Re: Mulberry gone, now what?
- From: "Dave Cridland" <dave@xxxxxxxxxxxx>
- Date: 7 Oct 2005 15:15:11 -0700
usenet@xxxxxxxxxx wrote:
> Dave Cridland <dave@xxxxxxxxxxxx> wrote:
> > > > > Personally I think this is the fundamental flaw in IMAP, it's not
> > > > > tightly enough specified so IMAP servers vary quite a lot with the
> > > > > result that it's well nigh impossible to write an IMAP client that
> > > > > works well with them all.
> > > >
> > > > Can you point to anything in the IMAP specification that illustrates this
> > > > point?
> > > >
> > > See http://dave.cridland.net/acap/polymer.html for example where it
> > > says:-
> > >
> >
> > Cool. Mark, see that? My webpage is a more important reference than the
> > specification. ;-)
> >
> That wasn't *really* what I was suggesting! :-)
>
I didn't think so, really.
> However I think you will find quite a few other IMAP clients have a
> similar attitude and therein lies much of IMAP's problem.
>
A similar attitude to what? All I say in the paragraph quoted below is
that I haven't tested every combination of every extension or optional
behaviour because there are lots of them, and that you should always
pick a server and client which, combined, do what you want.
I don't really see how that could be construed as thinking that IMAP is
badly specified.
> > > IMAP allows a great deal of latitude in much of the specification. As
> > > a result, some behaviour is rare, and relatively untested. As always
> > > with IMAP, you should select a server and client which interoperate
> > > well, supporting the features you require.
> > >
> >
> > Latitude is definitely not intended to mean the specification is in any
> > way problematic. IMAP is very highly specified. It does, however,
> > provide various protocol features which allow server to behave
> > reasonably differently within some areas, in order to support very
> > different server-side mail spools. That's not a fault in the protocol,
> > that's a benefit. (One that has, in many respects, outlived itself). In
> > addition, there's exciting choices like multiple commands in progress,
> > unsolicited reponses at any time, etc.
> >
> ... but it *is* a fault in the protocol from the user's point of
> view. I want my IMAP client to behave the same whatever server I'm
> using, that's what makes things easy for the user.
>
Ah, but as Mark pointed out, the original aim was to make the mail
system's layout appear the same no matter what client you were using...
That, too, makes things easy for people.
NAMESPACE changes this to effectively publish the server's hierarchy
arrangement, allowing a client author to present a consistent hierarchy
style between servers, should they choose.
> I see you say that you try and make your client seem the same to the
> user regardless of the IMAP server. That's an admirable aim but it
> feels you are trying to do it in spite of IMAP rather than as a result
> of using IMAP.
No, I'm just using the base specification plus one extension
(NAMESPACE) for that.
It was very easy to do. There are much more complex bits of Polymer,
such as sending a message efficiently, for instance. The hierarchy
normalization only took about two days, perhaps three. Message assembly
and transmission took several weeks.
The ACAP server took several months, and that really did feel like I
was doing it in spite of the state of ACAP, and certainly not because
of it.
Dave.
.
- Follow-Ups:
- Re: Mulberry gone, now what?
- From: John Beardmore
- Re: Mulberry gone, now what?
- References:
- Mulberry gone, now what?
- From: Yiorgos Adamopoulos
- Re: Mulberry gone, now what?
- From: kael
- Re: Mulberry gone, now what?
- From: usenet
- Re: Mulberry gone, now what?
- From: Mark Crispin
- Re: Mulberry gone, now what?
- From: usenet
- Re: Mulberry gone, now what?
- From: Dave Cridland
- Re: Mulberry gone, now what?
- From: usenet
- Mulberry gone, now what?
- Prev by Date: Re: Mulberry gone, now what?
- Next by Date: Re: Mulberry gone, now what?
- Previous by thread: Re: Mulberry gone, now what?
- Next by thread: Re: Mulberry gone, now what?
- Index(es):
Relevant Pages
|