The most recent episode of the always excellent
LugRadio features an interview with
Miguel. Mainly they talk about
Mono, though there is a small bit
in there on Ximian’s history.
The interesting point that one of the LugRadio guys brings up is
this: what if, instead of paying 25 people to work full-time on Mono,
Novell instead payed them to implement
f-spot,
Beagle(which is
awesome), and iFolder in C,
Python or whatever?
No doubt these projects could be further along, but would their growth
rate be higher? Implementation is probably just as easy with Python as
with C#, so its possible that there could be just as much momentum
behind a Python implementation of Beagle as there is behind the C#
Beagle.
Desktop applications are not a reason to get behind Mono. in the end,
the user is going to see the same (gtk+ ;) ) interface. The benefit in
Mono is for the developer. Try this: open your favorite package
management software (I’m crazy, I still use dselect), and search for
dbus. What do
you get? dbus, dbus-qt, dbus-python, dbus-glib, dbus-cli. The first
four give you APIs for C, C++ using QT, Python, and C using glib. Four
packages for four languages (three if you want to be technical), 4
different APIs to learn. The last package, dbus-cli, lets you access
dbus from any language implemented on the CLI, and the interface will
always be the same, after you mangle it a bit for your chosen
language.