Modifying Class Object

Alf P. Steinbach alfps at start.no
Sun Feb 14 02:34:11 CET 2010


* Benjamin Kaplan:
> On Sat, Feb 13, 2010 at 7:50 PM, Alf P. Steinbach <alfps at start.no> wrote:
>> At this point consider whether it's possible to implement Pascal in Haskell.
>>
>> If it is possible, then you have a problem wrt. drawing conclusions about
>> pointers in Pascal, uh oh, they apparently can't exist.
>>
>> But if it is not possible to implement Pascal in Haskell, then Haskell must
>> be some etremely limited special-purpose language, not Turing complete  --
>>  is that acceptable to you?
>> <quote>
>>
> 
> You're actually just proving his point here. It doesn't matter what
> model Haskell uses, a version of Pascal implemented in Haskell has
> pointers.

Yes, that kills his argument.

As you note, the Haskell bit is completely irrelevant.

So his use of Haskell implementation as a "no pointers" argument is completely 
bogus, a fallacy.


> Likewise, regardless of what model the implementation of
> Python uses, Python itself doesn't have pointers, it has objects and
> names.

Well that's wrong for at least one general meaning of "pointer", but why quibble 
about terminology?

Names in Python refer to objects.

Those references can be copied via assignment.

That's (almost) all.

And it provides a very short and neat way to describe pass by sharing.


Cheers & hth.,

- Alf



More information about the Python-list mailing list