Well recently i looked at imap's rfcs. Add, add, add ... lets add more
features.. is not excuse for not getting the design right in first place.
Imap is good, but spoiled by the users... nowdays developers get lost
fixing/adding/modifying the interactive part of the application, forgeting
about the core.
FWIW sometime ago I did something very very basic to what you want to do,
i hacked my way through using sqlrelay, adding some basic auth mechanism and
extra table and 3d db api... pop3/imap servers were going through it too..
well i realised the easiest thing that could be done was to use something in
the middle.. to be honest speed was improved.. more granularity .. therefor
more options.. share anything you want the way you want.. basically
unlimited.
cheers
p.s. what happens when kswapd in linux dies?
Eelco van Beek - IC&S writes:
Yeah i was thinking of adding some commands to POP (for selecting,
deleting and adding folders). This could fit between the 'extended'
protocol capabilities of POP. I'm going to think about it some more :)
To reply to all :
The big problem with imap is that the protocol is totally grown out of
shape. The irony is that only 20% of the functionality of imap as a
commandset is being used by 80% of the mailclients. The Imap server is
doing a lot stuff that isn't even nescessary for most clients and that
takes time. That why i wanted to create a native client that does nothing
more than required using fast, dbmail-structrured queries. I think speed
improvement over imap would be more than 250%.
Best regards,
Eelco
On 5-jan-04, at 20:51, Micah wrote:
I don't see how it's possible without writing a server daemon that has
access
to the database. Otherwise the client will have access to everyone's
mail.
Once you start doing that, you're almost reinventing the wheel, so why
bother?
I would bet if you gave it some thought you could make a server that was
faster than pop or imap, but is it worth it?
What about including some dbmail specific commands for the existing pop
server
that would take advantage of capabilities of the system. Your client
could
use them, but the overall server would still be compatible with the pop
standard (or IMAP for that matter..)
my half a cent..
-Micah
On Mon January 05 2004 11:25 am, Eelco van Beek - IC&S wrote:
Hi All,
I want to create a native dbmail client. Not a webbased client but a
full e-mail application for both windows and os X. My main problem is
authentication. Since i won't be using pop or imap i don't really know
how to handle authentication. The MySQL authentication meganism won't
suffice since it only authenticates on Tables, not on rows..
I've been thinking about this for quite a long time but i don't see a
real solution (except to create an intermediate language. The problem
with that is that some new wheels are needed to be re-invented..).
Anybody?
Best regards,
Eelco
_________________________
E.J.A. van Beek
ICT Manager
IC&S
T: +31 30 6355730
F: +31 30 6355731
PGP-key:
www.ic-s.nl/keys/eelco.txt
_______________________________________________
Dbmail mailing list
Dbmail@dbmail.org
https://mailman.fastxs.nl/mailman/listinfo/dbmail
_______________________________________________
Dbmail mailing list
Dbmail@dbmail.org
https://mailman.fastxs.nl/mailman/listinfo/dbmail
_________________________
E.J.A. van Beek
ICT Manager
IC&S
T: +31 30 6355730
F: +31 30 6355731
PGP-key:
www.ic-s.nl/keys/eelco.txt
_______________________________________________
Dbmail mailing list
Dbmail@dbmail.org
https://mailman.fastxs.nl/mailman/listinfo/dbmail
