gnus-message-archive-method is confusing

gnus-message-archive-method is confusing

[Kai Großjohann]

It turns out that this gnus-message-archive-method business is
confusing the heck out of some people.  Many people seem to think that
they should set this variable to their normal mail backend, but fail
in doing it right.  But it would have been so simple if they just used
fully qualified group names in gnus-message-archive-group!

Ideas?

kai
-- 
~/.signature: No such file or directory

Re: gnus-message-archive-method is confusing

[Paul Jarc]

Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:
> It turns out that this gnus-message-archive-method business is
> confusing the heck out of some people.  Many people seem to think that
> they should set this variable to their normal mail backend, but fail
> in doing it right.  But it would have been so simple if they just used
> fully qualified group names in gnus-message-archive-group!

I expect FQGNs probably would be less error-prone - you can't
accidentally create a new server that way, can you?  But then someone
ought to change the docstring for gnus-message-archive-group;
currently, it discourages use of FQGNs.

How does gnus-outgoing-message-group relate to all this?  Is it
obsoleted by g-m-a-g?  Or used for something slightly different?


paul

Re: gnus-message-archive-method is confusing

[Kai Großjohann]

On Fri, 20 Jul 2001, Paul Jarc wrote:

> How does gnus-outgoing-message-group relate to all this?  Is it
> obsoleted by g-m-a-g?  Or used for something slightly different?

If I remember correctly, I had an exchange like this with Lars:

Me: So, what's the difference between gnus-outgoing-message-group and
gnus-message-archive-group?

Lars: The former is older.

Me: So, is gnus-outgoing-message-group going to go away?

Lars: No.

I'm not sure what to make of this...  IMVHO the two ways are not
needed.  But on the other hand it seems that using
gnus-message-archive-group means that the Gcc header is not inserted
in all cases.  Hm.  Ah, yes, maybe it has to do with the gcc-self
group parameter.  Yeah, I think with gnus-outgoing-message-group and
the gcc-self parameter you get two Gcc headers, but with
gnus-message-archive-group and the gcc-self parameter you get only one
(the gcc-self one).

Can anyone confirm this?

kai
-- 
~/.signature: No such file or directory

Re: gnus-message-archive-method is confusing

[ShengHuo ZHU]

Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:

> On Fri, 20 Jul 2001, Paul Jarc wrote:
> 
>> How does gnus-outgoing-message-group relate to all this?  Is it
>> obsoleted by g-m-a-g?  Or used for something slightly different?
> 
> If I remember correctly, I had an exchange like this with Lars:
> 
> Me: So, what's the difference between gnus-outgoing-message-group and
> gnus-message-archive-group?
> 
> Lars: The former is older.
> 
> Me: So, is gnus-outgoing-message-group going to go away?
> 
> Lars: No.
> 
> I'm not sure what to make of this...  IMVHO the two ways are not
> needed.  But on the other hand it seems that using
> gnus-message-archive-group means that the Gcc header is not inserted
> in all cases.  Hm.  Ah, yes, maybe it has to do with the gcc-self
> group parameter.  Yeah, I think with gnus-outgoing-message-group and
> the gcc-self parameter you get two Gcc headers, but with
> gnus-message-archive-group and the gcc-self parameter you get only one
> (the gcc-self one).
> 
> Can anyone confirm this?

By reading the code, I think you are right.

ShengHuo