Gianni Mariani said:
Text based languages are so a thing of the past.
I strongly disagree. And of course, if that was true, there would not be any
anymore.
I have a vision that in the future, code will be developed using a
more graphical and collaborative model. "Source" control will be
managed using different views and releases are done by combining
views. The source propagates and merges automatically. Lifetime
management is implicit since the compiler implements the appropriate
form because it has full knowledge of the code and how to reap
objects.
I'm working on it now.
That's a great idea. It sounds revolutionary. Good luck with that. I, OTOH,
think that there is nothing wrong with a text-based programming and that it
is even most desireable for a number of reasons (which I won't get into much
here and now). I agree that the evolution of the languages has proceded so
slowly that they have to change drastically or become relics. I too think
that the best way to show the power of text-based programming languages is
to create a new one because "all the current ones suck so bad"
. Seriously
though, they are tied to past (as you say) but a new language is enabling to
new software. C++ should probably go into "maintenance-only mode" and much
of the resources should go toward a new language. (Oh, wait... perhaps
"designed by committee" language is exactly what I don't want another of!).
I too spend most of my "programming time" drawing boxes, cirles and arrows
etc., but I don't see it going from there directly into machine language for
the same reason I don't like programming to "big black boxes": it becomes to
proprietary (it's like working for someone else: ewwww!). A "modern"
language that enables that paradigm as an intermediate form is still the
core if you ask me: pick your poison, graphical development or hand coding
directly with the language.
John
