Re: slow SMTP send blocks emacs

[Richard Lewis <richardlewis@fastmail.co.uk>]

At Sat, 20 Mar 2010 14:30:39 -0400,
Per B. Sederberg wrote:
> 
> At Sat, 20 Mar 2010 14:22:56 -0400,
> Matt Price wrote:
> > 
> > Hi folks,
> > 
> >  the SMTP server for my main account is often quite slow to respond to
> >  send requests, and wl seems to sort of hang, waiting until the
> >  transaction is complete.  With another mua this would be fine -- I
> >  would just go over to emacs and work while the message was sending.
> >  However, wl is *inside* emacs -- so I'm unable to use my main work
> >  tool during the delay time.  
> > 
> > So I suppose the question is:  does anyone else experience this?  And
> > is there a way around it?  Thanks as always,
> 
> What do you mean by slow?  It usually takes my SMTP server (gmail)
> about 3 seconds to send a message, which is annoying, but not so bad
> that I've gone searching for an alternate solution.  Nor have I tried
> other SMTP servers.  Nonetheless, I would welcome a way to make it
> more speedy :)  
> 
> So, to the list, what do you deem as an acceptable amount of time to
> send a message and what do you actually experience with wl?
> 
I use two SMTP servers: one university one which I can only access
from inside the university network. This is always very fast with
wl. I also use another authenticated one which came with an imap email
account I use. This one tends to take a second or two to send
messages.

However, I do occasionally experience the very slow sending messages
problem. At home I use ADSL with a painfully slow uplink. So when I
send large messages (e.g. over 100K or so), I experience an
appreciable delay during which my Emacs is unusable.

One day, in the distant future, someone will make a decision about how
Emacs lisp with have threading support added. And then all the
thousands of package maintainers will gradually start altering their
code to take advantage of this multi-threading capability.

But until this happens, I guess our only solution is to use a mail
sending technique which relies on an external process, rather than
pure elisp. Not sure what's available on that front?
-- 
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Richard Lewis
ISMS, Computing
Goldsmiths, University of London
Tel: +44 (0)20 7078 5134
Skype: richardjlewis
JID: ironchicken@jabber.earth.li
http://www.richard-lewis.me.uk/
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
+----------------------------------------------+
| Support open access to scholarship           |
| http://freeculture.org/ http://www.doaj.org/ |
+----------------------------------------------+