From kde-core-devel  Wed Apr 05 12:28:26 2006
From: Carsten Pfeiffer <carpdjih () mailbox ! tu-berlin ! de>
Date: Wed, 05 Apr 2006 12:28:26 +0000
To: kde-core-devel
Subject: Re: Solid and Akonadi? (was: Re: Discontinuing kcmwifi for
Message-Id: <200604051428.27096.carpdjih () mailbox ! tu-berlin ! de>
X-MARC-Message: https://marc.info/?l=kde-core-devel&m=114424019122728

On Monday 03 April 2006 14:22, Will Stephenson wrote:

> The Akonadi daemon acts as a local cache; it's always available to Kontact
> etc.  The parts that talk to remote servers will be separate processes
> again, so these need to be suspend- or cleanly killable in the case of
> impending network disconnection.  The centre of Solid/network status
> integration will therefore be in the scheduler, this will handle status
> changes from Solid and user requests to 'Work On/Offline'.

I'm afraid simply killing or suspending processes that deal with remote 
resources might not be sufficient. There needs to be a local persistent 
storage facility as well.

Imagine you work with documents, that are associated with a remote resource 
and your network is down (not uncommon, when e.g. working in a train). Now 
your battery runs out and you need to shut down your laptop. You want to have 
all your data stored locally until you next have a network connection again. 
Then you want to be asked whether your data should be uploaded to the remote 
resource.

Also see https://bugs.kde.org/show_bug.cgi?id=85897

Cheers,
Carsten Pfeiffer