[Kde-pim] Raindrop and Akonadi: Scope for collaboration

[David Ascher]

On 10/26/09 3:50 PM, Stephen Kelly wrote:
> Hi,
>
> Your announcement last week about the Raindrop Project caught my attention for
> its scope and for its timing.
>    

Sorry for the delay in answering.
> I am a developer on the Akonadi project - the KDE4 platform for next
> generation PIM applications.
>
> http://pim.kde.org/akonadi/
>
> (Link for the Akonadi people: http://labs.mozilla.com/raindrop )
>
> Akonadi is a platform independent PIM framework designed to be accessed with
> any/multiple APIs and to cache and share data between PIM applications. As it
> happens, I was recently working on a proof of concept for getting PIM data
> into a web browser in the form of a web application.
>
> http://steveire.wordpress.com/2009/10/24/akonadi-goes-web2-0/
>    

congrats on the announcement!
> As Akonadi and Raindrop face the same challenges and attempt to solve similar
> problems, I thought it would be a good idea to investigate scope for
> collaboration, or at least to tell you that we exist :).
>    

Indeed!
> * Completing the python-akonadi API and porting Raindrop to use Akonadi.
> This would allow the maximum potential for sharing of data and collaborating,
> and the best user experience on platforms with mixed/multiple apis and
> applications. Of course I'm dreaming here, but if Thunderbird uses Akonadi
> too, there would be easy mixing/mashing between all KDE PIM applications,
> Raindrop and Thunderbird.
>    

At this point I don't think _porting_ of Raindrop to anything is 
likely.  I'm not sure what Akonadi's long-term goals are, but we're 
trying to make Raindrop as flexible to deploy as possible.  Note that in 
particular we _don't_ want to build a universal PIM, which does seem to 
be Akonadi's goal.

> * Writing a bridge for transferring data between Akonadi and Raindrop. There
> is already work in progress on a CouchDB resource for Akonadi, allowing any
> Akonadi using application to access and manipulate data from a couch.
> This would potentially bring the same benefits as above, but with added
> potential difficulties such as conflicts and no reduction of code duplication
> (imap, twitter implementation etc).
>    

Right, someone could write what we call a "protocol" to get data to and 
from Akonadi into and out of Raindrop. We've focused on those for 
messages, but we will someday figure out how Raindrop wants to talk to 
sources of "contact" data as well.

> * Sharing information on UX such as conflict resolution dialogs, managing
> multiple resources for the same type of data and how much of that should be
> abstracted away, and technical issues such as how to deal with broken IMAP
> servers, concurrency and security.
> This one obviously doesn't involve any code sharing and has the least direct
> benefit for the user.
>
>    

That seems like a very diverse set of topics, some of which could have 
shared code, and some of which not.  I'm not sure what to say about it 
as a result =).

> * I also notice that Mozilla is part of the nepomuk project.

Uh, I have no idea what nepomuk is, or how Mozilla is a part of it. Do 
you have details?

--david

_______________________________________________
KDE PIM mailing list kde-pim at kde.org
https://mail.kde.org/mailman/listinfo/kde-pim
KDE PIM home page at http://pim.kde.org/