Re: gnus makes emacs lose response

[Richard Stallman]

      When you redial, the computer attached to the
    modem bank has no clue who are (and might even be a different computer
    than before), so you usually get a different IP address, making TCP
    connections established with the old IP address unusable.  Either way,
    OpenSSL probably isn't even aware of all this commotion in the lower
    layers of the networking stack.

That being so, if you're using WiFi then OpenSSL should not die when
the connection drops, but if you're using PPP, then OpenSSL should
die when the connection drops.

Emacs is the wrong place to make the decision.  Emacs is not in a
position to know whether the connection was by phone or by wireless.
If Gnus could impose a timeout, it would fix the problem for phone
connections and cause a spurious problem for wireless.

How can we make the right thing happen in both cases?  Some part of
the system has to be responsible for making the right thing happen in
each of these cases.  Somehow OpenSSL needs to be able to determine
that a low-level connection is irrevocably broken, so it can die.
This may require changes in Linux and GNU Libc as well as OpenSSL,
but I don't know.

We need to report this to the people who can fix it right.

In the mean time, the solution for Emacs is to make sure C-g works.