Curious assignment behaviour

Tim Peters tim.one at home.com
Wed Oct 10 07:25:29 CEST 2001


[Dale Strickland-Clark]
> Thanks for bringing me up-to-date with this debate.
>
> I'm not sure I sympathise with the philosophy that if you can't please
> everyone then don't please anyone. An argumentative group can thus
> bring progress to a stand-still.

Take the hint and follow the link I provided:  if you don't study the
debates for yourself, you're going to reduce them to shallow
characterizations (like, say, that one) based on imagination.  If Guido has
a reputation for anything, fear of not pleasing everyone ain't it.

> For just about any operator you can come up with an example that needs
> parenthesise to behave in the 'natural' way rather than the way forced
> by precedence.

Sure.  It's a bad sign when examples aren't strained or contrived, though.

> Assignment in expressions, particularly in conditional statements can
> reduce complexity and simplify legibility.

While a stiff dose of poison can be good for some illnesses -- absolutes
aren't particularly illuminating.

> You can always write code that's difficult to understand and I would
> argue that the recently added list comprehension syntax can be a real
> pig to decypher but I wouldn't be without it now.

Sure and sure, but "difficult to understand" wasn't the issue so much as
"surprising in ordinary cases" and even "not helpful enough often enough to
be worth the costs".

> Not introducing a feature because it might make some code more
> difficult to understand is on a par with not letting adults buy knives
> in case they cut themselves.

It was more akin to not letting adults piss on the toaster because it might
short out and cause a fire <wink>.

opposing-arguments-that-were-actually-made-would-probably-be-more-
    interesting-ly y'rs  - tim





More information about the Python-list mailing list