now looking at MTBC> things that are nothing to do with what I originally launched from MTBC> emacs. This is especially bad with things like web browers, Acrobat MTBC> Reader, etc., where, when I view different things from outside emacs, MTBC> those attempts to view things attach to the existing emacs-spawned MTBC> process instead of spawning a new viewer, and then they want to die with MTBC> an application that never had anything to do with them. Yeah, that's annoying. You can do the disassociation in a shell/Perl/etc. script. Also there are programs to launch the URL or file in the OS's default URL handler, e.g. in Ubuntu there's a way to do it. When you launch the URL or file that way, the handler won't be associated with Emacs. With temporary files from a message attachment, however, this won't work because they get deleted after the external viewer is done. I personally use and prefer w3m so I don't experience this issue. Ted _______________________________________________ info-gnus-english mailing list [email protected] http://lists.gnu.org/mailman/listinfo/info-gnus-english
