Hi,

> On Fri, 2 Jan 2004, Christof Drescher wrote:
> > So I could also live with these ghost messages. What about stating
> > something like "ghost messages (as defined in RFC2180) are a valid
> > response, which should force clients to NOOP to catchup".
>
> Whether or not that strategy will get you into trouble is questionable
> (there are believable reports that it will not).
>
> Returning NO got other server implementors into trouble on multiple
> occasions.

OK, I can accept it.

If I break 90% of the clients with a NO response and only 40% with a NIL
message (though violating cache semantics), I'll take the latter if you
recommend it. But then, please, make it a "valid" response by mentioning
it in RFCxyz, so the number 40 is decreasing in the future.

> > I just want clarification for the issue, be it a "NO [NOOPFIRST] Fetch
> > response" or "valid NIL message response".
>
> You got a clarification!
>
> You will not tempt fate if you either:
>  . forbid shared expunge
>  . keep the message text around until the last client that has the mailbox
>     open is notified about the expunge

Both possibilities do not solve the problem if I want an immediate expunge.

> The other strategies tempt fate.
>
> Why subject yourself to the risk, when there is such a clear answer that
> does not tempt fate?  Why spend time on this one little issue, when there
> are so many other issues in writing an IMAP server that are much harder
> and have much less clearly-defined outcomes?

Because I did not yet encounter these issues, though I'm in the last few
coding lines? Might be your RFC was so good, might be I'm a darn lucky
coder. :-)

> Don't you see that you are making a bigger issue than it actually is?  A
> client will see an externally-induced expunge sooner or later.  With most
> well-written clients, it tends to be sooner (3 minutes or less).  The real
> issue is whether or not the user gets an unpleasant error message.

Great. What do you suggest - other than ignoring the premise "expunge
immediately"?

> For better or worse, the chance of getting [NOOPFIRST] widely implemented
> is about the same as Schroeder declaring that Bush was right about Iraq,
> and that he (Schroeder) will switch parties from SPD to CDU.  It could
> happen; people of a certain belief system may think that it should happen;
> but it's not very likely!!  :-)

Wisdom, your name is Mark. ;-)

Christof

Reply via email to