Re: "Mission Statement"
| From: | Sascha Schumann | Date: | Tue, 21 Nov 2000 00:04:47 +0000 |
| Subject: | Re: "Mission Statement" | ||
| References: | 1 | Groups: | php.lang |
| Request: | Send a blank email to php-lang+get-36@lists.php.net to get a copy of this message | ||
On Tue, 21 Nov 2000, Zeev Suraski wrote:
> At 01:51 21/11/2000, Sascha Schumann wrote:
> >On Tue, 21 Nov 2000, Zeev Suraski wrote:
> >
> > > I'd say that it exists to develop the language specification for PHP
> > > period. I don't see how portability or vendor independence have anything
> > > to do with it...
> >
> > Here is a short explanation for both terms, in the context of
> > last week's netmask thread (on zend-cvs).
> >
> > Portability:
> >
> > The ``hack'' relies on the host platform's long being
> > 32-bit which makes it non-portable. This particular
> > feature and similar ones should not be part of the
> > specification.
>
> So I'm not sure I agree about the portability part. Reality is much more
> important than perfection.
Would you want to standardize a feature which works on
Linux/ia32, but not on Linux/Alpha?
That is reality, as far as I am concerned, and something
which we should not simply put aside.
> There could be many things that were born as side effects of
> implementation, but if they're here to stay, there's no reason to make them
> a part of the language. It should be possible to implement just about
> anything under a different implementation as well. At any rate, I guess
> we'll have to see about this one as time goes by.
Well, we will have to evaluate the impact of a particular
feature and come to a decision on whether it is a valid
behaviour which we want to continue or a mere side-effect.
Most things in PHP are well understood, but there are some
areas (like reference passing) which are neither well-
understood (by the users) nor well-defined.
- Sascha