Zend PHP Framework Coming Soon ...

[jayboots]

You're free to leave any time.

...and miss out on this brilliant repartee?

[Quadzoola]

I disagree... The certification isn't worth the paper it's written on. I'm going to get hammered for making that comment but that's my thoughts on it, once you compare it against the certification that Sun offers.

High school level? Try pre-school level, maybe that is closer to it. It's laughable, if it weren't so lacking...

Not planning on 'hammering' you per se, but I think you're missing the point of the certification here. It at least presumes a certain level of understanding of the language itself--granted using php effectively is another story. I do give it weight when hiring new programmers as do other companies and I think that in and of itself contradicts your opinion.

It's also nice to quote on projects and be able to state that all your programmers have A, B and C certification. Now we know that some certifications hold more wieght than others but as far as clients are concerned it makes them feel more at ease, at least this has been my experience.

[bartb]

http://www.zend.com/collaboration/

[arborint]

http://www.zend.com/collaboration/

There's still not anything of substance there. They did put up a forum -- I guess figuring it would keep people busy for a while.
At some point they need to state whether the ZPF is based on an existing framework that one of the "partner companies" donated or it is an all new effort.

[asmictech]

In any capacity the framework will be, i would suggest Zend makes will great. There are many out there as alraedy stated. PEAR and so on. I only think they should have something that encompasses PHP-GTK i.e. windows application(GUI) and the existing PEAR. Or is it re-inventing the wheel?

[akrabat]

It's all vapour until I see some code...

[Selkirk]

Based on Andi Gutmans' comment on my blog, I think its still a work in progress:

I hope we will be able to put up some short code examples of some completed work in the next couple of weeks; but it will take a while longer to be fully baked.

The FAQ has been updated with a timeframe:

How/When can I join?
We're now working on setting up the collaboration infrastructure and engagement guidelines. We expect to have them ready by January 2006. In the meantime, we will use this medium to share progress on the project development.

[arborint]

In reading Andi Gutmans' comment, the immediate warning sign is that he only discusses their "database interface". It sounds like they are still back in days when database abstraction layers were all the rage. Quite honestly unless they are going to do something like Hibernate, then the "database interface" is one of the last things on the list of framework needs. I predict that the "20%" they plan to build will be nothing more than a PEAR replacement (hence why PEAR is out of PHP6).

[HarryF]

In reading Andi Gutmans' comment, the immediate warning sign is that he only discusses their "database interface".

It's a comment on a blog, not a white paper. There's been some talk about security being something this framework will address - you'll have to scan planet-php for comments on the Zend conference.

It sounds like they are still back in days when database abstraction layers were all the rage. Quite honestly unless they are going to do something like Hibernate, then the "database interface" is one of the last things on the list of framework needs.

Disagree - it depends one what you're trying to abstract. Think what Jeff's done with WACT's DB drivers illustrates the point e.g.

$username = DBC::getOneValue('SELECT username FROM users WHERE ID = 1'); 

A database "simplification layer" (vs. an ORM, SQL abstraction tool or otherwise, most all implementations of which start to get painful the moment you have a real world use case beyond basic CRUD).

I predict that the "20%" they plan to build will be nothing more than a PEAR replacement (hence why PEAR is out of PHP6).

Could well be right but why not? PEAR as a whole is basically beyond salvation (despite having some gems) because things like error handling are "wrong" and deeply embedded, version control has become a mess, the test framework (PHP_Unit v1) which most of the code should use has been abandoned by its author etc. etc.

Put another way, the main mistake they could make is implementing anything more than very very basic controllers - it's here most disagreement happens - and that sounds like the point Andi is trying to make.

And if they've got guys involved like those that worked on the SDO extension, I'd be optimistic - considering the tutorial, I personally wouldn't know how to sensibly go about something that does this;

Data objects can automatically record their change history that can then be used by a DAS to detect collisions when applying changes back to an enterprise information system.

These are not your typical PEAR developer.

Anyway;

It's all vapour until I see some code...

True - till then it's not worth speculating. Of course if they'd just dumped some code, perhaps we'd be complaining there was no warning

Long term (and it may not be all that long) all this Zend + IBM + corporations means, if you want it, there'll be plenty of cubicle jobs for PHP programmers - steady pay etc. Not bad if that web startup doesn't take off. Zeev and Andi will hopefully get some kind of financial reward one day, although there's not much money to be made these days if your primary customers are developers but they deserve it - they've worked hard for it (and yes - so have others).

[hippypink]

Of course something like http://phpOnTrax.com (php on trax) tries to provide the best of all worlds.
A. No need to learn Ruby,
B. Yet follows the Ruby framework almost perfectly, allowing programmers to go between ruby and php--hopefully--and use a great MVC.
C. Supports PHP5
D. works with PEAR libraries
E. Broader support since php is more widely used
F. Not controlled by Zend (good/bad thing?)

yes, there are other mvc php frameworks, but none so elegant or true to rails, oo, etc...

I still see this as ultimately just a stepping stone to something far better.

[Carl]

This would not surprise me in the least since the only way of getting a PHP framework to to do what Ruby on Rails does would be to have complete API built into php extensions. This (to answer a question in another thread) is the reason there is no defacto framework in PHP. In order to have control and enforce the architecture the Apache server would have to have power over what the code writer does. This can only be accomplished when the coder is using only what is provided by the framework, as in Rails development.

This is why the present "framework" projects (blue shoes, solar, prado and the rest ) are not frameworks but abstraction layers at the very most. At present the only true framework project is Midgard www.midgard-project.org where the core structure is control and implemented through php extensions and apache modules. Unfortunately this project suffers for developeritis, where the developers are too nerdy to think about making the system popular or widely available to the masses by pushing the system to web hosting circles. In otherwords a terrible marketing plan.

In walks Zend. It would benefit the Zend company to help and join in with the Midgard project rather than start a new and similar one. Everything is already in place the only thing they would need to do is to put their support into making the php extensions necessary to run Midgard readily available in PHPs installation the way they do with SQLite. I am not sure if Midgard qualifies as true MVC (more like front controller) but this would not be hard to fix with Zends involvement. It may not even be necessary to fix anything if the system were given a helping hand by having the Zend engine control part of the MVC.

So there you have it. If Rasmus and company see fit a PHP Midgard Framework to rival Ruby on Rails could be ready to go in less than a year.

[lastcraft]

Hi.

It is just me, or have Zend/PHP finally woken up again. Both the Eclipse integration, terribly undervalued until now, and the more inclusive release process are great news. Not sure about the framework though. A class library (the soap ext. and PDO are a start) is needed first.

But, hey, I feel happier this week .

yours, Marcus

[Selkirk]

I think the combination of low adoption rates for PHP 5 and the popularity of Rails has given Zend an epiphany.

I'm only guessing, but I bet that their sales force is finding that once they do the work convince a company to consider alternatives to Java/.Net, that they are seeing them slide right past PHP for Ruby.

I've written about the the value of MVC frameworks on my blog and here at sitepoint. I stand by my prediction that in 2007 writing a web application without an MVC framework will seem quaint and be rare.

The enterprise web development top tier is:
Java + (Struts | Tapestry | Spring | JSF) + (iBatis | Hibernate)
(C# | VB) + .NET

Leave the top tier and you have:
Ruby + Rails
PHP + ???
Python + ???

PHP needs at least one mature, well known framework.

I think the process of writing ZPF will reveal things about PHP language itself that need to be fixed. There are reasons that there are no mature frameworks for PHP already. I think ZPF will end up php 6 only.

I'm highly optimistic about PHP's future. I see alot of good things coming from Zend and from the PHP internals team. An opcode cache in PHP 6 is a litmus test for me. With it and a few other things, I think PHP 6 is going to be the natural successor to PHP 4.

[33degrees]

I think the combination of low adoption rates for PHP 5 and the popularity of Rails has given Zend an epiphany.

There was an article in
The Register that discusses Zend's recent attempts to push PHP into the enterprise. Of particular interest are the two last paragraphs.

The renewed push for PHP comes after analyst Evans Data Corp (EDC) reported a substantial decline in the use of PHP by developers in Europe, the Middle East and Africa (EMEA). In August, it was reported that PHP interest had dropped 25 per cent year-over-year with 40 per cent of developers saying they had no plans to evaluate or use PHP for future projects.

EDC blamed the drop on the failure of PHP to penetrate the enterprise developer market.

It's clear that Zend have worken up. Let's just hope it wasn't on the wrong side of the bed

[McGruff]

That's encouraging. If it's enterprise developers they're chasing maybe we'll have a chance to persuade them to do something test-driven.

[Dr Livingston]

But, hey, I feel happier this week

You do? Wasn't that long ago that you were spouting off that PHP was broken and unworkable

[bigduke]

I only hope Zend doesn't turn out to be the next Sun Microsystems. Buggers stole java's control from us Ironically, if this happens I'll switch to Java and maybe even change the kinda apps I develop. I get this bad feeling from teh last Q on the Q and A section "What do our customers think about all of this?" I wonder what took them so long to even come up with this decision for an enterprise framework. I also noticed that SAP, Zend's partner isn't even in the list of companies who have already joined hands on this prj.

On the "extreme simplicity" phrase: Yeah! why not lets turn that one little pure language where you could actually write some nice code into VB as well :-P (Mega pun intended)

But in the end, if this comes through, I hope to get more respect from them .NET buggers. :-|

[sweatje]

I only hope Zend doesn't turn out to be the next Sun Microsystems. Buggers stole java's control from us

How can one steal something which is already owned becuase they created it?

[sweatje]

Some additional tidbits on the network itself:
http://netevil.org/node.php?nid=633

[33degrees]

Thanks for the link, sweatje. I'd recently changed my feelings about this framework to "Cautiously optimistic", and the little info contained in the post is encouraging.

[McGruff]

Seems that it will be unit tested OOP after all

[Ezku]

I'd recently changed my feelings about this framework to "Cautiously optimistic", and the little info contained in the post is encouraging.

My reaction was the same. This will get interesting.

[bigduke]

How can one steal something which is already owned becuase they created it?

My bad on that! What I really meant was i hope they don't come up with a "we pwnz j00" attitude and completely commercialize the framework itself.

[sweatje]

Jump to the W

Java is a reflective, object-oriented programming language developed initially by James Gosling and colleagues at Sun Microsystems. Initially called Oak (named after the oak trees outside Gosling's office), it was intended to replace C++, although the feature set better resembles that of Objective-C. Java should not be confused with JavaScript, which shares only the name and a similar C-like syntax. Sun Microsystems currently maintains and updates Java regularly.

And we all know we can trust everything we read on the net so case closed

[akrabat]

From the link that Jason posted:

"One of the goals for the project is to keep every clear and simple to use, without forcing you to adopt the entire framework throughout your application; it doesn't impose itself on your app, and doesn't require any configuration files to deploy and use."

and

"Yes, the Zend Framework is PHP 5 OOP based, with unit testing. There's also documentation"

I'm also in the cautiously optimistic camp now.