At 22:26 30/07/2004, Edin Kadribasic wrote:

On Friday, Jul 30, 2004, at 21:16 Europe/Copenhagen, Zeev Suraski wrote:

My 2c - we shouldn't discuss the implementation in the same context as whether we should have it, because it moves the focus away from the source of the problem, which is not implementation. We're way beyond the times when we add something to PHP just because we can... I'm not saying we shouldn't discuss it, just not in the same thread.

I pretty much fail to see the reason to discuss this with you as you have more or less already vetoed it. No amount of arguments on how useful this tool is in certain situation is is going to convince you, isn't it? It is after all you who has such delicate sense of what are the needs of "the community".

I think it's a shame you resort to sarcasm instead of actually responding the points I (and others) are making. I went over the thread and most of what I could find was references to other places with examples that show why GOTO is so necessary, but couldn't actually find too many of these places.


I think it would help if you (or anybody else) tried to present the case for GOTO in a clear way, giving a set of situations where you believe that GOTO is necessary and cannot be resolved in a different, equivalently-good way, while answering the issues that I (and others) have raised, instead of discounting them as irrelevant. WTF factor and likelihood of abuse ARE factors, so far the only response from the pro-GOTO camp is that we should just ignore these factors, which is IMHO not an acceptable approach. In my opinion the pro-GOTO camp doesn't have real answers to these concerns, which is exactly why I believe that GOTO is not a good idea for PHP. If you (or any of the other pro-GOTO people) cannot address these concerns in a convincing way, then you're right - there's no point discussing it, at least not beyond the context of an academic implementation discussion.

And please, stop the 'dictator!' accusations. If whatever came out of my mouth was immediately implemented in PHP, we'd have libxml2 bundled and SOAP support built-in (just two examples, there are plenty more, and that's just fine).

Zeev

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



Reply via email to