Well I'm workin on the PDF features =P
You've heard me/us, and I guess you guys need to make a call. Lots of people
think two-step is important, and are asking for it's inclusion. I hope you
guys reconsider so that at launch we have a viable two-step solution to put
in the tutorials, books and sample applications.
K
----- Original Message -----
From: "Andi Gutmans" <[EMAIL PROTECTED]>
To: "Kevin McArthur" <[EMAIL PROTECTED]>; "Matthew O'Phinney"
<[EMAIL PROTECTED]>; <[email protected]>
Sent: Monday, June 25, 2007 10:40 PM
Subject: RE: Fw: [fw-general] Two-Step View, subclassing controller, etc
Hi Kevin,
I have talked to a large amount of PHP developers who have either started or
looking to start adopting Zend Framework.
For the majority of these people who depend on and buy into our various
goals it is more important for them to have a stable code base which they
can start leveraging in their critical applications as opposed to having the
perfect framework on day 1. With that in mind I have mentioned many times on
this list that this is the intention for 1.0.
There is definitely life after 1.0. Right after the release we will start
working on the next set of functionality and I believe the release cycle for
many of that does not have to be particularly long. As long as we retain BC
we can start getting functionality out in minor releases (e.g 1.1, 1.2,
etc...) while we might do a couple of mini releases (e.g. 1.0.1, 1.0.2) in
parallel for bug fixes.
So far the feedback re: ZF has been very positive and I believe overall it
will continue to be positive. We are also working on improving the content
of the Web site and the go-live messaging plan in order to get the key
points across. Part of this will also be a high-level roadmap which will
make it clear that live doesn't stop at 1.0. There is more that we all want
to get done. Will there be people bickering here and there. Sure and you
always see that with any technology/framework.
As long as we continue staying true to our goals and work towards that I
believe we will be extremely successful. Yes, it's at a slower pace than
many would want to see but that slower pace is one of our biggest
advantages. ZF is a very high-quality piece of software and I believe we are
setting new standards for quality in our user-base
So things like forms, view layout, tooling, pdf features, etc.. all very
important things will definitely happen. I probably have the longest list of
all of things I'd like to get done and think are important :) But I believe
for the benefit of the user-base we are better off delivering those
incrementally after 1.0 rather than delaying for a few months. And yes, in
order to meet our quality-goals and well thought out and broadly reviewed
architecture it would take much longer than just svn committing the source
code that people have written for this.
Andi
-----Original Message-----
From: Kevin McArthur [mailto:[EMAIL PROTECTED]
Sent: Monday, June 25, 2007 8:20 PM
To: Matthew O'Phinney; [email protected]
Subject: Re: Fw: [fw-general] Two-Step View, subclassing
controller, etc
Ok, I'll bite.
> There's plenty of interest. The issue right now is getting a stable
> and tested version of ZF out the door that developers can
rely on. We
> are in RC status, which means, loosely, no new features, and
> bugfixes/security fixes only. (I say loosely, as there have
been a few
> new features creeping in, but most have been backwards
compatible or
> augment existing
> functionality.)
RC3 wasnt BC to RC2. And RC1 introduced a massive new feature
(the view
renderer) which we needed to address it's implications.
Somehow the viewrenderer was ok to add in an RC but the
response to it isn't? Thats bogus.
> To give an idea of what I mean by "getting a stable and
tested version
> of ZF out the door," I personally addressed over 30 issues in the
> tracker between RC2 and RC3. There are still a few lingering issues
> and a few classes that need better test coverage. But if we add new
> features such as layout and partials support *now*, that
means pushing
> the date back.
Zend_Layout can be applied to the SVN today, in its current
form, without breaking ANY b/c (it patches right in) and
without modifying the dates. It has been working for weeks.
> Some may ask, why not push the date back and get *all* the
features in?
> Simple: when do we decide we have all the features? when
does the API
> freeze? There's a really great article about time-based release
> management versus "release when ready":
Why was there no list of goals to make a 1.0. Why was 1.0 a
date not a level of completeness. People are waiting for the
framework to be complete, not labeled with a version number.
If it were up to me ZFW 1.0 would have layouts, Zend_Form,
and probably a more unified set of APIs for
Zend_Request/Response and the differences between fetchAll in
Zend_Db_Table and with Zend_Db_Select fixed and standardized
such that the result of fetching is always the same (one
returns an object with properties another an array now)
Slapping a 1.0 label on it doesnt make the framework any more done.
> What it comes down to is: a lot of developers are waiting to use ZF
> until it has its first stable release, and continuing to
push that off
> into the future will only delay ZF uptake -- and thus
contributions to
> the project.
If you release too early (which many of think this is) you
risk setting the reputation of the framework [never get a
second chance to make a first impression]. As-is, while it's
excellent, it's going to get bad reviews for a lot of
legitimate but easily resolved reasons.
> There is definitely room for new features and polishing --
that's why
> there *will* be life after 1.0. Stay tuned after the
release -- there
> are plenty of proposals and ideas just waiting in the wings
for after
> this milestone.
What this says to me is that 1.0 just means what 0.8 and 0.9
meant. No stable release.
What hasn't been fully considered is that printed
documentation will be written for 1.0, tutorials generated
and they will all use a workflow many of us find less than
ideal. Where they use two-step, it will be either an
unofficial solution (layout, view-enhanced) or the
exceedingly complex custom-plugin approach.
This will set the learning on framework in a
header.tpl/footer.tpl non-two-step standard with no realistic
solution for form building [the most common action for any web app].
I don't mean to sound critical (I'm a big FW proponent), but
there's a lot of us who don't agree with the current course.
It would be painless to add layout, it would be even better
to fix the other 4-5 big issues before we set the API in stone.
$0.02
K
>> Pádraic Brady wrote:
>>
>> or this...lol
>>
>>
>>
http://svn.astrumfutura.org/zendframework/trunk/library/Propos
ed/Zend/View/
>> Helper/Partial.php
>>
>> Never have so many heads bumped the same wall...;). I do think
>> cross-module
>> partials are useful. The problem doing it is
configuring a new View
>> assuming it has no interaction with a controller. The
>> ViewRenderer would
>> likely work though it's lodged in the Controller.
>>
>>
>>
>> Pádraic Brady
>> http://blog.astrumfutura.com
>> http://www.patternsforphp.com
>>
>>
>> ----- Original Message ----
>> From: Ralph Schindler <[EMAIL PROTECTED]>
>> To: [email protected]
>> Sent: Monday, June 25, 2007 10:54:11 PM
>> Subject: Re: [fw-general] Two-Step View, subclassing
controller,
>> etc
>>
>> Matthew Weier O'Phinney wrote:
>> >
>> > Partials have been on my to-do list for a while now. They
>> actually don't
>> > require any changes on the view level; you can very simply
>> create a new
>> > View object in the helper, setup the environment
from the old view
>> > object (minus the variables), assign variables as
passed, and then
>> > render the "partial" view. I just need to write good
test cases for
>> > them, and determine the syntax for pulling them.
>>
>> You mean like this: ;)
>>
>>
http://svn.ralphschindler.com/repo/Xend/library/Xend/Layout/Vi
ewHelper/
>> Partial.php
>>
>>
>>
>>
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
━━━━━━━━━━━━━
>> Get the free Yahoo! toolbar and rest assured with the added
>> security of
>> spyware protection.
>>
>>
>>
>>
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
━━━━━━━━━━━━━
>> Got a little couch potato?
>> Check out fun summer activities for kids.
>>
>
> --
> Matthew Weier O'Phinney
> PHP Developer | [EMAIL PROTECTED]
> Zend - The PHP Company | http://www.zend.com/
