Wheewww....

The project looks wonderful....

to bad it uses all that STUFF I am not currently up on....;)

The graphics interface of course looks quite powerful depending on how
well it is designed and implimented. The problem being that most such
interfaces lack a proper Metaphorical Construct, to make the intuitiveness
of the interface practical to use.

It's a bit like Apple versus DOS in the old days,
the Mac came out with an intuitive interface and walked all over the early
DOS machines until "WINDOZE" came out.

Most HDL implimentations have attempted to work from the "Language" to the
interface, rather than attempting to define a Metaphor, and designing the
language to impliment the metaphor. The result, has usually been a tool
that was hardly an improvement on the original language, especially since
flashing up WINDOZE utilizes a high proportion of processing time just for
overhead.

While linux is a cheap operating system, and many people who are anti-MS
are using it. Planning a Graphics Interface should not be taken as a
trivial task, even if there are languages such as VB and VC that
trivialize the windows interface.

Just my $0.02 (CDN) worth

                                GREY

GRAEME SMITH                         email: grysmith@freenet.edmonton.ab.ca
YMCA Edmonton

Address has changed with little warning!
(I moved across the hall! :) )

Email will remain constant... at least for now.


On Sun, 16 May 1999, Jamil J. Weatherbee wrote:

[snip stuff about web page]... Good idea Just trying to cut bandwidth
>
> My proposed tool would:
>
> 1) Be Free, GNU copyleft or the like
> 2) Start with a simple "compiler" written with GCC, running under
> dos/windows/unix, but developed under linux/bsd
> 3) Expand with a graphical interface, maybe tcl/tk maybe java to describe
> the tasks
>
> 4) implement both procedural type code, and state machines
> 5) the user would simply define the processor type and speed, although I
> would probably start with the x14 instruction set pic, like 12c671 etc.
> because that would keep the cost of trying it out/testing very inexpensive
>
> 6) The user would specify the frequency that tasks run and timing etc.
> just in seconds/microseconds and the compiler deals with different
> clockrates.
>
> 7) ideally i see a software where all this could be easily described
> graphically, but in a very powerful fashion, almost "like" and HDL for
> PICs.
>
> If you are interested in such a project to benefit the entire PIC
> community please email me: jamil@scottibox.com
>