Re: What is the roadmap?

["John Stowers" <john stowers gmail com>]

On Feb 16, 2008 3:09 AM, Brent Gueth <creeva gmail com> wrote:
> I'm curious for personal sake what is the roadmap for services to be added?
>
> Do you maintain an internal list that isn't published beyond the bug
> requests for features /service additions?

Not explicitly. We did have a TODO on the wiki, but I deleted it and
moved everything to bugzilla as enhancement requests. The basic medium
term goals, at least in my head, are
* I will be releasing a 0.4.0 stable-ish relase around March, after
GNOME comes out. At that point I will probbably branch, There are no
worrysome bugs currently reported against Conduit that I forsee
preventing this
* I will then start switching us over to using GIO for all File I/O. I
expect this to break things
* Around  March I will also propose Conduit for inclusion into GNOME
* To raise the chances of us getting included in GNOME I will try and
encourage existing, and new contributors, help me integrate Conduit
into other parts of the desktop. The big bullseye target here is to
replace Tomboy sync with us, but there is also scope to integrate
Conduit into nautilus, and eye of GNOME a lot better
 * I will also be applying for summer of code to add opensync/syncml
support to Conduit
 * I will be going to GUADEC, I hope to be speaking about Conduit

The big uncertainty is the GNOME proposal part. It could go well, or
it could go really badly (remember beagle v. tracker)

> If I or someone finds a service that supports a python API and we have
> access to Python API code would this be a good starting place for
> members to go to submit ideas?  Would it be better to start a bit more
> from scratch?  I'm just curious on how those submitting requests (like
> the spurl guy the other day) can help speed up the process if they can
> go and do some of the leg work for the programmers from the beginning.

You are correct. It is very helpful if the bug reporter can do basic
things like finding out
1) If an API is available
2) If there already exists python bindings to it

I am very willing to add support for dataproviders which use an
external library to do the hard work because then there is less
maintenace responsibilty on us. If the service is super popular then
it may be worth us writing our own bindings for it (like thomas did
for box.net,picasa and facebook, JC2k started doing for opensync and
ipod, I did for evolution and am now doing for syncml).

In terms of new dataproviders contributors, there is some low hanging fruit
* There are already excellent python bindings to amazon S3 [1]
* Likewise for Remember the Milk [2]

Anyway, thats about all I can think of ATM. I would be interested to
hear what other things Conduit developers are working on. For example
I know that Conduit is working very hard on getting Conduit running on
maemo/nokia n800

Regards.

John

[1] http://bugzilla.gnome.org/show_bug.cgi?id=510110
[2] http://bugzilla.gnome.org/show_bug.cgi?id=510108
>
> Take you time in responding - this being more of a pondering e-mail
> then anything of importance.
> _______________________________________________
> Conduit-list mailing list
> Conduit-list gnome org
> http://mail.gnome.org/mailman/listinfo/conduit-list
>