Poll: what version of PHP for ZF2?

[weierophinney]

Hey, all --

As discussed today in the IRC meeting, and summarized in the log,
there's an open question of which PHP version we should use as our
minimum supported version.

I've started a poll page here:

    http://framework.zend.com/wiki/pages/viewpage.action?pageId=46792838

to summarize the ideas floating around. Per our meeting, I'd like to
open the floor for discussion. Post comments to the page, reply to this
thread, or discuss on #zftalk.2. Let us know what _you_ want, and more
importantly, _why_ -- try to give concrete reasons.

Statements like "we'll kill adoption rates," or "it'll take us another
six months to complete" may seem very true, but without evidence to back
them up are speculation.

On the flip side, statements like, "we have migrations already in place,
and doing this will cost <X amount of time and money>", "our hosting
provider has indicated they will not adopt 5.4 until <date in future>",
"traits will reduce the amount of code we need to produce and maintain
by <X percent>", etc. give everyone concrete data to work off of.

I've locked the poll for now so we can have discussion and give time for
everyone to inform themselves of all points of view. We'll open it up in
a week for actual voting.

--
Matthew Weier O'Phinney
Project Lead            | [hidden email]
Zend Framework          | http://framework.zend.com/
PGP key: http://framework.zend.com/zf-matthew-pgp-key.asc

--
List: [hidden email]
Info: http://framework.zend.com/archives
Unsubscribe: [hidden email]

[Tomáš Fejfar]

What about using traits - in my understanding they're like implemented
interfaces - and compiling them into some 5.3-safe build of ZF2. For 5.3
the trait implementation would be moved to the class and trait class will
be changed to same-named interface.

That would allow use of trait without future refactoring and also use of
5.3. The dark side of this is, that it will look strange without the
knowledge of the build process. But that's IMO only a small problem.

Some frameworks have quite successfully managed to do this when switching
from 5.2 to 5.3 (i.e. namespaces)


On Wed, Feb 1, 2012 at 11:59 PM, Matthew Weier O'Phinney
<[hidden email]>wrote:

> Hey, all --
>
> As discussed today in the IRC meeting, and summarized in the log,
> there's an open question of which PHP version we should use as our
> minimum supported version.
>
> I've started a poll page here:
>
>    http://framework.zend.com/wiki/pages/viewpage.action?pageId=46792838
>
> to summarize the ideas floating around. Per our meeting, I'd like to
> open the floor for discussion. Post comments to the page, reply to this
> thread, or discuss on #zftalk.2. Let us know what _you_ want, and more
> importantly, _why_ -- try to give concrete reasons.
>
> Statements like "we'll kill adoption rates," or "it'll take us another
> six months to complete" may seem very true, but without evidence to back
> them up are speculation.
>
> On the flip side, statements like, "we have migrations already in place,
> and doing this will cost <X amount of time and money>", "our hosting
> provider has indicated they will not adopt 5.4 until <date in future>",
> "traits will reduce the amount of code we need to produce and maintain
> by <X percent>", etc. give everyone concrete data to work off of.
>
> I've locked the poll for now so we can have discussion and give time for
> everyone to inform themselves of all points of view. We'll open it up in
> a week for actual voting.
>
> --
> Matthew Weier O'Phinney
> Project Lead            | [hidden email]
> Zend Framework          | http://framework.zend.com/
> PGP key: http://framework.zend.com/zf-matthew-pgp-key.asc
>
> --
> List: [hidden email]
> Info: http://framework.zend.com/archives
> Unsubscribe: [hidden email]
>
>
>

[Pádraic Brady]

Hi all,

For me it's a straightforward decision, and while I respect others
will wish to adopt PHP 5.4 sooner rather than later, in my case
adoption will not occur until some time in 2013 when we've had the
opportunity and time to wait for vendor distributions, migrate code,
update training, and get around to planning any new greenfield
projects to coincide with those. At present, which is what counts
today, we've already fixed a path based on PHP 5.3 with an option to
start using ZF2 as it approaches release date so projects can be
rolled out as ZF2 gets over its initial few minor versions. If anyone
thinks that's a conservative approach, you should speak to businesses
who adhere only to LTS distributions.

Since I can't realistically hold up paying work, the most probable
solution for me will be to keep our existing plans in place: Quit ZF1
towards the Summer for upstarts and switch to another framework for
PHP 5.3 - which basically means Symfony 2 because anyone who mentions
CodeIgniter will be summarily thrown out the nearest window ;).

Ideally, I'd like to see ZF2 remain with PHP 5.3 as its minimum
requirement. Those agile enough to adopt PHP 5.4 quickly can still use
the exact same ZF2 and also leverage off PHP 5.4's new features in
whatever way they wish. We could also add some optional PHP 5.4 traits
to ZF2 for those programmers to use where it makes sense to do so.

All the best,
Paddy


On Wed, Feb 1, 2012 at 10:59 PM, Matthew Weier O'Phinney
<[hidden email]> wrote:

> Hey, all --
>
> As discussed today in the IRC meeting, and summarized in the log,
> there's an open question of which PHP version we should use as our
> minimum supported version.
>
> I've started a poll page here:
>
>    http://framework.zend.com/wiki/pages/viewpage.action?pageId=46792838
>
> to summarize the ideas floating around. Per our meeting, I'd like to
> open the floor for discussion. Post comments to the page, reply to this
> thread, or discuss on #zftalk.2. Let us know what _you_ want, and more
> importantly, _why_ -- try to give concrete reasons.
>
> Statements like "we'll kill adoption rates," or "it'll take us another
> six months to complete" may seem very true, but without evidence to back
> them up are speculation.
>
> On the flip side, statements like, "we have migrations already in place,
> and doing this will cost <X amount of time and money>", "our hosting
> provider has indicated they will not adopt 5.4 until <date in future>",
> "traits will reduce the amount of code we need to produce and maintain
> by <X percent>", etc. give everyone concrete data to work off of.
>
> I've locked the poll for now so we can have discussion and give time for
> everyone to inform themselves of all points of view. We'll open it up in
> a week for actual voting.
>
> --
> Matthew Weier O'Phinney
> Project Lead            | [hidden email]
> Zend Framework          | http://framework.zend.com/
> PGP key: http://framework.zend.com/zf-matthew-pgp-key.asc
>
> --
> List: [hidden email]
> Info: http://framework.zend.com/archives
> Unsubscribe: [hidden email]
>
>

--
Pádraic Brady

http://blog.astrumfutura.com
http://www.survivethedeepend.com
Zend Framework Community Review Team

--
List: [hidden email]
Info: http://framework.zend.com/archives
Unsubscribe: [hidden email]

[Derek Miranda]

I couldn't agree more.

Sent from my iPhone

On Feb 1, 2012, at 7:27 PM, Pádraic Brady <[hidden email]> wrote:

> Hi all,
>
> For me it's a straightforward decision, and while I respect others
> will wish to adopt PHP 5.4 sooner rather than later, in my case
> adoption will not occur until some time in 2013 when we've had the
> opportunity and time to wait for vendor distributions, migrate code,
> update training, and get around to planning any new greenfield
> projects to coincide with those. At present, which is what counts
> today, we've already fixed a path based on PHP 5.3 with an option to
> start using ZF2 as it approaches release date so projects can be
> rolled out as ZF2 gets over its initial few minor versions. If anyone
> thinks that's a conservative approach, you should speak to businesses
> who adhere only to LTS distributions.
>
> Since I can't realistically hold up paying work, the most probable
> solution for me will be to keep our existing plans in place: Quit ZF1
> towards the Summer for upstarts and switch to another framework for
> PHP 5.3 - which basically means Symfony 2 because anyone who mentions
> CodeIgniter will be summarily thrown out the nearest window ;).
>
> Ideally, I'd like to see ZF2 remain with PHP 5.3 as its minimum
> requirement. Those agile enough to adopt PHP 5.4 quickly can still use
> the exact same ZF2 and also leverage off PHP 5.4's new features in
> whatever way they wish. We could also add some optional PHP 5.4 traits
> to ZF2 for those programmers to use where it makes sense to do so.
>
> All the best,
> Paddy
>
>
> On Wed, Feb 1, 2012 at 10:59 PM, Matthew Weier O'Phinney
> <[hidden email]> wrote:
>> Hey, all --
>>
>> As discussed today in the IRC meeting, and summarized in the log,
>> there's an open question of which PHP version we should use as our
>> minimum supported version.
>>
>> I've started a poll page here:
>>
>>    http://framework.zend.com/wiki/pages/viewpage.action?pageId=46792838
>>
>> to summarize the ideas floating around. Per our meeting, I'd like to
>> open the floor for discussion. Post comments to the page, reply to this
>> thread, or discuss on #zftalk.2. Let us know what _you_ want, and more
>> importantly, _why_ -- try to give concrete reasons.
>>
>> Statements like "we'll kill adoption rates," or "it'll take us another
>> six months to complete" may seem very true, but without evidence to back
>> them up are speculation.
>>
>> On the flip side, statements like, "we have migrations already in place,
>> and doing this will cost <X amount of time and money>", "our hosting
>> provider has indicated they will not adopt 5.4 until <date in future>",
>> "traits will reduce the amount of code we need to produce and maintain
>> by <X percent>", etc. give everyone concrete data to work off of.
>>
>> I've locked the poll for now so we can have discussion and give time for
>> everyone to inform themselves of all points of view. We'll open it up in
>> a week for actual voting.
>>
>> --
>> Matthew Weier O'Phinney
>> Project Lead            | [hidden email]
>> Zend Framework          | http://framework.zend.com/
>> PGP key: http://framework.zend.com/zf-matthew-pgp-key.asc
>>
>> --
>> List: [hidden email]
>> Info: http://framework.zend.com/archives
>> Unsubscribe: [hidden email]
>>
>>
>
>
>
> --
> Pádraic Brady
>
> http://blog.astrumfutura.com
> http://www.survivethedeepend.com
> Zend Framework Community Review Team
>
> --
> List: [hidden email]
> Info: http://framework.zend.com/archives
> Unsubscribe: [hidden email]
>
>

--
List: [hidden email]
Info: http://framework.zend.com/archives
Unsubscribe: [hidden email]

[xoops]

Wish PHP 5.4+

On Thu, Feb 2, 2012 at 6:59 AM, Matthew Weier O'Phinney <[hidden email]>wrote:

> Hey, all --
>
> As discussed today in the IRC meeting, and summarized in the log,
> there's an open question of which PHP version we should use as our
> minimum supported version.
>
> I've started a poll page here:
>
>    http://framework.zend.com/wiki/pages/viewpage.action?pageId=46792838
>
> to summarize the ideas floating around. Per our meeting, I'd like to
> open the floor for discussion. Post comments to the page, reply to this
> thread, or discuss on #zftalk.2. Let us know what _you_ want, and more
> importantly, _why_ -- try to give concrete reasons.
>
> Statements like "we'll kill adoption rates," or "it'll take us another
> six months to complete" may seem very true, but without evidence to back
> them up are speculation.
>
> On the flip side, statements like, "we have migrations already in place,
> and doing this will cost <X amount of time and money>", "our hosting
> provider has indicated they will not adopt 5.4 until <date in future>",
> "traits will reduce the amount of code we need to produce and maintain
> by <X percent>", etc. give everyone concrete data to work off of.
>
> I've locked the poll for now so we can have discussion and give time for
> everyone to inform themselves of all points of view. We'll open it up in
> a week for actual voting.
>
> --
> Matthew Weier O'Phinney
> Project Lead            | [hidden email]
> Zend Framework          | http://framework.zend.com/
> PGP key: http://framework.zend.com/zf-matthew-pgp-key.asc
>
> --
> List: [hidden email]
> Info: http://framework.zend.com/archives
> Unsubscribe: [hidden email]
>
>
>

--

Taiwen Jiang (aka D.J.)

Build Xoops Engine
http://www.xoopsengine.org
web and mobile application platform

CTO for EEFOCUS.com
Leading social platform for electronics professionals

[ottovw]

In my oppinion, zf should stop moving the target and rather invest the
time in building a solid product.

Shipment is an important feature, either way ;)

Otto

2012/2/2 D. J. <[hidden email]>:

> Wish PHP 5.4+
>
> On Thu, Feb 2, 2012 at 6:59 AM, Matthew Weier O'Phinney <[hidden email]>wrote:
>
>> Hey, all --
>>
>> As discussed today in the IRC meeting, and summarized in the log,
>> there's an open question of which PHP version we should use as our
>> minimum supported version.
>>
>> I've started a poll page here:
>>
>>    http://framework.zend.com/wiki/pages/viewpage.action?pageId=46792838
>>
>> to summarize the ideas floating around. Per our meeting, I'd like to
>> open the floor for discussion. Post comments to the page, reply to this
>> thread, or discuss on #zftalk.2. Let us know what _you_ want, and more
>> importantly, _why_ -- try to give concrete reasons.
>>
>> Statements like "we'll kill adoption rates," or "it'll take us another
>> six months to complete" may seem very true, but without evidence to back
>> them up are speculation.
>>
>> On the flip side, statements like, "we have migrations already in place,
>> and doing this will cost <X amount of time and money>", "our hosting
>> provider has indicated they will not adopt 5.4 until <date in future>",
>> "traits will reduce the amount of code we need to produce and maintain
>> by <X percent>", etc. give everyone concrete data to work off of.
>>
>> I've locked the poll for now so we can have discussion and give time for
>> everyone to inform themselves of all points of view. We'll open it up in
>> a week for actual voting.
>>
>> --
>> Matthew Weier O'Phinney
>> Project Lead            | [hidden email]
>> Zend Framework          | http://framework.zend.com/
>> PGP key: http://framework.zend.com/zf-matthew-pgp-key.asc
>>
>> --
>> List: [hidden email]
>> Info: http://framework.zend.com/archives
>> Unsubscribe: [hidden email]
>>
>>
>>
>
>
> --
>
> Taiwen Jiang (aka D.J.)
>
> Build Xoops Engine
> http://www.xoopsengine.org
> web and mobile application platform
>
> CTO for EEFOCUS.com
> Leading social platform for electronics professionals

--
List: [hidden email]
Info: http://framework.zend.com/archives
Unsubscribe: [hidden email]

[Mike Willbanks]

Hello,


> to summarize the ideas floating around. Per our meeting, I'd like to
> open the floor for discussion. Post comments to the page, reply to this
> thread, or discuss on #zftalk.2. Let us know what _you_ want, and more
> importantly, _why_ -- try to give concrete reasons.
>

I am 100% against a *full* 5.4 implementation at this point in time.

Adoption of PHP 5.4 will take a long time; it is always the truth.  We have
been working on getting 5.3 into production on all of our web servers for
quite sometime.  We actually have 3 different versions of PHP in production
at this very second (not something I am extremely happy about).  One is 5.2
and the others are two different versions of 5.3.  I have a feeling even
although we would be ready for 5.4 internally as a development team.  I do
not foresee this getting into production for another 6+ months.  This means
that along with the re-work; we would be scrambling to implement servers
with 5.4 on them using a .0 release (I specifically will not upgrade until
a .1 or .2 release due to general complications).  At this specific point
in time we are looking for doing a beta release of software this month
running zf2.

I am alternatively for a 5.4 implementation of certain traits; when we do
get there it would aide in a reduction of code duplication as well as
allowing us to simplify maintenance - especially given the example of the
event manager.  I could see several other usages as well that would make
our implementation easier and sometimes quicker by the utilization of
traits.

Regards,

Mike

[robertbasic]

Good morning all!

Besides being +1 on Pádraic and Mike, don't forget that we decided on
faster releases, 18 months if memory serves right? Hopefully 5.4 will
have better adoption rates than 5.3, and after those first 18 months
we'll all be "aye" for 5.4 :)

Regards,
Robert

--
List: [hidden email]
Info: http://framework.zend.com/archives
Unsubscribe: [hidden email]

[Mike A]

Pádraic Brady wrote

For me it's a straightforward decision, and while I respect others
will wish to adopt PHP 5.4 sooner rather than later, in my case
adoption will not occur until some time in 2013 when we've had the
opportunity and time to wait for vendor distributions, migrate code,
update training, and get around to planning any new greenfield
projects to coincide with those. At present, which is what counts
today, we've already fixed a path based on PHP 5.3 with an option to
start using ZF2 as it approaches release date so projects can be
rolled out as ZF2 gets over its initial few minor versions. If anyone
thinks that's a conservative approach, you should speak to businesses
who adhere only to LTS distributions.

Since I can't realistically hold up paying work, the most probable
solution for me will be to keep our existing plans in place: Quit ZF1
towards the Summer for upstarts and switch to another framework for
PHP 5.3 - which basically means Symfony 2 because anyone who mentions
CodeIgniter will be summarily thrown out the nearest window ;).

Ideally, I'd like to see ZF2 remain with PHP 5.3 as its minimum
requirement. Those agile enough to adopt PHP 5.4 quickly can still use
the exact same ZF2 and also leverage off PHP 5.4's new features in
whatever way they wish. We could also add some optional PHP 5.4 traits
to ZF2 for those programmers to use where it makes sense to do so.

I concur.

There are two important facets of this bombshell: technical and business.

The technical facet primarily relates to the above answer and 5.4 in reality being way off for most hosts. There are others here much more knowledgeable who can test and debate technical considerations, like traits.

The business case has far wider implications. That could be why after word very quickly began to spread last night the IRC chat room swelled somewhat. Developers will be concerned about their income, and firms will have more than a little concern regarding plans already made but not implemented.

The case goes something like this inside the boardroom: "The thing on which the proposed systems will run has not come to market yet, has not faced widespread practical use, and the framework developers want to do WHAT?"

That's an uncertainty position.

Commercial decision makers don't like uncertainty. Going forward with uncertain technologies requires a gamble -- firms don't like gambling. Nor do investors, executives with legal responsibility, senior managers whose job could be on the line, outside advisors and consultants, or developers who rely on wages to feed their families. Technical passions belong outside the boardroom.

The notion that ZF2 should found upon 5.4 flies in the face of sound business judgement. A notion that it could drive 5.4 is a technical passion and will face aggressive opposition by competent managers. Any notion that ZF2 could come with 5.4 shiny new nice things, like traits, will diminish now accepted announcements over the past year or so that ZF2 founds upon PHP 5.3. By comparison, given the bombshell under discussion, Symfony 2 appears business-like, credible and superior -- each powerful magnetic factors to decision-makers.

Hey, it's not for me to judge the credibility of a development group producing the world's most superior framework. I wouldn't, and not just because I admire their technical genius and passion. But damage my credibility by convincing me ZF2 is for PHP 5.3, and as a result me convincing business decision-makers to run with it, and then changing the game, will guarantee me jumping up and down and making an awful fuss. I've made business decisions and commitments based upon ZF2 appearing for an accepted platform.

How serious is all this, from a business context I mean?

In some cultures, like China, contemplaters of the drastic decision before us would simply lose face: their credibility shot to pieces. In most business cultures the appearance of giving credence to a potential  decision will raise caution for a considerable time. These, and past difficulties that have alienated developers (like documentation and forms), do not help positively lift awareness for Zend Framework at a time when at least in part ZF2 should overcome negative aspects of its predecessor.

In an attempt to damp my outrage at the potential for ZF2 on 5.4 I've now spoken with some hosts. None of them will take on 5.4 this year. They will see the business case for 5.3 first. Perhaps an argument for virtual hosting exists to support the 5.4 debate. I am considering it. I've yet to see the case made though. To me, a roll-out on 5.4 is at least a year away due to host reticence. I'm not waiting another year to arrive at a position where I can begin project business decision-making processes based upon ZF2.

Is my business position in the majority? I'd be surprised if it isn't, and the majority sits at least 90+%.

In conclusion, for the business reasons mentioned, I say drop ZF2 for 5.4 discussion now to limit damage. Disregard technical arguments as immature in the business world. Kill the poll. Let the dust settle. Perhaps revisit the subject in March or later on the basis of exploring value-added features suiting 5.4 developers, but strictly as additions to ZF2 on 5.3. At the same time perhaps explore ZF3 for 5.4, even if it does run in conjunction with ZF2 development (the maintenance issue being a technical red herring). Or risk pleasuring competitors like Symfony by damaging the Zend Framework brand for a long time.

Can we now keep focussed upon ZF2 for PHP 5.3?

[Marco Pivetta]

First, I voted for 5.3, and that's not because I refuse awesomeness, but
because 5.4 has more dis-advantages than features right now (extensions
stability/support/distro packages/builds).

What I'd like to just note here is that we often bring the discussion to
the "hosting providers". I just wanted to cut that out, as zf2 applications
will most likely not land on shared hostings. This already happens on any
doctrine-based application, and as far as I know (as these apps require
caches that are not available on shared hosting solutions).

So I guess the problem is just with corporates. ("only" sounds like a joke
here :| )

Marco Pivetta

http://twitter.com/Ocramius

http://marco-pivetta.com



On 2 February 2012 09:02, Robert Basic <[hidden email]> wrote:

> Good morning all!
>
> Besides being +1 on Pádraic and Mike, don't forget that we decided on
> faster releases, 18 months if memory serves right? Hopefully 5.4 will
> have better adoption rates than 5.3, and after those first 18 months
> we'll all be "aye" for 5.4 :)
>
> Regards,
> Robert
>
> --
> List: [hidden email]
> Info: http://framework.zend.com/archives
> Unsubscribe: [hidden email]
>
>
>

[Artur Bodera]

On Wed, Feb 1, 2012 at 11:59 PM, Matthew Weier O'Phinney
<[hidden email]> wrote:
>
> As discussed today in the IRC meeting, and summarized in the log,
> there's an open question of which PHP version we should use as our
> minimum supported version.

I've re-thought the whole thing last night.

This statement still holds true: for many (including me), migrating
from zf1 to zf2 is a challenge by itself. It is new, it's cutting edge
etc. and it requires new version of PHP.

 * 5.3, which obviously for some linux distros was just recently
introduced, brings in (among others) extended Spl and namespaces.
 * 5.4 is 5.3 + traits (as a generalization on how zf2 could benefit from it).

Unfortunately 5.4 is at "feature freeze", so we won't get any new
gadgets besides what we already know. Short array syntax is irrelevant
for a framework. Same for array dereferencing. Closure rebinding is a
nice thing, but I doubt we'll have many use cases for that - which
means we can use it if phpver > 5.4 and rely on things like reflection
and closure rebuilding otherwise. There are also those small
convenience and performance changes like static method calling -
again, with little relevance to a framework.


Bottom line - 5.4 for zf2 is all about traits.

That said it's a decision between:
 1. Having traits and much lower zf2 usage rate.
 2. Having much higher zf2 usage rate because of current 5.3 market
penetration, but no traits.

I've been waiting for traits in php5 for like 10 years now and worked
my way around them. ZF2 in current trunk does the same thing - it just
"architects" around them. This means that traits currently fight for
the right to be called anything more than a "convenience". I know that
_proper_ usage of traits will make the codebase cleaner, BUT current
zf2 refactoring efforts already made it much cleaner and SOLID as
compared to zf1. Today, I see only a handful places _inside_ zf2 where
traits would be used and therefore - I see little overall value.

Outside of zf2 traits are (again) a convenience for developers to
add-on features, which helps a lot with complex concepts like
"pluggable" and "observable". This is a true value of that proposal,
but again - I do not feel that this (albeit very nice to have)
convenience can outweigh a heavily diminished ZF2 adoption rate due to
slow 5.4 market penetration.

Thanks for reading.

A.

--
      __
     /.)\   [phone]
     \(./   [hidden email]

--
List: [hidden email]
Info: http://framework.zend.com/archives
Unsubscribe: [hidden email]

[Shawn Stratton]

It may also be worth noting that someone could easily fork ZF2 to replace
the faux traits with the real deal and that fork could be reintegrated at a
later time, or with community effort maintained.  The reality is though
that by the time 5.4 becomes the adopted standard we will probably be
discussing ZF3.

Just my two cents.

Shawn Stratton
Systems Architect
Discovery Communications, Inc
On Feb 2, 2012 4:36 AM, "Artur Bodera" <[hidden email]> wrote:

> On Wed, Feb 1, 2012 at 11:59 PM, Matthew Weier O'Phinney
> <[hidden email]> wrote:
> >
> > As discussed today in the IRC meeting, and summarized in the log,
> > there's an open question of which PHP version we should use as our
> > minimum supported version.
>
> I've re-thought the whole thing last night.
>
> This statement still holds true: for many (including me), migrating
> from zf1 to zf2 is a challenge by itself. It is new, it's cutting edge
> etc. and it requires new version of PHP.
>
>  * 5.3, which obviously for some linux distros was just recently
> introduced, brings in (among others) extended Spl and namespaces.
>  * 5.4 is 5.3 + traits (as a generalization on how zf2 could benefit from
> it).
>
> Unfortunately 5.4 is at "feature freeze", so we won't get any new
> gadgets besides what we already know. Short array syntax is irrelevant
> for a framework. Same for array dereferencing. Closure rebinding is a
> nice thing, but I doubt we'll have many use cases for that - which
> means we can use it if phpver > 5.4 and rely on things like reflection
> and closure rebuilding otherwise. There are also those small
> convenience and performance changes like static method calling -
> again, with little relevance to a framework.
>
>
> Bottom line - 5.4 for zf2 is all about traits.
>
> That said it's a decision between:
>  1. Having traits and much lower zf2 usage rate.
>  2. Having much higher zf2 usage rate because of current 5.3 market
> penetration, but no traits.
>
> I've been waiting for traits in php5 for like 10 years now and worked
> my way around them. ZF2 in current trunk does the same thing - it just
> "architects" around them. This means that traits currently fight for
> the right to be called anything more than a "convenience". I know that
> _proper_ usage of traits will make the codebase cleaner, BUT current
> zf2 refactoring efforts already made it much cleaner and SOLID as
> compared to zf1. Today, I see only a handful places _inside_ zf2 where
> traits would be used and therefore - I see little overall value.
>
> Outside of zf2 traits are (again) a convenience for developers to
> add-on features, which helps a lot with complex concepts like
> "pluggable" and "observable". This is a true value of that proposal,
> but again - I do not feel that this (albeit very nice to have)
> convenience can outweigh a heavily diminished ZF2 adoption rate due to
> slow 5.4 market penetration.
>
> Thanks for reading.
>
> A.
>
> --
>       __
>      /.)\   [phone]
>      \(./   [hidden email]
>
> --
> List: [hidden email]
> Info: http://framework.zend.com/archives
> Unsubscribe: [hidden email]
>
>
>

[weierophinney]

-- Pádraic Brady <[hidden email]> wrote
(on Thursday, 02 February 2012, 12:27 AM +0000):

> For me it's a straightforward decision, and while I respect others
> will wish to adopt PHP 5.4 sooner rather than later, in my case
> adoption will not occur until some time in 2013 when we've had the
> opportunity and time to wait for vendor distributions, migrate code,
> update training, and get around to planning any new greenfield
> projects to coincide with those. At present, which is what counts
> today, we've already fixed a path based on PHP 5.3 with an option to
> start using ZF2 as it approaches release date so projects can be
> rolled out as ZF2 gets over its initial few minor versions. If anyone
> thinks that's a conservative approach, you should speak to businesses
> who adhere only to LTS distributions.

Thanks -- the above is exactly the sort of feedback I was looking for.

> Since I can't realistically hold up paying work, the most probable
> solution for me will be to keep our existing plans in place: Quit ZF1
> towards the Summer for upstarts and switch to another framework for
> PHP 5.3 - which basically means Symfony 2 because anyone who mentions
> CodeIgniter will be summarily thrown out the nearest window ;).
>
> Ideally, I'd like to see ZF2 remain with PHP 5.3 as its minimum
> requirement. Those agile enough to adopt PHP 5.4 quickly can still use
> the exact same ZF2 and also leverage off PHP 5.4's new features in
> whatever way they wish. We could also add some optional PHP 5.4 traits
> to ZF2 for those programmers to use where it makes sense to do so.

This is where I'm leaning today after the various discussions.

> On Wed, Feb 1, 2012 at 10:59 PM, Matthew Weier O'Phinney
> <[hidden email]> wrote:
> > Hey, all --
> >
> > As discussed today in the IRC meeting, and summarized in the log,
> > there's an open question of which PHP version we should use as our
> > minimum supported version.
> >
> > I've started a poll page here:
> >
> >    http://framework.zend.com/wiki/pages/viewpage.action?pageId=46792838
> >
> > to summarize the ideas floating around. Per our meeting, I'd like to
> > open the floor for discussion. Post comments to the page, reply to this
> > thread, or discuss on #zftalk.2. Let us know what _you_ want, and more
> > importantly, _why_ -- try to give concrete reasons.
> >
> > Statements like "we'll kill adoption rates," or "it'll take us another
> > six months to complete" may seem very true, but without evidence to back
> > them up are speculation.
> >
> > On the flip side, statements like, "we have migrations already in place,
> > and doing this will cost <X amount of time and money>", "our hosting
> > provider has indicated they will not adopt 5.4 until <date in future>",
> > "traits will reduce the amount of code we need to produce and maintain
> > by <X percent>", etc. give everyone concrete data to work off of.
> >
> > I've locked the poll for now so we can have discussion and give time for
> > everyone to inform themselves of all points of view. We'll open it up in
> > a week for actual voting.
> >
> > --
> > Matthew Weier O'Phinney
> > Project Lead            | [hidden email]
> > Zend Framework          | http://framework.zend.com/
> > PGP key: http://framework.zend.com/zf-matthew-pgp-key.asc
> >
> > --
> > List: [hidden email]
> > Info: http://framework.zend.com/archives
> > Unsubscribe: [hidden email]
> >
> >
>
>
>
> --
> Pádraic Brady
>
> http://blog.astrumfutura.com
> http://www.survivethedeepend.com
> Zend Framework Community Review Team
>
> --
> List: [hidden email]
> Info: http://framework.zend.com/archives
> Unsubscribe: [hidden email]
>
>

--
Matthew Weier O'Phinney
Project Lead            | [hidden email]
Zend Framework          | http://framework.zend.com/
PGP key: http://framework.zend.com/zf-matthew-pgp-key.asc

--
List: [hidden email]
Info: http://framework.zend.com/archives
Unsubscribe: [hidden email]

[Lars Kneschke-2]

Am 02.02.2012 00:05:55, schrieb Matthew Weier O'Phinney:
Hello!
> On the flip side, statements like, "we have migrations already in place,
> and doing this will cost <X amount of time and money>", "our hosting
> provider has indicated they will not adopt 5.4 until <date in future>",
> "traits will reduce the amount of code we need to produce and maintain
> by <X percent>", etc. give everyone concrete data to work off of.



Speaking for the Tine 2.0 project. Currently we require PHP 5.2 as minimum, but recommend PHP 5.3.
For the next major release, scheduled for the end of this year, we plan to raise the minimum PHP version to 5.3 as we like to switch to ZF2.

If ZF2 would require PHP 5.4 we would skip the update from ZF1 for the next release and maybe also for the next but one. I don't expect that most of our users/customers have switched to PHP 5.4 already until the end of the year.
--
​Lars KneschkeMetaways Infosystems GmbHPickhuben 2-4, 20457 Hamburg
Tel: [phone] - 521Fax: [phone] - 921Mobile: [phone]Email: [hidden email]: www.metaways.de
Metaways Infosystems GmbH - Sitz: D-22967 TremsbüttelHandelsregister: Amtsgericht Lübeck HRB 4508 AHGeschäftsführung: Hermann Thaele, Lüder-H. Thaele

[till-2]

On Thu, Feb 2, 2012 at 2:59 PM, Shawn Stratton <[hidden email]>wrote:

> It may also be worth noting that someone could easily fork ZF2 to replace
> the faux traits with the real deal and that fork could be reintegrated at a
> later time, or with community effort maintained.  The reality is though
> that by the time 5.4 becomes the adopted standard we will probably be
> discussing ZF3.
>
> Just my two cents.
>

I agree with that. And that's why I'd say going 5.4 is important and the
right step.

Others have mentioned already that zf2 apps will unlikely run on shared
hosting anyway. I don't see an issue with that. For most/all distributions
there are gonna be 5.4 builds really soon. Not getting the version is not
gonna be the issue.

Question is if you want to use something like dotdeb.

If you are semi-serious you are building your own packages anyway since
many distributions unfortunately tend to screw it up or lag hopelessly
behind.

Till

[Alessandro Pellizzari]

Il Thu, 02 Feb 2012 16:14:36 +0100, till ha scritto:

> If you are semi-serious you are building your own packages anyway since
> many distributions unfortunately tend to screw it up or lag hopelessly
> behind.

Building your own PHP packages means keeping pace with security, testing
all packages by yourself, maintaining the packages and resolve conflicts.
All on your own.

I use (debian) packages exactly to avoid such a burden. Debian devs are
much better than I am in packaging.

I think that if ZF2 will depend on PHP 5.4, only a small portion of
developers will adopt it until most of the major server distro will
deploy it. Sometimes next year, I think.

+1 for PHP 5.3

Bye.



--
List: [hidden email]
Info: http://framework.zend.com/archives
Unsubscribe: [hidden email]

[robertbasic]

On Thu, Feb 2, 2012 at 4:14 PM, till <[hidden email]> wrote:
>
> Others have mentioned already that zf2 apps will unlikely run on shared
> hosting anyway.

But it's not unusual to have a client with his own servers which are
running some older (older than 5.4 in this case) version of PHP.
Telling him "If you want to work with me, you'll need to update your
servers to 5.4" will most definitely leave me without that project.

--
~Robert Basic;
http://robertbasic.com/

--
List: [hidden email]
Info: http://framework.zend.com/archives
Unsubscribe: [hidden email]

[weierophinney]

-- Lars Kneschke <[hidden email]> wrote
(on Thursday, 02 February 2012, 03:01 PM +0000):

> Am 02.02.2012 00:05:55, schrieb Matthew Weier O'Phinney:
> Hello!
> > On the flip side, statements like, "we have migrations already in place,
> > and doing this will cost <X amount of time and money>", "our hosting
> > provider has indicated they will not adopt 5.4 until <date in future>",
> > "traits will reduce the amount of code we need to produce and maintain
> > by <X percent>", etc. give everyone concrete data to work off of.
>
>
>
> Speaking for the Tine 2.0 project. Currently we require PHP 5.2 as
> minimum, but recommend PHP 5.3.  For the next major release, scheduled
> for the end of this year, we plan to raise the minimum PHP version to
> 5.3 as we like to switch to ZF2.
>
> If ZF2 would require PHP 5.4 we would skip the update from ZF1 for the
> next release and maybe also for the next but one. I don't expect that
> most of our users/customers have switched to PHP 5.4 already until the
> end of the year.

Thanks, Lars -- good feedback.

--
Matthew Weier O'Phinney
Project Lead            | [hidden email]
Zend Framework          | http://framework.zend.com/
PGP key: http://framework.zend.com/zf-matthew-pgp-key.asc

--
List: [hidden email]
Info: http://framework.zend.com/archives
Unsubscribe: [hidden email]

[papayasoft]

And there are still a few of us out here who still live predominantly on
shared hosting - warts and all - that would love to stay current with
ZF2. Staying at 5.3 will definitely be a help for guys like us... well,
guys like me... ;-)

[First contribution to the list!]

=========================
David Weinraub
[hidden email]


On 2/2/2012 10:22 PM, Robert Basic wrote:

> On Thu, Feb 2, 2012 at 4:14 PM, till<[hidden email]>  wrote:
>>
>> Others have mentioned already that zf2 apps will unlikely run on shared
>> hosting anyway.
>
> But it's not unusual to have a client with his own servers which are
> running some older (older than 5.4 in this case) version of PHP.
> Telling him "If you want to work with me, you'll need to update your
> servers to 5.4" will most definitely leave me without that project.
>

--
List: [hidden email]
Info: http://framework.zend.com/archives
Unsubscribe: [hidden email]

[Tomáš Fejfar]

I hope NO ONE is considering dropping support for 5.3. But I can see very
real way how to support 5,3 and also make use of 5,4 features (= changing
files by some sort of build script).


On Thu, Feb 2, 2012 at 5:27 PM, David Weinraub - PapayaSoft <
[hidden email]> wrote:

> And there are still a few of us out here who still live predominantly on
> shared hosting - warts and all - that would love to stay current with ZF2.
> Staying at 5.3 will definitely be a help for guys like us... well, guys
> like me... ;-)
>
> [First contribution to the list!]
>
> =========================
> David Weinraub
> [hidden email]
>
>
>
> On 2/2/2012 10:22 PM, Robert Basic wrote:
>
>> On Thu, Feb 2, 2012 at 4:14 PM, till<[hidden email]>  wrote:
>>
>>>
>>> Others have mentioned already that zf2 apps will unlikely run on shared
>>> hosting anyway.
>>>
>>
>> But it's not unusual to have a client with his own servers which are
>> running some older (older than 5.4 in this case) version of PHP.
>> Telling him "If you want to work with me, you'll need to update your
>> servers to 5.4" will most definitely leave me without that project.
>>
>>
> --
> List: [hidden email]
> Info: http://framework.zend.com/**archives<http://framework.zend.com/archives>
> Unsubscribe: zf-contributors-unsubscribe@**lists.zend.com<[hidden email]>
>
>
>