home.social

#programminglanguagedesign — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #programminglanguagedesign, aggregated by home.social.

  1. Let me add several notes (but no spoilers):

    One part of the exercise is designing an approach to process sequences generically that can work with the Loop Facility.
    Dave Tenny's post upthread describes one possibility.

    Another part is in working out how to extend the Loop Facility itself so one can say

    (loop for x <being the elements of> some-sequence ...)

    where coming up with a suitable keyword for the placeholder is a subpart of the exercise.

    Yet another part is thinking about the trade-offs involved in introducing such a subclause.
    This includes understanding how hardware and compiler technologies have changed in about half a century since the Loop Facility was first designed, so reasonable trade-offs have also changed.

    And a metanote repeating the well-known: the value of such exercises is in _doing_ them, not in obtaining the final result.

    #CommonLisp
    #ComputerProgramming
    #Lisp
    #ProgrammingLanguageDesign

    @davetenny

  2. Let me add several notes (but no spoilers):

    One part of the exercise is designing an approach to process sequences generically that can work with the Loop Facility.
    Dave Tenny's post upthread describes one possibility.

    Another part is in working out how to extend the Loop Facility itself so one can say

    (loop for x <being the elements of> some-sequence ...)

    where coming up with a suitable keyword for the placeholder is a subpart of the exercise.

    Yet another part is thinking about the trade-offs involved in introducing such a subclause.
    This includes understanding how hardware and compiler technologies have changed in about half a century since the Loop Facility was first designed, so reasonable trade-offs have also changed.

    And a metanote repeating the well-known: the value of such exercises is in _doing_ them, not in obtaining the final result.

    #CommonLisp
    #ComputerProgramming
    #Lisp
    #ProgrammingLanguageDesign

    @davetenny

  3. Let me add several notes (but no spoilers):

    One part of the exercise is designing an approach to process sequences generically that can work with the Loop Facility.
    Dave Tenny's post upthread describes one possibility.

    Another part is in working out how to extend the Loop Facility itself so one can say

    (loop for x <being the elements of> some-sequence ...)

    where coming up with a suitable keyword for the placeholder is a subpart of the exercise.

    Yet another part is thinking about the trade-offs involved in introducing such a subclause.
    This includes understanding how hardware and compiler technologies have changed in about half a century since the Loop Facility was first designed, so reasonable trade-offs have also changed.

    And a metanote repeating the well-known: the value of such exercises is in _doing_ them, not in obtaining the final result.

    #CommonLisp
    #ComputerProgramming
    #Lisp
    #ProgrammingLanguageDesign

    @davetenny

  4. Let me add several notes (but no spoilers):

    One part of the exercise is designing an approach to process sequences generically that can work with the Loop Facility.
    Dave Tenny's post upthread describes one possibility.

    Another part is in working out how to extend the Loop Facility itself so one can say

    (loop for x <being the elements of> some-sequence ...)

    where coming up with a suitable keyword for the placeholder is a subpart of the exercise.

    Yet another part is thinking about the trade-offs involved in introducing such a subclause.
    This includes understanding how hardware and compiler technologies have changed in about half a century since the Loop Facility was first designed, so reasonable trade-offs have also changed.

    And a metanote repeating the well-known: the value of such exercises is in _doing_ them, not in obtaining the final result.

    #CommonLisp
    #ComputerProgramming
    #Lisp
    #ProgrammingLanguageDesign

    @davetenny

  5. Let me add several notes (but no spoilers):

    One part of the exercise is designing an approach to process sequences generically that can work with the Loop Facility.
    Dave Tenny's post upthread describes one possibility.

    Another part is in working out how to extend the Loop Facility itself so one can say

    (loop for x <being the elements of> some-sequence ...)

    where coming up with a suitable keyword for the placeholder is a subpart of the exercise.

    Yet another part is thinking about the trade-offs involved in introducing such a subclause.
    This includes understanding how hardware and compiler technologies have changed in about half a century since the Loop Facility was first designed, so reasonable trade-offs have also changed.

    And a metanote repeating the well-known: the value of such exercises is in _doing_ them, not in obtaining the final result.

    #CommonLisp
    #ComputerProgramming
    #Lisp
    #ProgrammingLanguageDesign

    @davetenny

  6. [Can't divide integers with `/' in Dylan (apparently).]

    Right.

    Myself, I see the language design logic here, but I also see that this particular design trade-off would have been best avoided.
    In my opinion, in this context, it would have been slightly better (not _good_, just _better_) if the language didn't have a `/' operator at all.
    If one wants to force the programmer to be very specific about an operation, one wouldn't use a well-known operator that is already widely used with a more general semantics.
    Cultural compatibility matters.

    #ComputerProgramming
    #ProgrammingLanguageDesign
    #ProgrammingLanguages

    @mdhughes

  7. [Can't divide integers with `/' in Dylan (apparently).]

    Right.

    Myself, I see the language design logic here, but I also see that this particular design trade-off would have been best avoided.
    In my opinion, in this context, it would have been slightly better (not _good_, just _better_) if the language didn't have a `/' operator at all.
    If one wants to force the programmer to be very specific about an operation, one wouldn't use a well-known operator that is already widely used with a more general semantics.
    Cultural compatibility matters.

    #ComputerProgramming
    #ProgrammingLanguageDesign
    #ProgrammingLanguages

    @mdhughes