Re: Look and Feel

[Nicolas Roard]

Le 13 févr. 05, à 09:56, Gregory John Casamento a écrit :

> > Apple spends millions of dollars getting its interface right. Even
> > Steve Jobs who manages to force a single button mouse down our throats
> > realized that an made the change.
>
> Steve bent to the whim of the Mac faithful, who were unwilling to 
> change.  Make
> no mistake this had nothing to do with any revelation on the part of 
> Steven P.
> Jobs suddenly understanding the error of his left handed scrollbar 
> ways.
>
> To quote Avie Tevanian, concerning the typical Apple user's reaction 
> to the
> NeXT IU: "It's as if we're in the land of the blind offering eyes and 
> they balk
> at the color."

Nice quote ;-)

> > Critical issues like this must be very well thought through before
> > implementation.
> >
> >
> > That aside, I must say that I am VERY dissapointed to see KDE and
> > GNOME take such a lead over a GNUSTEP based environment.
> >
> > It will be easier to sell (GNUSTEP/Objective-C) to developers if some
> > of apple's well thought out interface ideas are adopted.
>
> Me too, but I'm not sure that the reasons you cite are the reasons why 
> it
> happened. :)   The main reason GNOME and KDE "took the lead" is 
> because of
> developer stupidity, yes that's right, stupidity.
>
> Many devs don't want to learn anything new, and a lot of devs wanted a 
> C++ or a
> Java based environment because that's what was most "popular."

I *really* don't think that. Sure, some people only want one language. 
But the majority of the people that works on GNOME and KDE are far from 
being "stupid". GNUstep is *only recently* something a gui developer 
can really use. You can't blame people for choosing something that 
*works* like KDE or GNOME or XFCE or whatever, over GNUstep. Wasn't 
miguel de icaza involved a bit in gnustep at one point ?

We can regret that more people didn't participate in the project from 
the beginning;
but 1) the language wasn't well known which surely didn't help -- how 
many of us took a look to Erlang, Ada, OCAML ? a bit 
hypocrit/unrealistic to call any person that don't know ObjC "stupid" 
2) the NeXT development way wasn't well known by most of linux users -- 
hence it was very difficult to explain why it was really worth it 
instead of jumping to KDE (for example) 3) OOP in general wasn't really 
well known, and it's still a bit the case.

If I take my case, I only became interested in GNUstep 3-4 years ago 
when I saw GNUMail and realised, hey, you can really program things 
with GNUstep now. I admit I was very curious and interested by NeXT, 
and when I knew about the GNUstep project (perhaps 5-6 years ago) I 
wanted to know more. But then, well, it was quite complex to install, 
it wanted to have this display postscript thing, etc. All in all it 
appeared as something too complex to be involved in for the moment.. 
and I think a lot of people were in the same position. On the other 
hand, KDE was working quite well. And Qt is a very nice toolkit -- 
borrowing some ideas from ObjC with their signal/slots mechanism...

The main problem at the time, imho, is that even people who could agree 
that the project was sound, could have been discouraged by the sheer 
amount of work needed to complete it. KDE or GNOME on the other hand 
were already running (sortof) because they were built on top of 
existing libraries. I remember people were saying "yes, we're 
doing/using KDE, but it's temporary, in a couple of years from now 
we'll have GNUstep and that will rock". In 96-97.

There's no need to call people stupid while everything could be 
explained by the way the events unfolded... and I really don't think 
beeing arrogant will help the project to attract people (strange like 
Mac users, Smalltalkers, Lisp-ers, NeXT devs, etc., can be arrogant 
sometimes -- don't you think it's not exactly helpful for your "cause" 
? Beeing *right* doesn't help it seems).

--
Nicolas Roard
"Any sufficiently advanced technology is indistinguishable from magic."
 -Arthur C. Clarke