Without the key information of what you're thinking of developing and
what the target environments are any advice will be vague and
approximate at best because it is always a case of horses for courses.
That said I can say definitively from the free software point of view
not Java or anything targeting Gnome's Mono engine. They're both
proprietary originated systems and have a range of patents on them
which might bite the free software community later on. This is also
the FSF advice.
If you're setting about writing GUI applications then the general
wisdom at present is you should use a language with automatic memory
management - handcrafted memory management does less well on larger
projects and makes for additional coding and debugging time.
So combining this with Qt as you've mentioned for RAD you might try
Python, KDE's qdevelop with its python plug-ins, Qtdesigner etc and
later in development profile the program for performance hot spots
and recode them in C/C++. But the best advice for performance has
always been to look at your algorithms first, poor ones are what eat
the CPU cycles usually.
Now for some opinions on related matters which others might find
contentious.
Free software and open source software has a lot of apps which are
developed for one widget set or another. The various desktop
environments have each a smaller range of apps targeting them. I
suspect this is because people fall into the 'popularity trap' RMS
mentioned and want their app to run in a wide range of desktop
environments including proprietary ones. IMO free desktop
environments are in need of a full range of apps specific for them so
that we have the hope of a future 'killer app' bringing the cause new
users. So choose a DE.
Further I believe free software developers should deliberately ignore
users of non-free systems. If users of such system want a specific
free software app, let them port a fork of it themselves - the free
software community should not spend time on it. The idea that we
have enough free software developers is a myth, it was known in the
late 80s that the demand for software development exceeded the world
economy's ability to pay for it and the situation is worse now[1].
So IMO free software developers should not waste their time, which is
in a sense a community resource, on pandering to people who RMS
justifiably calls fools.
Combining the above two with RMS's GNU 30th speech point of 'people
are picking low hanging fruit.' Don't do the 'me too' thing of
writing yet another media player etc. It's mere vanity. I'm not
saying you have to write a large super app. But e.g. go look in a
computer store at some of those apps that cost $50 or under. Things
like a GUI outliner (which there are a selection of for the Mac), or
a clone of 'Write Your Own Novel (Professional)' are a much better
proposition. They add to the variety of free software. Remember FSF
canon allows you to use non-free software in order to code a
replacement.
Similarly free software is especially in need of more games.
Lastly, a word on the supposed 'wisdom' that developers should use a
bleeding edge distro like Parabola GNU/Linux. I find it very hard
not to indulge the British habit of swearing at such nonsense. I
have over 30 years in software and am under various NDAs etc so I
can't give details but I have seen sums with lots of zeros go down
the tubes from this practice. Developers must have a very stable
development machine. Use a chroot/LXC container/Parabola under KVM
or whatever for the target environment, but keep your actual
development desktop O/S stable, the less change the better. So set
up you development environment on Trisquel, hand compile any
additional libraries you need and set up some virtualised targets.
You'll need to for the other GNU FSDG distros once you've released
your app anyway.
[1] If you think about this - there not being enough money to pay for
the software we need you will realise that the open source
community's purported reasons for success are vacuous. The reality
it is because the code can be shared / modified etc as guaranteed by
the four freedoms they are riding on an economic incentive to make
software for a lot less money. They may have a lot of users but they
are fooling themselves if they think there's any other reason than
money for them being there. The free software community OTOH has
grown and is about an ethical commitment, the fact that there will
always be free riders doesn't detract from our achievements, we've
never pretended to ourselves we have some quality magic or anything
like that.
Leny
