[NNTP] LIST EXTENSIONS (again)

[Russ Allbery]

Mark Crispin <MRC at CAC.Washington.EDU> writes:

> I did not do this out of a desire to be gothic.  I did this because it
> was the only thing that worked with all the servers that real users
> encountered.  There are, indeed, users who could not use NNTP with Pine
> until they added the /user=xxx to force authentication before MODE
> READER.

> It sounds to me as if you telling me that I should do MODE READER first,
> and tell my users who set /user=xxx to force AUTHINFO first "sorry, the
> IETF NNTPEXT working group says this is broken, tell your server admin
> to fix his server."

> This ain't gonna happen.

I think that, for your software, what makes the most sense is to just
document that Pine doesn't work with any server that requires MODE READER
unless that server implements the new LIST EXTENSIONS protocol that we're
hammering out now (at which point you can use that to decide whether you
have to do MODE READER).

The point, however, that several of us have been trying to make is that
you could currently simply remove MODE READER from Pine entirely and it
would be functionally equivalent to what you're doing right now for any
server that requires either TLS or AUTHINFO.

-- 
Russ Allbery (rra at stanford.edu)             <http://www.eyrie.org/~eagle/>