perm filename PASCAL.HAK[P,JRA] blob sn#128587 filedate 1974-11-07 generic text, type C, neo UTF8
COMMENT ⊗   VALID 00002 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	pascal barfs
C00005 ENDMK
C⊗;
pascal barfs

pedestrian
 alternative to
  some
   computational or
    arithemetical
     languages


1. compiler based

2. debugging  ≡  %$#&%-ing dumps!!!!   come on!!!


4. integer labels?...nicely nmeumonic
     and why the fuck declare them.

5. where are co-rountines, etc..."modern language" features
     or do we have pascal(i) i→∞. if proper runtime control is not
     built in, then there will be serious difficulties in extension!!

6. ree 5: procedure-valued vars??? who knows
     yes, finally(p 73-74); but unclear what is done about global variables.
     also not clear if we can return functions...most likely not.


8. what is semantics of and[x,y]??? i.e. is it call by val. or l-r as lisp
    ah,ha, p 18. implementation dependent.neat!!

8a.   "true or F(x) is undefined it F has side effects"..what if F does not??
         determination of side-effects is very,very difficult!!!

9. writeln example on p15(user's man): how does it know about formatting?

10. "the most fundamental of statements is the assignment ..." bullshit

11. scalar types are a good idea.

12. packed arrays are a bad idea; this kind of crap is implementation!!

13. pointers suck (looking ahead)

14. records are ok. not to crazy about syntax, but ...
     humm..assignment to components w.o. copying.

15. initial thought: no provision for user i/o conventions on records
     as part of record definition. gota have it!

16. how useful are set vars...dunno

17. ho,ho "pointers are simple tools for the construction of .. data structures."
    and a gun is a simple tool for trimming your fingernails.