[Kroupware] Web interface frontend
Andreas Jellinghaus
kroupware@mail.kde.org
30 Sep 2002 09:59:47 +1200
the server development is 100% closed.
This is what i'm saying. no matter how many
mails i write, you dont leak any details,
and you do not participate in any discussions
beside marketing statements.
> The concept paper is a "living" document and will be subject to permanent
> change!
berhard stated quite clearly, there is not much room for changes.
the document got only a markup in more than two weeks, that is not living.
beside the docbook source is not availabel?
you want me to lookup the dilbert concerning "living documents"?
about the termi site:
its a very well defined term used by anyone in the field of distributed
computing. the default scenario is this: a company has a head office and
a few local offices, wants to allow the local offices to access
functions like email, but does not want to have that traffic on the slow
network connection between the offices.
the default solution is to deploy mail servers in that local offices,
but with central administration, and an easy interface to decide where
which mailbox is located. or even provide automatic/heuristic mail
routing to make the mailboxes available where there are needed.
combined with a client that does not to be reconfigured (e.g. for people
visiting different localtions).
site stuff is not easy. quite obvious you cannot replicate all mailboxes
to all sites. also people do not want a central mailserver, that will be
unavailable if the network connection fails. also traveling people
accessing their data from different sites causes headaches.
and then talk to someone who knows replicated databases and the problems
of two databases without a link, both getting changes, and then trying
to merge. or N databases. exchange is a includes database.
or consider the complexity, if one server is not big/fast/... enough
to hold all your mails, but you need to merge several servers, and the
last thing you want is to reconfigure every user to different settings.
also you dont want to manualy configure what user will be on what
server. exchange does that: join several servers with a more or less
"do not care" approach.
if you ever wondered why exchange is difficult to use: it has advances
features most people do not need, and thus a lot of unnecessary
complexity. but when you need these features, there are few
alternatives.
if your statement to implement everything exchange does, then there
is a lot of stuff to look at and learn.
oh, even while it is completely unrelated your statement
"we believe to have found a nice solution." is quite typical for
the reason i complain: a vague statement, not sharing any details
with everyone else. what have you found out? why not share your
findings with us. this years slogan at the linuxtag was:
open your heart, open your mind, open your source!
andreas