Trouble in Zend Framework Land?

[akrabat]

Another perspective is that Zend's announcement at ZendCon is no different than many of Apple's announcements at WWDC. For example, Apple announced its Intel plans last year. They wanted developers to be ready. Zend announced its framework plans last year. They wanted developers to be ready.

Not a fantastic comparison as Apple made available "beta" computers for developers so that they could get ready...

Zend may have been better to either not mention it or provide a fairly good date on when something would be released.

[McGruff]

the "PHP way" ....

The php way desperately needs to improve, I think - or rather it needs to develop a bit at the "enterprise" end. Php is a bit like a bus you can get on and off again wherever you like. That's a good thing but there need to be some clear signposts for those who want to learn more. My hope is that the Zend framework will do a lot to promote OOP and testing.

I recently joined a team advising on how to develop a site. Basically, if we want decent software we'll have to write it ourselves. I'd like to look forward a year or two from now to a time when there might be a range of solid apps to choose from, built with Zend or other, similar frameworks, which will satisfy the test-infected amongst us.

[firepages]

No one is suggesting copying Java.

But you are and you do ,constantly, implicitly or otherwise.

Whatever happens Marcus I think you (and a few other regulars here) should brace yourselves to be dissapointed with the Zend framework.

Don't take this the wrong way but IMO thats a good thing~ if you thought that the framework was brilliant that would probably mean that it was inaccessible/irrelevant for most regular PHP users.

[lastcraft]

Hi...

But you are and you do ,constantly, implicitly or otherwise.

But I don't do Java . I'm a C++'er... (and possibly a Rubyist nowadays) .

Don't take this the wrong way but IMO thats a good thing~ if you thought that the framework was brilliant that would probably mean that it was inaccessible/irrelevant for most regular PHP users.

Actually I take that as a complement . I'm not after a clever framework though, I just want the best for the users.

Really, I'd quite like to see code that was accessable. SimpleTest does a lot of things differently from the fiendishly clever JUnit, just because it would have made it inaccessable. I actually think PHP is brilliant in this regard...and moan all the time, because it could be so much better with a bit of bug fixing and finishing off.

I wouldn't recommend that the ZPF introduce dependency injection for example, and I think the choice of ActiveRecord is a good one. On the other hand, it wouldn't pay to have a glass ceiling, so I think Zend have to tread a bit carefully. PHP is not Ruby. If you spray static methods around, there are no mixins to dig you out of the hole.

How about this alternate syntax...?

$transaction = new ZUnitOfWork();
$people = $transaction->Person->findAll(); 

This would get around the save() problem of the ActiveRecord, you only have one finder (so only one extra line of code) and the fluent interface makes it just as syntactically sweet as the static.

yours, Marcus

[33degrees]

Perhaps the problem is that they're not being secretive enough. If no one knew about the framework, all of this speculation and criticism wouldn't exist. They could be like Apple and only announce something after it had been completed.

Another perspective is that Zend's announcement at ZendCon is no different than many of Apple's announcements at WWDC. For example, Apple announced its Intel plans last year. They wanted developers to be ready. Zend announced its framework plans last year. They wanted developers to be ready.

Except that Intel based Macs are not an open source web-app development framework. Besides, Apple offered developer kits to anyone who was willing to pay for them. Zend wants people to be ready? Then give us something to prepare with, not syntax examples that can't work.

Linus gave us "release early, release often," but that doesn't mean he released Linux to the world before he was ready. There's also the fact that his project wasn't immediately popular - he still had very few contributors for the first year or two, and it took him more than two years to release 1.0. In other words, the conditions were entirely different.

The two situations are very different, yes; when Linux was released to the public, the internet was just starting to catch on, most people had never heard of open source, and there was no clear demand for an open source operating system. Zend, on the other hand, already have all that playing in their favor. Besides, Linux's lack of popularity at first doesn't change the fact that it's release early, release often schedule was a prime reason for it's eventual success.

Try to contribute a patch to the Linux kernel now, and see if you still think it's the best example to further your argument. The same goes for Apache, MySQL, Perl, and just about any other successful open source project.

No, the difference isn't that those projects are successful, it's that they're mature, and the reason they're mature is that many many people poked through the code and submitted patches. Besides, I'm not saying the should be accepting patches, I'm just saying they should let us see the code.

In fact, PHP is known for having the most open development process of the major open source technologies, and this is often a source of criticism. This is the group of developers with which Zend is working, so I highly doubt they're blind to the advantages and disadvantages of their approach.

I don't think the criticism is directed at their openess as much as how they've handled the process. People may complain about things that have been added to the language, but isn't that the fault of the people who accept the patches? Moreover, developing a language isn't the same as developing a framework, it's much easier to fix mistakes in a framework than in a language.

Anyway, Zend surely has their reasons for choosing to keep everything under wraps at the moment, and I imagine that they didn't take that decision lightly. But I do think that they haven't been handling this as well as they could.

[dagfinn]

No one is suggesting copying Java. We don't have to copy Rails either.

For the record, it's the PHP developers who have been most enthusiastically copying Java, as evidenced by the new features in PHP 5.

[pachanga]

Off Topic:

How about this alternate syntax...?

$transaction = new ZUnitOfWork();
$people = $transaction->Person->findAll(); 

How about a separate thread for that ? The idea is very interesting!

[33degrees]

Getting back to the issue of static finders vs. finder objects, I was just thinking that a good solution would be to implement the functionality in a finder object, but add static methods as wrappers to the finder object in the active model; best of both worlds?

[kyberfabrikken]

How about a separate thread for that ? The idea is very interesting!

A different slant on persistence ?

[dgx]

Workaround: correct self:: behaviour

[Ren]

Think I'll wait till it gets patched properly.

Atleast one patch for it exists atm, so hopefully will be fixed soon.

Back to the original topic, anyone think that the Framework is going to make an appearance before end of February, as Andi suggested on the PHP podcast?

[shiflett]

Back to the original topic, anyone think that the Framework is going to make an appearance before end of February, as Andi suggested on the PHP podcast?

I do. :-)

[Selkirk]

Well, I doubt its going to be announced today or tommorrow. A Zend announcement on Monday would be most welcome. Apple has a big announcement on Tuesday. Wednesday is March. We'll know soon enough.

Anyway, its not like the framework will turn into a pumpkin at the stroke of midnight on the 28th.

[Dr Livingston]

Well, I don't want to be a party pooper but I can't see any announcement this spring, maybe this summer... ?

[Ren]

Well, I don't want to be a party pooper but I can't see any announcement this spring, maybe this summer... ?

It was in the podcast http://podcast.phparch.com/main/inde...sodes:20060127 where Andi mentioned he hoped there be something for us to see in Feb.

[Dr Livingston]

Still nothing definite though what I did hear though, is the reservations of Ruby On Rails within the conversations... I've stated in the past that I have my own reservations, well I can't be all that far off, by following my gut instincts huh

So people, I wouldn't wipe PHP of the map just yet then...

[Selkirk]

Mar 1. Oh, well. These things happen.

[Ren]

Or don't

[DerelictMan]

Mar 1. Oh, well. These things happen.

Maybe this is supposed to be it? >:-)

Just kidding, just kidding...

[Ren]

Ah, its arrived..

http://framework.zend.com/

[thr]

No AR? Guess they didn't warp time and fixed the self:: "bug" in 5.0.2+

[Ren]

No AR? Guess they didn't warp time and fixed the self:: "bug" in 5.0.2+

Still alot of debate how exactly to fix it, and what it should be called...

this::
self::
owner::
caller::

Etc...

[thr]

Yes I know, but I find it more then amusing that they showcased code which didn't work enough of that tho... the ZF looks well.. decent, the search and mail functions are very cool tho

[Ren]

Yes I know, but I find it more then amusing that they showcased code which didn't work enough of that tho... the ZF looks well.. decent, the search and mail functions are very cool tho

Yes, already plowed into the Search, seems to be creating an index fine, but not getting results back

[akrabat]

enough of that tho... the ZF looks well.. decent, the search and mail functions are very cool tho

The front controller does what it says on the tin, though I had to write my own router as the Zend_Controller_Router assumes that the first element of request_uri is the controller's name.

Can you do mod_rewrite magic so that the http:://localhost/test/zend/controller/action results in a $_SERVER['REQUEST_URI'] = '/controller/action' ? Not that modifying the router was hard.

I couldn't find any documentation on the Controller, so I stuck up my notes on www.akrabat.com. Of course, I might have completely mis-used it!