At 23:59 11/08/2010, Josh Davis wrote:
Not sure what kind of impact we're talking about here. Currently,
there's no scalar type hinting and there will never be a consensus
around strict XOR weak. Having an implementation that allows both
while reusing a familiar syntax (parentheses as a way typecast) looks
like the best (and perhaps only) way to move on.

Josh,

No disrespect, but I can't explain why it's bad to have both implementations any better than I already did. If you don't see it, chances are that no matter how much I try to explain - it won't help - we probably see things too differently for us to ever agree on it. Let's end it by saying that a great deal of people here think it's horrible to introduce strict typing to PHP period. Whether we also add anything else at the same time doesn't change that one bit.

Moving forward with both is certainly not the only option, I'd say (given the paragraph above) that it's not an option at all. At the very least, there's one other option which is doing nothing. And that's assuming we can't reach widespread consensus that auto-converting type hints are bad.

The opposition to strict typing was that it's 'evil', and we mustn't introduce it into PHP. The opposition to auto-converting type hints was that strict typing is supposedly better.

Now that strict typing is pretty clearly off the table - how would those that supported it vote between doing nothing at all and 'settling' for auto-converting type hints? That's the real question on the table now.

For the record, I'm fine with both options, although personally I'd go for the latter.

Zeev

--
PHP Internals - PHP Runtime Development Mailing List
To unsubscribe, visit: http://www.php.net/unsub.php

Reply via email to