why ; ?

Yigal Chripun yigal100 at gmail.com
Tue May 13 14:31:10 PDT 2008


BCS wrote:
> Reply to Yigal,
> 
>> personally I don't understand why we still use text files to represent
>> code.
>>
> 
> I's to long a story to post here, but my day job is working on a new
> programming language. For several years it was a non text based
> language. Suffice to say, it just couldn't be made to work.
> 
> 
I'll be glad to here that sometime when there's time, if you care to
tell it.
anyway, that doesn't mean anything. this just shows that your particular
 language couldn't be made to work by your company. that's not a proof
in any way.
the little theorem of Fermat was calculated by super computers for a
huge set of numbers, but until it was formally proved it wasn't formally
true. the story of the light bulb shows also how despite hundreds of
failures Edison continued his work until he did manage to create a
light-bulb.
on the other hand, as I said in the original post, there are already
available programming languages that are not text based. even those math
(latex) languages are included in this. You can even include UML in this
if you use a weaker definition, since there are (many) tools that
generate code from UML diagrams.
GUI is a prime candidate for a visual language. most people already use
tools like eclipse's visual editor to draw the GUI and have the IDE
generate the code for them. I don't think there are any Java programmers
left (outside the educational system) that actually hand code their GUI
code. Eclipse does a better job of it and the development is faster.
this is just evolution taking it's course.
once upon a time humans used assembly. I think the last assembly
language designed to be used by humans was that of the PDF-11. today the
compiler generates it for you. you still can use assembly directly of
course, but the current design is driven by the needs of compiler
writers and the assembly language is no longer built to accommodate mere
humans.

http://en.wikipedia.org/wiki/Visual_programming_language




More information about the Digitalmars-d mailing list