Does Python really follow its philosophy of "Readability counts"?

[Steven D'Aprano]

On Thu, 22 Jan 2009 19:10:05 +0000, Mark Wooding wrote:

> Steven D'Aprano <steve at REMOVE-THIS-cybersource.com.au> writes:
> 
>> On Thu, 22 Jan 2009 15:12:31 +0100, Bruno Desthuilliers wrote:
>>> Steven D'Aprano a écrit :
>>>> But if you have free access to attributes, then *everything* is
>>>> interface.
>>>
>>> Nope.
>>
>> How could anyone fail to be convinced by an argument that detailed and
>> carefully reasoned?
> 
> Well, your claim /was/ just wrong.  But if you want to play dumb: the
> interface is what's documented as being the interface.

But you miss my point.

We're told Python doesn't have private attributes. We're told that we're 
allowed to "mess with the internals", we're *encouraged* to do so: Python 
gives you the freedom to do so, and any suggestion that freedom might be 
reduced even a tiny bit is fought passionately. When people ask how to 
implement private attributes, they're often told not to bother even using 
single-underscore names. When it is suggested that Python should become 
stricter, with enforced data hiding, the objections come thick and fast: 
people vehemently say that they like Python just the way it is, that they 
want the ability to mess with the internals.

You even argued that you disliked data structures implemented in C and 
preferred those written in Python because you have more ability to mess 
with the private attributes. In context, I had just mentioned that lists' 
internals were inaccessible from Python code. I neglected to give an 
example at the time, but a good example is the current length of the 
list. Consider the experience of Microsoft and Apple. No matter how often 
they tell people not to mess with the internals, people do it anyway, and 
always believe that their reason is a good reason.

And Python culture encourages that behaviour (albeit the consequences are 
milder: no buffer overflows or core dumps).

Add to that the culture of Open Source that encourages reading the source 
code. You don't need to buy a book called "Undocumented Tips and Tricks 
for Python" to discover the internals. You just need to read the source 
code.

And then you have at least two places in the standard library where 
_attributes are *explicitly* public:

http://bugs.python.org/issue3152

Given this permissive culture, any responsible library writer must assume 
that if he changes his so-called "private" attributes, he will break 
other people's code. In principle it could break just as much code as if 
he didn't even bother flagging them with a leading underscore, which is 
probably why many people don't even bother with _names.

In other words, if you make it easy for people to mess with your 
internals, if you have a culture that allows and even encourages them to 
mess with your internals, then you don't have internals. Everything is de 
facto public.

 
> You can tell that your claim is simply wrong by pushing it the other
> way.  If everything you have free access to is interface then all
> behaviour observable by messing with the things you have access to is
> fair game: you can rely on cmp returning one of {-1, 0, 1} on integer
> arguments, for example.
> 
> But no: the Library Reference says only that it returns a negative, zero
> or positive integer, and /that/ defines the interface.  Everything else
> is idiosyncrasy of the implementation, allowed to change at whim.

And yet people still assume that cmp returns -1, 0 or 1. Even Guido 
himself makes that mistake occasionally. Quoting from PEP 285:

"...you might be tempted to believe that cmp() also returned a 
truth value, whereas in reality it can return three different values 
(-1, 0, 1)."

http://www.python.org/dev/peps/pep-0285/

No, cmp() can return an infinite number of values. It just never does, at 
least not yet, but it might. But when Guido himself says that cmp() can 
return three values, can you blame people for acting as if cmp() can 
return three values?

Here's a thought experiment for you. You've suggested that the values 
returned by cmp() are allowed to change "at whim". Okay, let's do it: 
make a patch that changes cmp() to return -17, 0 or 53, and promise to 
support it for at least three years. Try to get it accepted on python-
dev. What do you expect they will say?

My money is on them saying "No, this will pointlessly break code for no 
good reason. Rejected."



-- 
Steven