2013/4/29 Martin Owens <doctormo@...400...>:
Over all I think the dbus signal work would need to be picked up and completed properly. At least there's a good starting point for anyone who wants to get involved.
I was thinking about this recently and I think the best way to provide the relevant functionality is to simply embed a Python interpreter in Inkscape and provide bindings through Boost.Python, Swig, or some other method. In keeping with the 'worse is better' philosphy, this approach has a chance to actually end up being usefully implemented in a reasonable amount of time.
The only desirable feature of the DBus-based solution is that you can in principle use any language to implement an extension that uses the DBus API. At the same time it has many disadvantages: poor compatibility with Windows, poor prospects for future cross-platform compatibility (there are plans to move DBus into the kernel), higher implementation complexity and much higher overhead. Even allowing arbitrary languages is not a clear-cut advantage: if someone wrote a great extension in a language that no-one on the core team knows and proposed it for inclusion into Inkscape, we would have a choice between rejecting useful functionality and alienating a potential contributor or accepting something we can't maintain.
Just my 2 cents, not looking to start a holy war.
Regards, Krzysztof