Re: various independent "mail servers" and the nnml backend

[Uwe Brauer <uwe_brauer@mat.ucm.es>]

>>>>> "Ted" == Ted Zlatanov <tzz@lifelogs.com> writes:

    Ted> On Mon, 18 Jul 2005, uwe_brauer@mat.ucm.es wrote:
    >>>>>>> "Ted" == Ted Zlatanov <tzz@lifelogs.com> writes:
    >> 

    Ted> They shouldn't.  Try the `a' `e' method to add the servers instead of
    Ted> gnus-secondary-select-methods.  Unfortunately I don't have this set up
    Ted> myself, so I can't vouch for the correctness of Gnus with two nnml
    Ted> servers, but if you run into further problems I'll take a look to make
    Ted> sure the underlying code is doing the right thing.

    Ted> You can also do `a' from the Server buffer, add an nnml server, then
    Ted> hit `e' to edit the parameters and add the nnml-directory.
    >> 
I just deleted all my .newsrc newsrc.eld files and restarted gnus, I
add a nnml server, 
so the corresponding buffer looks like
(nnml "NNML")   

However  independent whether I edit the server in order to obtain something
like this

(nnml "NNML-Archive" (nnml-directory "/home/oub/Mail/NNML-Archive"))

When I enter the nnml server he shows me groups which do not exist. So
for some reason gnus reads some old information which resulted from my
experiments. But what.

Another question: it seems that I can use the secondary select method
in order to select a server, but I could also to it in the way
described above, even for nnimap or nntp servers. Which method is
better?

Thanks 

Uwe