Re: Goodbye for now, kmail

On Sat, May 06, 2017 at 12:38:37PM +0200, Martin Steigerwald wrote:
> Anders Lund - 06.05.17, 12:19:
> > P?? Sat, 06 May 2017 12:07:22 +0200
> > 
> > Daniel Vr??til <[email protected]> skrev:
> > > You have Trojita and Sink, both support IMAP and don't use Akonadi.
> > > There's no way to add IMAP to KMail without using Akonadi.
> > 
> > No? I can't imagine this is due to technical reasons, so it must be
> > religious. So a fork may be the way to go?
> 
> Anders, it *is* due to technical reasons.
> 
> Current KMail only speaks to Akonadi. I go as far as saying: Without major 
> rework it is locked to it. So if you want to add IMAP without Akonadi to KMail 
> you also need to do the integration with KMail, which I believe would be quite 
> invasive change.
> 
> Either you would completely move back to pre-akonadi times of KMail which is a 
> no-go for current KMail developers or you??d have to maintain *two* completely 
> different IMAP access methods within KMail, including an Akonadi and a non-
> Akonadi access method. I don??t see how this would be maintainable by the 
> current team.

So we are in a situation where KMail speak only to akonadi, which in
turn does not work well at the only thing he need to offer to kmail,
manage the messages. 

At this point, a pre-akonadi kmail or something not tied to akonadi
are not a really appalling ideas, if you ask me.

Technical reasons aside, the big problem here is that the KDEPIM
developers have not learned anything from the KDE4 initial release
fiasco, so they repeated it: release basically an alpha version as
"ready for the end users", which in my experience is way far from the
thruth.

And as aggravating they made it a one way road, given that as today
kmail has not a way to export all the mails in a simple way (to move
from kmail to claws I needed to pass throug mutt to be able to import
the mailboxes).


 
> And again the most important question:
> 
> Who would do the fork? You? If not, whomelse? I don??t see *anyone* 
> volunteering. Do you?

Obviously nobody try a fork. For once, kmail is so heavily tied to
akonadi that a full rewrite is probabily simpler, and people
just use something else, albeit not integrated with KDE, that just work.


bye
Gianluca