Re: Compiler writers will love this language

vbdis@aol.com (VBDis) wrote in message news:03-06-018...
> , ericmuttta@email.com (Eric) schreibt:
>
> >However seeing that I am a polite and considerate language designer, I
> >am going to consult you and seek your advice on certain aspects of X,
> >so as to make your life easier when it comes to implementation.
>
> What kind of language do you have in mind? If it is a general programming
> language, what does it make different from already existing languages?


I am thinking of a general purpose, multi-paradigm programming
language. Your second question "what makes it different from existing
languages?" is an interesting one, but unfortunately it cannot be
answered fully within this space because


1) there is loads I can say and
2) I dunno if the comp.compilers folks would agree that this is the
right place to give the details.


If I was to say one thing about it, I think it would be that it gives
greater accessibility to power - power to clearly and quickly express
solutions to problems that often need to be solved in pairs.


> What's the difference between "designing" a language, and writing
> down its grammar?


Hmm. To me, "designing" a language, is thinking about problems that
need to be solved and then crafting a tool (ie the language) that is
suited to solving those problems. Writing down the grammar is indeed
part of the design (because you have to ask "how do I want the syntax
to look like?") but I would class that as being more about "language
specification" ie (formally) describing the design on paper.


>
> IMO the designer of a language should study existing languages first,
> and reUSE existing designs, before trying to reINVENT the wheel.


That's solid advice and indeed, I believe we get new ("good")
languages because someone studied existing languages and found them to
come up short. I think for two possibly similar languages X and Y,
there is a general sentence that says "Y was created because X
did/didn't do A, B and C".


Unfortunately a lot of the time language designers adopt existing
designs without questioning them and we have age old design mistakes
continually propagating into new languages. You will tend to find that
if a language designer created a language Y, and they were previously
experienced with language X, many of the faults in Y can be traced
back to X. One of the biggest things robbing us of great new advances
in language design is the fact that we use language X for so long,
that all its faults become transparent to us and propagate to the new
languages we later design.


For instance, take C and C++. Backward compatability is an important
goal and sometimes inevitable evil, but why was is it *so* important
when it came to C++?


Admittedly, that's where C++'s greatest strengths and weeknesses lie,
but I have always wondered what the world would be like if Mr.
Stroustrup had loosened up a little bit on the backward compatability
goal :)


Cheers, E.
[Language design is on-topic insofar as it affects compiler design and
implementations. But don't argume about where the semicolon goes.
-John]

Return to the comp.compilers page.
Search the comp.compilers archives again.