[r6rs-discuss] [Formal] Allow compilers to reject obvious violations

[Thomas Lord]

Per Bothner wrote:
>
> If you need a place-holder for code that you haven't written yet,
> you use something like:
>   (error "not yet implemented")
>

That's a nice strategy in *some* situations.   The strategy
Pascal suggests is, as well.



>> A language specification shouldn't make any assumptions about the 
>> software development "methodology" that I want to use.
>
> Why not?  

s/a language specification/ the specification of scheme/

and consider the intro to the spec, the history, etc.






> I think it is good if a language specification makes it
> easier to follow good practices - for whatever one believes are
> good practices, of course!
>

You're going to force me to invoke that silly rule about
usenet discussions by quoting stallman about "fascist
with a read-only mind" if you aren't careful ;-)


> Gcc, one of the larger and more complex Free Software programs
> out there, years ago switched to building with many warnings
> check for - and warnings teated as errors.  This, we've learned,
> is a good idea.
>

GCC == C?  (Roughly true but more or less to the detriment
of both and entirely for bogus outcomes of microeconomic
processes dominated by people behaving badly.)


> Anyway, I'm not sure there is much point in arguing over this
> - implementors will do what they want to.

Hmm.  Now *there's* a good perspective to reflect in a
language standard.


Bitter as I wanna be,
-t