Zend PHP Framework Coming Soon ...

[arborint]

OK, maybe not coming soon, but there have been some stories around the Internet about a possible framework from Zend along with Eclipse IDE integration.

I noticed that Ruby is already there: Ruby Eclipse Integration so Zend is playing catch-up.

Do you think Zend have anyone on par with David Heinemmeier Hannson of 37signals to design the thing? DHH is heavily into Agile Development which is not as popular in PHPland.

testedit

[33degrees]

It would be somewhat surprising for Zend to develop an Eclipse plugin, since that would directly compete with their own offering, but I have often wondered why Zend hasn't developped a framework, or at least thrown their weight behind an existing one. Both Microsoft and Java offer platforms to go with their languages, and while you can't directly compare something like J2EE with Rails, it's a lot more than Zend is offering.

And forget agile development, most of phpland hasn't even heard of design patterns, and truth be told PHP doesn't make it particularily easy to implement them.

testedit

[OfficeOfTheLaw]

And forget agile development, most of phpland hasn't even heard of design patterns, and truth be told PHP doesn't make it particularily easy to implement them.

Sadly, on an XP mailing list I subscribe to whenever PHP is brought up the jokes begin.

[mubs]

I hope they don't build another Framework. There are so many frameworks out there already.

Perhaps they could adopt one, possibly one of the many Rails inspired frameworks that exist for PHP already.

Personally, I'm very happy with Fusebox, and don't need the many "quick start" features of Rails, and if I needed something more MVC/OO, I'd take a closer look at Mach-II.

Mubs

[Selkirk]

I have to agree that PHP isn't getting much agile respect. I've also been Speculating about ZPF.

I think an eclipse plugin makes sense for Zend. Zend Studio is already written in java. Why pedal your own bicycle when you can hop on the bus? I don't use eclipse or Zend Studio, but I look at this part of the announcement as positive.

Regarding ZPF, there is so much that it could be, but all we have so far is a few vague quotes from press releases and press releases are famous for promising the world and delivering less.

[jayboots]

It sure sounds like a lot of focus on eclipse. I wonder if the plan is to integrate PHP as a scripting solution within the platform -- for example, so that you can build eclipse plugins with PHP scripts. It also sounds like they don't intend to compete with eclipse in the long-run but rather wish to harmonize their own IDE with eclipse by allowing it to use eclipse plugins and perhaps, vice-versa. For example, use an eclipse UML editor and have the model exported to your Zend IDE or reverse your code from the IDE back into the model in your eclipse session.

At anyrate, it doesn't sound like the "framework" has anything in particular to do with solutions based code. Instead it seems like it all revolves around the tooling.

Oh, why am I commenting on a rumour?

[Selkirk]

Not a rumor any more: PHP Collaboration project. And yes its a framework.

Is it just me or does it sound like the framework isn't written yet and that this is announcing the project to build it?

[Dr Livingston]

Q. Describe the Zend PHP Framework. What is it and why is it needed?

A. ... We also think that it's very important this framework embody the "extreme simplicity" mindset that PHP itself exemplifies. We want to deliver a framework that helps solve developer problems and speeds development; not one that's overly complex and heavyweight. ...

This is a worry... Just how simply are they talking about? Are they talking procedural programming or not, as that isn't really clear to me. It's a worry if that is to be the case.

Applying object oriented programming methodologies, to me, does mean that there is at one level or other, a degree of complexity. Lack of information is abound at the moment of course - as you'd expect - but my thinking is, not to hold my breath.

[jayboots]

This is a worry... Just how simply are they talking about? Are they talking procedural programming or not, as that isn't really clear to me. It's a worry if that is to be the case.

Why is this a worry? No matter what it is, I don't see any mention of them dispatching armed soldiers to make you use this stuff at gun point.

[Selkirk]

I wonder if "extreme simplicity" is a code word for no objects?

Who is the I in this statement?

Q: Aren't there other existing PHP frameworks? Why is this one different?

A: While there are other existing PHP frameworks, they are all either very specialized or don't embody the "extreme simplicity" mindset I described above.

I can help but notice that every framework project starts out by saying that they are going to be "lightweight" and don't want to be like those other complex, heavy weight frameworks.

Then, the difficult real world use cases start coming in and as the framework matures it starts to look more and more like all the other frameworks that provide the same level of capability.

Its easy to be light weight if you don't do anything.

Being simple is very very very hard. If it were easy, there would be an abundance of existing "extremely simple" php frameworks to choose from.

Rails achieves its simplicity it by making assumptions and limiting the scope of what it tries to do.

However, tool vendors and committees are not traditionally good at limiting scope.

[chiefmonkey]

This is a worry... Just how simply are they talking about? Are they talking procedural programming or not, as that isn't really clear to me. It's a worry if that is to be the case.

Possibly but it would be a bit silly of them to invest all this time improving OOP in PHP5 (and PHP6) and then not use it for their own framework.

George

[33degrees]

This is a worry... Just how simply are they talking about? Are they talking procedural programming or not, as that isn't really clear to me. It's a worry if that is to be the case.

Applying object oriented programming methodologies, to me, does mean that there is at one level or other, a degree of complexity. Lack of information is abound at the moment of course - as you'd expect - but my thinking is, not to hold my breath.

This, to me, is the paradox of Zend pushing PHP into the enterprise. It may be my personal bias, but I've always felt that OOP is essential to developing big complex projects, and given PHP's currently poor OO support, I don't really see how far PHP will be able get in the enterprise space. PHP5 is definitely an improvement, but there's still a long ways to go, and much of it would mean breaking backwards compatibility; At that point, they'd probably be better off starting from scratch.

In a way, I think that PHP is a victim of it's own success. Witness the rate at which hosts are offering PHP5; for them, and for the vast majority of the web, PHP4 offers all they need. My prediction is that Zend are going to facing an increasingly uphill battle as they come out with new versions, while the majority of the web sticks with PHP4.

In the end, I think Zend's best bet would be to follow Microsoft's lead with ASP, and start a new, incompatible branch of the language. Let the procedural programmers keep working with v4 or 5, as they see fit, and push the new, improved PHP for the enterprise

BTW, Great link, Selkirk.

[arborint]

I take away three things from the PHP Collaboration Project announcement.

First that they have some companies on board. That could be good because it will get web app builders rather than tool builders involved in the process. They do make a point to say thay they will "actively engage with and support the efforts of individual developers" but I don't see a "Click Here to Join" button anywhere.

Second is that an unknown somebody has a vision of this framework and the focus seems to be on "Extreme Simplicity" whatever that means. It is a little worrisome because the reality is that useful frameworks are complex beasts that need excellent design to succeed. That's why I asked iniitally who the architects are -- because Zend seems to have been lacking in vision since PHP4 was released.

Finally, it seems that they are moving their development platform to Eclipse. I would guess that gentle encouragement from their new large partners and the high cost to maintain an IDE codebase are behind the decision. Now they can just focus on the PHP tools and leverage everything else in Eclipse. (I should note that I'm happy becuase I'm an Eclipse user).

[33degrees]

Why is this a worry? No matter what it is, I don't see any mention of them dispatching armed soldiers to make you use this stuff at gun point.

The issue is that people are starting to think that Zend are out of touch, and having them come out with a procedural framework would definitely reinforce that opinion. This is a pretty big move from Zend (albeit a very neccesary one, in my opinion), and if they 'screw it up' it's going to hurt PHP's credibility, and that's something that's going to affect all of us, regardless of whether we use the framework or not.

[jayboots]

The issue is that people are starting to think that Zend are out of touch, and having them come out with a procedural framework would definitely reinforce that opinion. This is a pretty big move from Zend (albeit a very neccesary one, in my opinion), and if they 'screw it up' it's going to hurt PHP's credibility, and that's something that's going to affect all of us, regardless of whether we use the framework or not.

Ok, butterfly effect and all that. In the end I believe that you are most affected by your own skills, opportunities and decisions. I also don't think that the pundits here are any wiser when it comes to writing/designing frameworks than Zend (IBM/Oracle/MySQL/...) might be and vice-versa.

BTW, which "people" are thinking that Zend is out-of-touch? That doesn't sound at all representative.

[sweatje]

Off Topic:

Not related to the framework topic, but I think anyone who watches the internals list has seen people with @zend.com email address get flamed as much as the next guy by the php core group. Zend certainly has a big dog in the race with PHP, but it is a far strech to say they control it.

[JaredWhite]

I like this news, and I think it could be a good source of standardization around core "framework" concepts which PHP desperately needs. That being said, I have a few concerns:

1. Why make this announcement and offer seemingly NO contact information? No e-mails, no forms, no forums, no mailing lists, nothing. How can you start an open source project with a single press release? Dumb.

2. I don't understand what the goal is. Is the goal to compete with Java in the enterprise? Is the goal to compete with RAD-style cutting-edge frameworks like Rails? Is the goal to create a culture of innovation and move the PHP world forward? Is the goal to do something very basic and blindingly-obvious just so every PHP project doesn't reinvent the wheel?

3. Is this going to make use of existing PEAR classes? If not, that seems like a gigantic waste of opportunity. Hopefully they aren't that out of it.

4. It makes me nervous when a Zend press release "puts down" existing PHP frameworks with nonsense about them being too specialized or too complex. There are some fine frameworks out there already that are general-purpose and simple to use. CakePHP is the first that comes to mind. Since that is PHP4-compatitble and Rails-inspired, I'd of course understand why Zend wouldn't get involved in that sort of thing, but why make this somewhat insulting remark? Not a good foot to start out on.

Anyway, I'll keep on eye on these developments, but, for now, I'm choosing to spend my development time on my own projects (announcements forthcoming ).

Jared

[Dr Livingston]

Why is this a worry?

It is a worry because, Enterprise level applications are by their very nature complex beasts, and the only way that from a development - and a business - point of view that that complexity can be expressed, is only by object oriented programming.

You have of course, other tools such as UML for example, but at the end of the day, it's about modeling the business isn't it? How on this Earth, can you express a business model using procedural methods?

Damn impossible if you ask me.

[33degrees]

BTW, which "people" are thinking that Zend is out-of-touch? That doesn't sound at all representative.

A small but realtively vocal minority; I'm specifically thinking of recent threads regarding bug reporting and the 'cannot reasssign $this' fiasco. I do realise that I'm mixing up the PHP Core group and Zend, and it's possible that there is less overlap between the two I'm assuming. The fact is that there are some people that aren't happy with the PHP development, and I'm assuming at least some of that is from Zend's involvement.

[bonefry]

I have to agree that PHP isn't getting much agile respect. I've also been Speculating about ZPF.

It's really disturbing kids today are so quick to proclaim a platform's death. It's also unclear to me why all this blogs pop out of nowhere and then are used as references in other blogs (if that guy said it ... it must be true ... afterall ... he heard it from a friend).

BTW, there isn't a single technical reason in that blog that prevent PHP from being used as a platform for "agile development".

And could you guess what other platform doesn't have a problem with "agile development" ? I will give you a hint ... which I don't trust myself ... but it's better than yours.

(just curious: who's Vincent Foley-Bourgon ?)

[sweatje]

It's also unclear to me why all this blogs pop out of nowhere and then are used as references in other blogs (if that guy said it ... it must be true ... afterall ... he heard it from a friend).

Perhaps it would be clearer to you if you knew that Selkirk, who posted the links, is the author of said blog. He was posting them as a shortcut way of saying "I have formulated some thoughts on this subject already, and they are over here if you care to look at them".

[bonefry]

I don't understand what the goal is. Is the goal to compete with Java in the enterprise?

The goal is simple and predictable ... to create buzz around Zend because they are loosing mementum (PHP is not the brightest start around ... anymore). About your assumption ... that would be a foolish goal.

[bonefry]

Perhaps it would be clearer to you if you knew that Selkirk, who posted the links, is the author of said blog. He was posting them as a shortcut way of saying "I have formulated some thoughts on this subject already, and they are over here if you care to look at them".

I realized that ... I was talking about links inside his blogs.

[sweatje]

I realized that ... I was talking about links inside his blogs.

Well, that is just the brave new world of blogging and freedom of communication. Anybody with minimal effort to locate hosting can create a ruckus with their opinions by blogging. I guess they at least have to be relevant enought to be found in the first place.

[arborint]

It's really disturbing kids today are so quick to proclaim a platform's death.

From what I can tell, you re the only one discussing that. So I guess "kids today" means you?

It's also unclear to me why all this blogs pop out of nowhere and then are used as references in other blogs (if that guy said it ... it must be true ... afterall ... he heard it from a friend).

What a crazy idea: people have opinions, they put those opinions one the web, these same people put links to each others opinions ... go figure ...