Re: Revisiting the Gnome Bindings
- From: Jamie McCracken <jamiemcc blueyonder co uk>
- To: Danilo Šegan <danilo gnome org>
- Cc: Thomas Vander Stichele <thomas apestaart org>, Havoc Pennington <hp redhat com>, sri aracnet com, gnome-desktop-devel <desktop-devel-list gnome org>
- Subject: Re: Revisiting the Gnome Bindings
- Date: Sun, 26 Sep 2004 11:33:51 +0100
On Sun, 2004-09-26 at 10:56 +0200, Danilo Šegan wrote:
> Jamie,
>
> Today at 2:21, Jamie McCracken wrote:
>
> > 1) Easy to program in and has naturally clear syntax and simple
> > structure. (that rules out C++)
>
> If that rules out C++, IMO it rules out both Java and C# as well
> (we're talking about syntax here, right?). Though, Java would have
> an edge here, because of not-requiring "virtual" keyword (or was it
> vice-versa?) — that's a big one for me when it comes to "natural
> clearness of syntax" and "simplicity of structure" ;-)
The problem with c++ is its complexity and you can program stuff in it
in a zillion different ways (most of them insane!). If we are to have an
alternative to C it should have a more simple structure and be more
suitable for beginners like Java/C#/Object Pascal (Freepascal/Delphi/
Kylix) Although Object Pascal would win here because its syntax is
amongst the clearest the point is merely that in an ideal world it would
be both clear syntax and simple structure - C++ is out because it has
neither. Java has an advantage too because most universities now teach
it instead of C.
>
> I'm not saying that all of these suck and don't fulfill this
> criteria, but rather that Java, C# fulfill it as much as C++ does
> (with neglectible error margin), and that it depends on *personal*
> views of what you consider clear syntax (for me, anything apart from
> assembly and Python is far away from it, though C is close because of
> it's resemblance to HL assembly :-).
>
> > 2) Has a damn good IDE with integrated form designer so its child play
> > for developers to use.
>
> This rules everything else basically out, except for maybe
> Delphi/Kylix which doesn't use Gtk+ ;-)
No Cause FreePascal and Lazarus is an open source clone of Delphi/Kylix
which has GTK bindings (http://www.lazarus.freepascal.org/). Sure it
would need updating and some work to make it suitable but thats just one
option.
> I haven't looked at
> MonoDevelop, but I doubt it provides useful-enough "integrated form
> designer" for Gtk#.
I believe MonoDevelop will incorporate an integrated Form designer at
some point - I dont know about Eclipse?
The real point here is that having a RAD language without a good IDE is
worthless. When choosing a language the quality of its IDE will be
critical.
>
> > 5) Is a proven RAD language. (going with the D language might be too
> > risky?)
>
> Uhm, nothing left. Unless we're thinking of really adopting Kylix :-)
Kylix is bad in my opinion - its a pale shadow of Delphi (which is the
best RAD tool ever developed IMO). However I see the options as : Java/
Mono/Object Pascal (Lazarus) assuming they all satisfy the IDE
requirements in the near future.
>
> > I also think time is against us here and waiting another year or two is
> > bad - we need to have consensus ASAP so that the next major version of
> > Gnome (3) can take advantage of RAD tools long before Longhorn rears its
> > ugly head.
>
> Consensus can't be gotten that easily, because it requires (almost)
> half of the community stepping down. And as with anything in Free
> Software, it comes down to mostly emotional issues, so for anything
> to happen, it will just have to happen.
>
> Also note that any "decision" taken by "core" Gnome project is not
> set in stone: other developers will still use whatever bindings they
> see fit, so I still expect more apps to appear in something like
> pygtk, than eg. java-gnome, even if we "dedicate" ourselves to it.
>
> We're not time-critical as you're trying to describe, since we can
> see more and more apps popping up without us ever prescribing any
> single bindings as the best one. It will certainly take much longer
> than "another year or two" for Gnome to switch it's basic
> infrastructure to anything other than C.
That would be unfortunate cause building a runtime for whatever RAD
language we choose is gonna take a lot of time and effort so the sooner
we get started the better. Writing GUIs in C sux and is a waste of
developer time - we could achieve so much more with RAD. The language
chosen does not really matter so long as it provides an easy way for
more people to contribute and accelerates development. Its really sad
when we have to spend so much developer time on tedious C coding
especially in GUIs.
jamie.
>
> Just my 2€/100 ;-)
>
> Cheers,
> Danilo
>
>
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]