[Python-Dev] PEP 318 needs a rewrite

Mon Mar 8 13:53:32 EST 2004

Instead of arguing in circles, can someone (or a small group) rewrite
the PEP to contain a summary of the arguments for and against various
forms?

I've been asked by a few folks to pronounce on the PEP, but as it
stands, the PEP doesn't adequately summarize the discussion, so I
can't pronounce.

I haven't been able to follow the discussion here in much detail, but
I've sampled bits and pieces, and I happen to like the Quixote form

  def <name> [<attributes>] (<arguments>): <body>

As a human reader, I like seeing the attributes before the arguments
because some attributes (like classmethod and staticmethod) affect the
interpretation of the argument list.  Also, I suspect that long
argument lists are much more common than long lists of attributes; I
don't like finding attributes hiding after the argument list.

I don't like having the attributes before the name, because that makes
it harder to find the name.  So from this perspective the Quixote form
is optimal IMO.

(Feel free to rewrite the PEP to amtch this opinion. :-)

--Guido van Rossum (home page: http://www.python.org/~guido/)