PEP 285: Adding a bool type

[Raymond Hettinger]

"Guido van Rossum" <guido at python.org> wrote in message
news:mailman.1017466974.6751.python-list at python.org...
> I offer the following PEP for review by the community.  If it receives
> a favorable response, it will be implemented in Python 2.3.

Bono's six hats:

Python author hat:   Votes -1 because all the books which have a section, a
quick intro to python, will have to add another couple of paragraphs
introducing the new type and differentiating it from ints.  The extra
paragraphs clutter the concept space.

Python helper/tutor hat:  Votes -1 because it will need a FAQ entry to
explain the rationale and provide examples of how to interchange it with int
types.  More than one question will come across the help desk.

Python YAGNI hat:  Votes -1 because scans of previously my written code do
not show any need for the bool.

Python vs other languages hat:  Votes -1 because this concept would make
more sense in strongly typed languages.  In contrast, Python offers easy
substitutability (dictionaries where shelves were used; stringio where files
were used, etc.) -- the many faces of truth and falsehood that now exist
should be kept.

Python accountant:  Where is the income stream?  What can be done after the
capital expenditure that couldn't be done before?

Python gambler:  Sure, why not, how many backward incompatibilities and
unexpected consequences could there be?  With the potential payoff so high,
you have to go for it!  <wink>


Raymond Hettinger, CPA




? Would implicit 'return Nones' at the end of function definitions be
interpreted as False
? Would the 'and' and 'or' operators now require boolean arguments