What I want for PHP 6 Christmas...

[Stormrider]

I agree with Tony here (Again, good to hear from you about this, being a 'student' of many of your articles and your website!) - PHP doesn't need this radical change. To be honest, the function naming is inconsistent, yes, but it really doesn't bother me that much. Having consistent naming doesn't make things easier to learn, you still have to learn names of things, and I have no trouble learning the function names as they are .

[paul_wilkins]

I for one welcome our namespace overloads.

Do you know how many times I've had to look up which way to use needle / haystack?

[losangelescarren]

Nice for google like. Because google prefer html page than php page now

[Jake Arkinstall]

There is no difference to search engines, lostangelescarren, that's a rumour. Besides, you could fool a search engine into thinking your page is html by using .htaccess redirects, but that isn't necessary.

@Tony, I understand the point you're attempting to get across, unfortunately the utter garbage surrounding your points somewhat lose you a lot of credibility. Maybe if you improve your attitude your arguments will be taken seriously (to those of us who aren't readers of your articles).

You evidently don't find the lack of consistency in the PHP function names annoying - sure, with experience you learn the differences - but the point is that, for a language that is meant to be easy for beginners, it certainly leads to a lot of manual-lookup for beginners. I'm not saying 'it should be like...', I'm saying it should be improved.

However, PHP mentality is currently about retaining backwards compatibility, so function renaming is certainly out of reach and isn't very practical for current PHP programmers. However, a fork-off of PHP with a more user-friendly library would be useful for those who just want something different. In fact, the current library of functions is something that is used to discredit PHP the most, in anti-PHP arguments. I'm not talking about changing the current language, because there are always people who, like you, like things the way they are.

There is no reason why a language can't "adapt". I have experience in PHP, .Net, C++ (the C library included) and Java - I can certainly say PHP is the messiest. Most people argue back saying 'its the best for its job' etc - however, that's IN SPITE of the mess behind it, not because of it. Removing old inconsistencies would be of certain benefit for those who do find the library annoying.

Besides, if languages didn't borrow things from others then OOP would be dead and Java, PHP, C++ and C# wouldn't exist. I'm not saying I want to turn PHP into C++ or Java, but I'm saying that PHP could do with a library comparable to them. If you find that ridiculous, then I hope you aren't a part of the development of the PHP language.

But seriously, please die down on your trigger-happy insults. This is a friendly forum; keep it that way, yeah?

[pgb101]

However, PHP mentality is currently about retaining backwards compatibility, so function renaming is certainly out of reach and isn't very practical for current PHP programmers. However, a fork-off of PHP with a more user-friendly library would be useful for those who just want something different. In fact, the current library of functions is something that is used to discredit PHP the most, in anti-PHP arguments. I'm not talking about changing the current language, because there are always people who, like you, like things the way they are.

Forking causes more political issues. PHP is installed a lot of the time to handle what could be said are low quality code bases by hosting companies. There is still divide in the PHP community whether OOP has any place, even though I would say most( if my job shouts are to go by all ) now require it. Things have changed a lot in the last few years and in London good OOP, database and PHP skills are quite hard to find. This would probably leave most hosting companies to settle for the version that leaves it as it is. It will be hard to move to 6 as it is. Every hack solution posted on a forum that becomes the accepted answer could possibly be one step back, especially if it gets into an application that is widespread. This site is good as like PHPBuilder, there is a level of quality policing. Cuddles are only worth so much.

Besides, if languages didn't borrow things from others then OOP would be dead and Java, PHP, C++ and C# wouldn't exist. I'm not saying I want to turn PHP into C++ or Java, but I'm saying that PHP could do with a library comparable to them. If you find that ridiculous, then I hope you aren't a part of the development of the PHP language.

I agree, languages evolve to fix problems that are re-occuring/tiring to handle in their domain. Often there are common problems that are language independant so need to be solved in a common fashion, so if there is an elegant solution already out there why not use it. The arguments for and against are already there if looked for.

Exceptions are good because it allows forced deferred decision that can then be handled at the appropriate level. Some should shut down all action, others should just be logged due to the functionality they handle is really just ice on the cake( ie banner handling ). For a decent argument on exceptions there is always the checked exceptions versus the unchecked exceptions. Error codes are too easily ignored, Exceptions are generally more readable as they have a distinct handling method different from the rest of the code and shout for attention. Empty catches really shout for attention. Though, like interfaces, using them well is not necessarily easy.

With any handed solution it is good to step outside of personal sanctity and try and imagine the world it was created in and why. If you are against it create an argument for, if you are for create one against. Development is rarely 100% right in approach all the time, that is what makes it fun. There is always improvement to be made, to yourself and to others around you which in turn improves the code base.

Coding for yourself is also very different than coding with 20 people. Between monkey see monkey do and group laziness quality does slip the more people added. The bar of group acceptibility gets lowered as there would be potentially war or people giving up. There is also a risk of process by commitee and that can get pretty tough, eloquence is not a requirement of good code writing.

I don't like blanket rules but some people do as they are the mentally easy way. In the short term at least.

Anyway it is quite old but this may have some value.
PHP6 notes
http://www.php.net/~derick/meeting-notes.html

Personally I want the type hinted return types.

[Tony Marston]

I just object to demands that PHP should be "just like language X" simply because some poor programmers do not have the mental capacity to deal with the fact that PHP is different from language X ...

Enough is enough. Such downtalking doesn't prove you're smart - it just proves you're very insecure about your intelligence and feel the need to try to prop it up by insulting others.

Why is that an insult? Surely the fact that some programmers can't handle the differences between different programming languages points to a lack of ability on their part, so I regard that as fair comment.

[Force Flow]

Why is that an insult? Surely the fact that some programmers can't handle the differences between different programming languages points to a lack of ability on their part, so I regard that as fair comment.

That's not always the case. Sometimes different languages do have features that would be useful to include in a language that doesn't have them. Wanting a useful feature in a language doesn't necessary mean a mental deficiency on the part of the developer.

[Michael Morris]

Please don't quote Tony - I put him on ignore so I wouldn't have to read his inflammatory crap.

As for forking PHP - that would take a lot of effort, effort better placed in creating a server side implementation of EMCA Script 5 / Javascript 2. Such an implementation would have the huge advantage of being able to share some code blocks between the server side and client side code (for example, blocks of code that perform field validation).

[pgb101]

That's not always the case. Sometimes different languages do have features that would be useful to include in a language that doesn't have them. Wanting a useful feature in a language doesn't necessary mean a mental deficiency on the part of the developer.

You have to remember Flagellation is the penance of all worthy code monkeys