PHP's hope for the future?

[dagfinn]

A lot of people, have made a lot of money (hard cash) from PHP, and thus in my view, people should have more respect. Is that asking too much huh?

What exactly is so disrespectful? I don't see disrespect, not in this forum at least. I see critcism, sometimes harsh, and I see the same thing from you and a few others in the opposite direction. Some have come dangerously close to portraying me and others as lazy whiners, but I can live with that.

It seems to me there is demand for unilateral and exaggerated "respect" which is part of the mindset that is perpetuating these problems. Mutal respect and giving everyone the benefit of the doubt is what we need.

[DougBTX]

What exactly is so disrespectful?

I think calling people "cowboy developers" is disrespectful, I've heard that a couple of times. Whether true or not, you can't say it is showing respect. Quite a few of the comments seem to be with loosing respect with people due to their attitudes.

Douglas

[Dr Livingston]

If none of what is said in these forums influences your programming style or habits, what do you come here for?

Of course these forums influence on how I develop, just like everyone else surely? That is why we come here, to take note on what others are doing, how they're doing it, etc. But my commenting wasn't really in regards to develop as such, more to PHP, which is what this discussion is about

That is very shortsigthed to be honest.

Maybe. Yes, your proberly correct but that is how I feel about the situation. People make a living out of PHP which therefore they should at the very least respect that much like anyother language etc that there are going to be problems.

What a lot of people forget is that moving from PHP4 to PHP5 did require a completely new solution. That solution for the most part is pretty stable, but what isn't stable (as I can tell) is basically backward compatability.

PHP5 had to support PHP4.x, and let's be honest here for a moment, with PHP5.x supporting a completely re-write, that was never going to be easy to support PHP4 as well.

The way some people are going about sabre rattling, you'd think that PHP was a mistake, or the community (one individual, or the collective group) are about to dump all over PHP.

If they dump all over PHP, then so what? PHP will still continue, the world will still be here tomorrow. In one form or another

I just don't know what all the fuss is about, what with what is actually happening in our world today. Anyways, hope I'm a bit more clearer this time around

If your best friend started doing heavy hardcore drugs, or committing horrible crimes, would you be "prepared to defend him"?

Nope.

Or would you do what's best for him and confront him about the wrong in his ways?

Of course you would at least try to help, but at the end of the day they can only help themselves.

[jayboots]

Isn't it lovely that anyone who has a free membership to a popular forum or a personal blog can stir up any hell they want about a community? The vitriol on internals being created by users who claim to be in the "top 10%" is both counter-productive and makes them (the ranting users) look really bad. It is one thing to ask for a new process and to present your needs in a reasoned manner with an explanation of why you have those needs. It is quite another to start a personal rant. Nor is having your proposals turned down or ignored a reason to start a rant.

If you are truly professionals--particularly in the enterprise--you don't run into the problems that people are crying like babies over. You have process of your own and you have testing and staging and you don't rely on your upstream to do the right thing, especially on upgrades. In other words, your apps don't break on upgrade because it never gets to that point. You people who are complaining about a FOSS project not meeting your needs and saying that that proves it is not professional or enterprise ready -- take a better look at your own practices and improve them. Then realize that crying and ranting will get you nothing.

Enterprises are businesses, they don't work on feeling. They work on fact and rationale. If facts don't match the rationale and the process can't handle the inputs, then you have decisions to make. You can *work with* your inputs to try to harmonize them with your process, you can change your process or failing all of that you can pick new inputs; however, you *do not* assume that you can dictate what the inputs will be. Especially, ESPECIALLY, when the input is a FOSS project made up mainly of volounteers! These are not your employees! Show respect!

[DougBTX]

The vitriol on internals being created by users who claim to be in the "top 10%" is both counter-productive and makes them (the ranting users) look really bad.

Spot on. I didn't recognise the name of the person who posted that, though I did see the SP link.

Douglas

[jplush76]

as someone who had to recently upgrade a 250K line application from php4 to php5 I know the headaches of all the breakages in backwards compatibility from the two versions and to see a point release cause this same headache is really dissapointing, ESPECIALLY with all the PHP6 talk regarding "lets clean house and make it non-backwards compatible again with PHP5" which is a hard sell in the enterprise world.

I'd be willing to bet there are still a small number of users using PHP5 even at this time. I now fear upgrading. If a vulnerability is reported I feel more comfortable changing code that uses that vuln instead of upgrading php versions. Its becoming a ****ing joke.

[Dr Livingston]

To bring PHP6 into the matter is not really the point, as it's not going to happen any time soon, but in saying that, with all the trouble people seam to be having, maybe PHP6 will do a better job of it.

I think the biggest mistake made with PHP5 was to support PHP4 in the first place, in my view.

Maybe in the future when you report a bug, it'll not be denounced as bogus huh? But to me the denouncement is all academic really, the point is that if you come across a bug, that you make the effort to report it, then it's in some form or another, documented.

That is more important, that other developers are made aware of it. The fact that nothing is done about it, is not here nor there. There are many limitations in our busy lifes, and time is one of them.

If you find a workaround to mask over a bug, then great, use the workaround with that PHP version. If at some point you upgrade, the bug may not even exist, but you should be, as a developer, prepared to script new code to work with an upgrade anyways.

Maybe the world has gone mad or something, I don't know?

[jayboots]

For the same reason, many people still use GCC 3.3 and 3.4 rather than 4.x. This also applies to other languages and applications developed in a certain version of that language. For crying out loud, your job is maintenance of your application. You likely get paid to do it. You should fear upgrading and morseso, you should plan your migration path and have adequate safe guards if upgrading is your intention. You certainly shouldn't expect someone else to make sure that everything just works for you despite the fact that they are *trying* to achieve that.

[siteartwork]

PHP5 had to support PHP4.x, and let's be honest here for a moment, with PHP5.x supporting a completely re-write, that was never going to be easy to support PHP4 as well.

Okay, but this is not the point. The point is, that a release of the 4.X branch is not BC, and people complain about it.

No one complaints about the (hard) work the core developers do, but about the way this whole referencing-problem (of the 4.x ! branch) was handled, and specially the way the core developers respondend to the reactions of the community.

[siteartwork]

For the same reason, many people still use GCC 3.3 and 3.4 rather than 4.x. This also applies to other languages and applications developed in a certain version of that language. For crying out loud, your job is maintenance of your application. You likely get paid to do it. You should fear upgrading and morseso, you should plan your migration path and have adequate safe guards if upgrading is your intention. You certainly shouldn't expect someone else to make sure that everything just works for you despite the fact that they are *trying* to achieve that.

True. But sometimes users must do an upgrade because of security holes etc.

[siteartwork]

If you are truly professionals--particularly in the enterprise--you don't run into the problems that people are crying like babies over.

Take this one as an example (and I mean example, this is not to start another offense to the php internals):

http://news.php.net/php.internals/18794

SquirrelMail has code like this all over the place:

$value = strtolower(array_shift(split('/\w/',trim($value))));

Here array_shift() does of course change the arg, so that is a potential
problem. And yes, that's a dumb way to do this, but people write code
like this.

So, developing PHP-Apps for the enterprise, does that mean I have to know the Zend Engine like the palm of my hand to do not nested function calls the core developers laugh about?

[jayboots]

So, developing PHP-Apps for the enterprise, does that mean I have to know the Zend Engine like the palm of my hand to do not nested function calls the core developers laugh about?

He wasn't laughing -- in fact, he was trying to find workarounds to continue to support that sort of code style.

More to the point, the take-away is that even if something is listed as a point-release (where BC breaks would normally not be permitted), do not rely on that. It is called testing and the first premise is that nothing can be trusted to work the way it supposed to work. Hey, they write bad code too just like we might and they make bad decisions sometimes. Protect yourself.

I agree that we need get the dev team to appreciate the need for applying security and bc break fixes separately -- but there are better ways to get that point across than going on the list and ranting like a lunatic. I find that very distatasteful.

[siteartwork]

but there are better ways to get that point across than going on the list and ranting like a lunatic. I find that very distatasteful.

Yes, you're totally right. I know this would sometime end in a "who-started-it"-discussion but I can understand the people going nuts about the stfu-comment of Derick.

[ethanp]

I hear a lot of people saying that PHP is dying, but how many websites and programmers still use it? I don't think the numbers are dropping at all. PHP is very easy to learn, and is still very powerful. Sure it has its problems, but doesn't everything?

Recently I was outraged when the company that produces Ruby on Rails said to use their system rather than PHP because it's dying. First I laughed, because who would believe this coming from the profit company that has designed Ruby on Rails. Then the ALA has a new backend in this evil programming language. That's when I was angry. PHP is a much platform for web based applications.

I will continue to teach others all my PHP knowledge and promote its usage. I could care less what a few bloggers have to say, or some crack-smoking company. PHP is the language on the web, and is the basis of hundreds of great open source products. PHP has a bright future once web host stop relying so much on cPanel to do their hard work, and have manually configured servers using PHP5.

[Luke Redpath]

I hear a lot of people saying that PHP is dying, but how many websites and programmers still use it? I don't think the numbers are dropping at all. PHP is very easy to learn, and is still very powerful. Sure it has its problems, but doesn't everything?

Recently I was outraged when the company that produces Ruby on Rails said to use their system rather than PHP because it's dying. First I laughed, because who would believe this coming from the profit company that has designed Ruby on Rails. Then the ALA has a new backend in this evil programming language. That's when I was angry. PHP is a much platform for web based applications.

I will continue to teach others all my PHP knowledge and promote its usage. I could care less what a few bloggers have to say, or some crack-smoking company. PHP is the language on the web, and is the basis of hundreds of great open source products. PHP has a bright future once web host stop relying so much on cPanel to do their hard work, and have manually configured servers using PHP5.

I take it you've not really looked at Rails right? The people who created RubyOnRails used to be PHP programmers. They created Rails because they were fed up with PHP. I'm glad they did, Rails is awesome.

Rails is as open-source and not-for-profit as PHP is. Your dismissal of RubyOnRails as "evil" is immature and laughable at best.

You have too much rage if you have enough to start venting it at something as innovative as Rails or the people behind it.

PHP is indeed the basis of some very good apps. But also the basis of thousands more bad ones.

[Dr Livingston]

The people who created RubyOnRails used to be PHP programmers. They created Rails because they were fed up with PHP. I'm glad they did, Rails is awesome.

Reading your post, and taking note of what Marcus had to say over at www.phplen.com on one blog post, to say the least, and to put it politely, that I was not too amused at what Marcus had to say.

Ruby on Rails has something going for it I think, but for me it's too early to say. At the moment I couldn't use Ruby, never mind if it's on rails or not. Personally I've taken the Python route, but not as a replacement for PHP, but merely to suppliment it.

If yourself, and for that matter Marcus want to move onto Ruby, then by all means do so. Please. I'll be better for it, you'll be better for it, and when I come to think of it, PHP will (proberly) be better for it

Enough said then.

[DougBTX]

I hear a lot of people saying that PHP is dying, but how many websites and programmers still use it?

That isn't what I hear - it seems more like it is treading water, filling in the gaps. PHP5 was all about the object model (though most of the attention seems to be on breakages..) and PHP6 looks to be all about Unicode (though for the things I do it'll just be support for strings) and bundling an encoder cache (which should be good if you dont use Zend or friends today) - so it is getting better, some tydying up, some papering over of problems. Not dead, just not much innovation.

Dougals

[DougBTX]

Reading your post, and taking note of what Marcus had to say over at www.phplen.com on one blog post, to say the least, and to put it politely, that I was not too amused at what Marcus had to say.

Is there anything specific that offended you in Marcus' post? Considering the post he was replying to, it was on the mark for me.

Douglas

[ghurtado]

If yourself, and for that matter Marcus want to move onto Ruby, then by all means do so. Please. I'll be better for it, you'll be better for it, and when I come to think of it, PHP will (proberly) be better for it

You may want to explain in more detail just how PHP will probably be better if Marcus decided to start using Ruby, 'cause I just don't quite get exactly what you mean.

[OzRoy]

Ignore the PHP developers for a minute. Forget about what they have said, or done for a second and just look at us the community.

The fact that this argument is happening at all is a sign that something is wrong. The fact that a lot of very intelligent people have decided to just give up on PHP shows that something is wrong.

Somewhere there has been a major break in communication. I believe it has resulted in an US and THEM attitude and as long as that is allowed to continue these problems will never disapear and will happen every time there is a new change.

If an online community is going to be successful then everyone on both sides has to be part of the same community. A perfect example would be the mozilla community. There is excelent communication between the users and the developers, and that is one of the main reasons for Firefox being so successful.

PHP has the potential to do the same. They have an excelent product, all they need is a little more effort on the community side of things. If they could build an 'official' community then a lot of their problems can be solved. It will become a resource where they can recruit more volunteers. Maybe more people to look after the bug database so that bug reports get treated with a bit more respect.

[dagfinn]

vitriol

I agree with 90% of your actual arguments, but I think your rhetoric (as in "crying like babies") is completely off. You're the most vitriolic IMHO.

You fail to mention who you're talking about, but to the extent you're referring to me, let me just say that you're arguing against a straw man. You seem to do that a lot. In the other thread, you were coming out of that mode, but now you're back.

[Carl]

Well this just confirms what I have been preaching for the last six months. php4 is at the end of it's cycle. It is deprecated and the only good fixes for it are coming from php5 development. It is time that the PTPs to take responsiblilty and stop php4 development and go on to php5 fully. php/fi -.php2-php3-php4-php5 Just because php suddenly became very popular with the release of version 4 does not mean that version is a "forever child".

The stress of maintaining a deprecated version of php and the replacement version is going to keep showing more and more in the dev team and if there is a 4.5 release it will very liklely throw cold water on to a sleeping PHP community. It needs to wake up so that it does not get caught like the Visual Basic community.

[Carl]

Oh, I for got to mention some of the hypocrisies that programming communities seem to enjoy when they come to headers with the language developers.

Unwilling to change syntax and structure to better the language but always willing to threaten or actually change to another language and learn a new syntax and structure.

Unwilling to accept periodical compatiblilty breaks but very willing to change to using newer functionality and leave old functions hanging never to be deprecated.

Unwilling to port to a newer version after bc but always willing to port to a "better" language.

I also forgot to mention that Perl is suffering from the syptoms that PHP will aquire if php4 development continues. Perls syntax has long been its weakness. but things like CPAN and other large projects have ensured that the PERL syntax never matured or changed for the better. The result is a stangnation that causes devlopers to jump ship and "move on". This is also the reason that the coming PERL 6 was created.

[epp_b]

I think this is a little blown out of proportion. Any software may require changes for applications supporting older versions. Consider Windows as an example. When Windows 2000 came out, it took a little while to get reasonable driver support. Generally, medium-to-larger programs that don't say they specifically support Windows XP generally don't. I know I have at least one Win9X-supporting application that runs quite poorly, if at all, under Windows XP.

As for PHP-specific notes, I've only had to make a few changes...only two or three unique instances in 16 different sites, to be exact.

[dagfinn]

You people who are complaining about a FOSS project not meeting your needs and saying that that proves it is not professional or enterprise ready -- take a better look at your own practices and improve them.

I think you've missed the point here. I've pointed this out, but I'm not pleading my own case. Nor is it about the troubles of PHP developers or any particular bug or problem. It's about what's communicated and how that will be seen by people who are outside PHP entirely.

I see a deep incongruency between these two:

• The idea that you can't really expect bugs to be fixed in a FOSS project. I think you've expressed this idea and so does the text on bugs.php.net.
• The idea that PHP is being "driven to the enterprise".

I can live with this, but can the "enterprise" people? I could be wrong, but I don't think they will accept the risk that's implied. They will choose instead to pay for software from a company that commits itself to fixing bugs.

BTW, I notice that Mysql has a "tutorial" on bug reporting that's almost identical to the PHP one but skips the part about how few people are fixing bugs and how you should be "excellent" to the bug fixers to motivate them.