Re: [jgen] All these changes in Joomla are killing me!

[Nick Savov]

Hi and welcome, Chris!

You're a few weeks behind, unfortunately/fortunately :)

Please see the following discussion, which started less than 3 weeks ago
and the last comment was about 5 days ago:
https://groups.google.com/d/msg/joomla-dev-general/bK4r4m3VEBg/h8j80SdE3YAJ

There are 300+ posts in there, so it's quite the read.

The changes from 1.6-3.0 are relatively minimal in comparison to 1.5 to
1.6. I had an extension that I wrote in 1.6 work without change in
1.7-3.0. Another one just needed one tiny thing changed that took less
than a minute to do. So in my opinion, Joomla is going a great job with
their API, at least since I started following things.

Anyway, hope you stick with it! Things are getting better! :)

Kind regards,
Nick

> I'm a (small) commercial extension developer, and I have to say that the
> constant rework of my code to keep up with changes in Joomla is a huge
> problem for me. It seems to me that the Joomla developers have lost sight
> of what Joomla is really about, and what real users really want. From my
> experience, the real users of Joomla are not using Joomla for fun, they
> are
> using it as a tool for some other purpose, mostly to run some kind of
> small
> businesses. From what my users tell me, updating to new levels of Joomla
> is
> an overhead that doesn't buy any benefit for their business. What they
> really do want is an infinite number of new features in my extensions.
> Many
> people are stuck on old versions of Joomla for one reason or another, so
> they want new features without having to update Joomla. I try to monitor
> my
> customers' usage to some extent, and I sense I am a very long way from
> being able to drop support for Joomla 1.5. So all my extensions have to
> work in Joomla 1.5, 1.6, 1.7, 2.5 and 3.0, and it has to be from a single
> code image because there is no way I can support multiple branches.
> Obviously, it is becoming more and more difficult, but it's what customers
> demand.
>
> My strategy for the future is to use less and less of the Joomla "API".
> Actually one of my extensions never got MVC'd, but it works very well
> indeed and was by far the easiest to migrate to Joomla 3.0. With the
> current MVC model now declared to be "legacy", this is actually a model I
> may well migrate all my extensions to. I want to isolate my code more and
> more from Joomla. It means writing a bit more code, but at least I know it
> will always work. I think it's really the only way I can keep going.
>
> I put "API" in quotes there because I don't really see Joomla as an API in
> the true sense. I've been a full-time commercial programmer for almost 40
> years so I have seen a few API's in my time. In my experience, an API is
> well documented, and long-lived. On the contrary, many Joomla functions
> require extensive reverse engineering in order to use them, and are of
> unknown longevity. In future I will just write my own code and keep the
> interface to Joomla as minimal as I can.
>
> --
> You received this message because you are subscribed to the Google Groups
> "Joomla! General Development" group.
> To view this discussion on the web, visit
> https://groups.google.com/d/msg/joomla-dev-general/-/e0N8JkEwxbUJ.
> To post to this group, send an email to
> joomla-de...@googlegroups.com.
> To unsubscribe from this group, send email to
> joomla-dev-gene...@googlegroups.com.
> For more options, visit this group at
> http://groups.google.com/group/joomla-dev-general?hl=en-GB.
>
>

[editor]

I think there are a lot of people in your situation - myself included.
Sadly it seems our concerns fall mostly upon deaf ears or they are met
with denial or derision - or both!

I am a big Joomla fan and I have huge respect to all those who have
worked on it. I cannot understand why they seem so determined to
over-develop Joomla into an early grave. I feel very sad about the whole
d�b�cle.

I felt particular empathy for what you say regarding avoiding Joomla
code in developing Joomla extensions. I am tending to do exactly the
same - for exactly the same reasons as you. Which rather begs the
question: "Why use Joomla at all?"

You may wish to study another thread running in joomla-dev-general
called "Joomla Exodus". Interesting to note that in its early posts, a
significant number of the core devs seemed in a state of denial and it
has taken quite a long time to convince them that there is an exodus at
all!

However my guess is the current exodus is nothing compared to what will
happen once J1.5 is no longer supported. In any event, the pain of J1.0
to J1.5 was enough for me. I suspect the same is true for many other
users and small commercial developers whose incomes depend on it.

Unless the "migration" from 1.5 to whatever the recommended LTS is at
the time is reasonably straightforward, I'm afraid many of us will be
forced, reluctantly, to jump ship.

Very sad, G.

[Nick Savov]

G., why don't you help out then and make things better? Joomla is run by
volunteers and anyone can contribute code to it. Why sit on the sidelines
when you can play?

I'd be happy to help you in getting started with contributing if you're
interested. Just let me know.

Kind regards,
Nick

[Gary Mort]

On Friday, November 23, 2012 9:41:39 PM UTC-5, ChrisG wrote:

My strategy for the future is to use less and less of the Joomla "API". Actually one of my extensions never got MVC'd, but it works very well indeed and was by far the easiest to migrate to Joomla 3.0. With the current MVC model now declared to be "legacy", this is actually a model I may well migrate all my extensions to. I want to isolate my code more and more from Joomla. It means writing a bit more code, but at least I know it will always work. I think it's really the only way I can keep going.

I put "API" in quotes there because I don't really see Joomla as an API in the true sense. I've been a full-time commercial programmer for almost 40 years so I have seen a few API's in my time. In my experience, an API is well documented, and long-lived. On the contrary, many Joomla functions require extensive reverse engineering in order to use them, and are of unknown longevity. In future I will just write my own code and keep the interface to Joomla as minimal as I can.

I suggest you look at Akeeba Backup...it seems to me to be an excellent model for how to bring sanity to your Joomla development as a commercial developer.

Akeeba Backup extends many of the Joomla classes and uses those extensions rather than the Joomla API directly.

So, instead of having

class MyComponentController extends JController  for example, he has

MyComponentController extends FOFController

and FOFController extends JController

This way if there are changes that breaks his code in JController, he can go ahead and and change his library once and all his code will work.  Either by adding removed code back to FOFController or whatever.

For myself, I'm working on my own library slightly differently:

class MyComponentController extends j/JController

In my code, at the top of each file I have:

namespace crafty/crafty

Taking some time to learn how to use namespaces and especially how to abuse them can help you moving forward as you can move all your function calls inside your own namespace and override them as needed.

As an example, I abhor JText::_(), it makes for really ugly code

echo JText::_('my string to be translated is here');

So instead I define some utility functions such as

namespace crafty/crafty

function l($string) 

{

   JText::_($string);

}

Now inside my code I can have 

echo l('my string to be converted'). ' ---- ' . l('another string to be converted');

It's handy, it's quick, and most importantly it reads logically.  lowercase l for language.   Since it doesn't pollute the global namespace all is good.

At the end of the day, I will agree, if your a commercial developer supporting multiple versions of Joomla - don't use the Joomla API directly - add a layer of abstraction so your code is cleaner and do all the messy stuff in your library.   Moreover, if you do that you can actually upgrade YOUR code to the latest Joomla API even if your running on an ancient 1.5 website.

[r...@osdcs.com]

Make your code standalone then with a simple bridge component to it.

Pretty sure that's jow jomres does it now. and possible nBill components

---

From: joomla-de...@googlegroups.com [mailto:joomla-de...@googlegroups.com] On Behalf Of ChrisG
Sent: Saturday, November 24, 2012 9:42 AM
To: joomla-de...@googlegroups.com
Subject: [jgen] All these changes in Joomla are killing me!

I'm a (small) commercial extension developer, and I have to say that the constant rework of my code to keep up with changes in Joomla is a huge problem for me. It seems to me that the Joomla developers have lost sight of what Joomla is really about, and what real users really want. From my experience, the real users of Joomla are not using Joomla for fun, they are using it as a tool for some other purpose, mostly to run some kind of small businesses. From what my users tell me, updating to new levels of Joomla is an overhead that doesn't buy any benefit for their business. What they really do want is an infinite number of new features in my extensions. Many people are stuck on old versions of Joomla for one reason or another, so they want new features without having to update Joomla. I try to monitor my customers' usage to some extent, and I sense I am a very long way from being able to drop support for Joomla 1.5. So all my extensions have to work in Joomla 1.5, 1.6, 1.7, 2.5 and 3.0, and it has to be from a single code image because there is no way I can support multiple branches. Obviously, it is becoming more and more difficult, but it's what customers demand.

My strategy for the future is to use less and less of the Joomla "API". Actually one of my extensions never got MVC'd, but it works very well indeed and was by far the easiest to migrate to Joomla 3.0. With the current MVC model now declared to be "legacy", this is actually a model I may well migrate all my extensions to. I want to isolate my code more and more from Joomla. It means writing a bit more code, but at least I know it will always work. I think it's really the only way I can keep going.

I put "API" in quotes there because I don't really see Joomla as an API in the true sense. I've been a full-time commercial programmer for almost 40 years so I have seen a few API's in my time. In my experience, an API is well documented, and long-lived. On the contrary, many Joomla functions require extensive reverse engineering in order to use them, and are of unknown longevity. In future I will just write my own code and keep the interface to Joomla as minimal as I can.

[Don]

So you're development strategy is to go back to 1.0 development days?

Sent from my iPhone

[r...@osdcs.com]

why would you say that? having your own api and briding it with joomla lets you bridge it to other systems too like drupal

then you only have to maintain the bridges when the cms upgrades

---

From: joomla-de...@googlegroups.com [mailto:joomla-de...@googlegroups.com] On Behalf Of Don
Sent: Sunday, November 25, 2012 12:21 PM
To: joomla-de...@googlegroups.com
Subject: Re: [jgen] All these changes in Joomla are killing me!

[Donald Gilbert]

Maintain a bridge or maintain your extension. I guess if that's what you prefer. But you really are opening yourself up to a lot of headaches using your own api. Because instead of just making your software better and adding features, you now have to make your software better and upgrade the bridges for any features you add, as well as keep up with all the changes of three separate platforms (assuming you made your extension available to WordPress, Joomla and Drupal). In that case, the idea that the API changes in Joomla are "killing you" becomes ludicrous. You made your bed, now lie in it.

OR, you could choose to use the Joomla API and keep up with the changes there. Everyone (in my experience) that complains about the API changes are worried about changes to the API that were clearly announced and discussed over 12 months ago. If they are unaware of those changes, I'm not sure anything would help them.

[Andrew Eddie]

On Sunday, 25 November 2012 06:58:21 UTC+10, Mr Goose wrote:

I think there are a lot of people in your situation - myself included.
Sadly it seems our concerns fall mostly upon deaf ears or they are met
with denial or derision - or both!

It's very hard not to take such sweeping statements personally.  The "deaf ears" is a red rag to a bull for me, but I'll get over it :)  I don't mind if you disagree with me, just don't tell me I'm not listening (because I will also try and look through the venting to see if there is something that can be improved, and there's usually a commit log to prove it after the fact).  Whatever the case, such statements aren't logically consistent with a contributor led, Open Source project.  Contribution is what drives and steers change.  You reap what you sow as the good book says.

 

I am a big Joomla fan and I have huge respect to all those who have
worked on it. I cannot understand why they seem so determined to
over-develop Joomla into an early grave. I feel very sad about the whole

d�b�cle.

It's very hard to know what you mean by "over develop" without giving any examples.  As for the early grave part, I actually somewhat agree.  The current architecture isn't going to survive much longer because, at a basic level, it's more or less the same principles that I was employing when writing Mambo 4.5.  The web has changed somewhat since Dec. of 2003.

 

You may wish to study another thread running in joomla-dev-general
called "Joomla Exodus". Interesting to note that in its early posts, a
significant number of the core devs

I'll just re-highlight my point in "that thread" that there is no such thing as "core devs" anymore.  That was an artefact of the Mambo and very early Joomla 1.0 days, ending more or less in January 2009 (give or take - it took the community about 6 months to get used to the idea).

 

seemed in a state of denial

I didn't see any of the "regular contributors" in any sort of state of denial.  Having a difference of opinion or holding different expectations does not translate into a state of denial.

 

and it
has taken quite a long time to convince them that there is an exodus at
all!

It was very clear that the reasons for why people thought there was an exodus were divided.  For example, it's not a bad reflection on the contributors of Joomla is people only need a simple WordPress site instead of using a Joomla stack.  An "exodus" in that case just means people don't need all the "stuff" that comes with the Joomla CMS.  All that means is the target market has shifted.

 

However my guess is the current exodus is nothing compared to what will
happen once J1.5 is no longer supported.

That more or less happened 2 months ago.  The horse has bolted and sky has not fallen (can't wait to see Skyfall though).  However, there is an opportunity for either volunteer led or commercial entities to pick up extended Joomla 1.5 on their own, and probably charge a premium for it.

 

In any event, the pain of J1.0
to J1.5 was enough for me.

Even though that *was* 4 years ago, a lot of people still agreed, including those contributing code to make thoroughly sure that upgrades, not migrations, would be the order of the day post 1.6.  This, unfortunately, is a reality of software development.

 

I suspect the same is true for many other
users and small commercial developers whose incomes depend on it.

Begs the question is contribution driven by passion or economics ... ahem.  Balance is required, but software developers and site implementors need to expect some change will happen sometime.

 

Unless the "migration" from 1.5 to whatever the recommended LTS is at
the time is reasonably straightforward, I'm afraid many of us will be
forced, reluctantly, to jump ship.

I believe that is the case talking to other real life Joomla users who have used jUpgrade, at least for the core. 

 

Very sad, G.

I get people need to vent, but those same people are in charge, to some degree, of their own destinies.

 

On 24/11/12 02:41, ChrisG wrote:
> I'm a (small) commercial extension developer, and I have to say that the
> constant rework of my code to keep up with changes in Joomla is a huge
> problem for me.

Can you refine what you mean by "constant"?  Are we talking every week?

 

It seems to me that the Joomla developers have lost
> sight of what Joomla is really about, and what real users really want.

It's possible, but as I said above, you really need to talk about Joomla "contributors", not just developers.  People who are not coders actually do steer the development process.  There is also ample opportunity for "you" to speak up about what's ticking you off.

 

>  From my experience, the real users of Joomla are not using Joomla for
> fun, they are using it as a tool for some other purpose, mostly to run
> some kind of small businesses. From what my users tell me, updating to
> new levels of Joomla is an overhead that doesn't buy any benefit for
> their business.

What "levels" are you referring to?  Security updates?  1.5 to 1.6 migrations?  Post 1.6 automatic upgrades?

 

What they really do want is an infinite number of new
> features in my extensions. Many people are stuck on old versions of
> Joomla for one reason or another, so they want new features without
> having to update Joomla.

I assume you are talking about updating from 1.5?

 

I try to monitor my customers' usage to some
> extent, and I sense I am a very long way from being able to drop support
> for Joomla 1.5.

Charge them accordingly for the trouble.  I would.

 

So all my extensions have to work in Joomla 1.5, 1.6,
> 1.7, 2.5 and 3.0, and it has to be from a single code image because
> there is no way I can support multiple branches.

Well, there are in fact many ways to support the multiple versions, but you probably have a preference for a small subset.  For as long as I've been involved in the project, we've been very upfront about trying to make it easy to support two versions (1.5 and 2.5, or 2.5 and 3.0).  If you want to go outside those parameters, you have to do the heavy lifting yourself.  It's unreasonable to demand that volunteers support more than is reasonable just so you can, presumably, keep your costs down.  I would also advise you to drop support for 1.6 and 1.7 as an inducement for customers to at least upgrade to 2.5.  If I were you, I would freeze the features of your 1.5 version and, indeed, put the in their own repository or branch.  Do your active development on 2.5+3 dual version packages.  

 

Obviously, it is
> becoming more and more difficult, but it's what customers demand.

Again, that's your call.  Charge them accordingly.  If you want to provide code examples, there are peers here that would help you work through the issues of streamlining your architecture.

 

> My strategy for the future is to use less and less of the Joomla "API".
> Actually one of my extensions never got MVC'd, but it works very well
> indeed and was by far the easiest to migrate to Joomla 3.0.

That's an option, but then you miss out on consistent form handling, form modifiers, and layout overrides just to name a few.  If *your* customers don't want those features that's fine.  Many other users do and will pass you over in the process.

 

With the
> current MVC model now declared to be "legacy",

We have probably failed to properly explain this, because there is quite a story behind it and what "legacy" actually means.  Suffice to say the fact that "legacy" is in the name (I actually suggested different names, but that's beside the point) doesn't mean you can't or should not use those classes.  The core CMS hasn't even worked out how to ween off them yet.

 

this is actually a model
> I may well migrate all my extensions to. I want to isolate my code more
> and more from Joomla. It means writing a bit more code, but at least I
> know it will always work. I think it's really the only way I can keep going.

I would suggest to you that you engage on the appropriate development lists and "talk code".

> I put "API" in quotes there because I don't really see Joomla as an API
> in the true sense.

I beg to differ.  An API exists - some parts better and worse than others but it exists nonetheless.

 

I've been a full-time commercial programmer for
> almost 40 years so I have seen a few API's in my time. In my experience,
> an API is well documented,

Documentation is contributed by volunteers.  There is lots out there.  I wish every developer that uses Joomla would give the project a few hours a month to contribute to the documentation of the code they use for free.  We'd have it knocked over in a very short period of time I'm sure.

 

and long-lived.

How long is "long"?  Joomla 1.0's API essentially lasted from Feb. 2005 (Mambo 4.5.2) and continued on past the release of Joomla 1.5 in Jan. 2008 via the legacy layer.  Joomla 1.5 only just made a soft crossing of end-of-life in Sep. this year so the longevity of that API was some seven years.  Joomla 1.6 was released in Jan. 2011 and this was really the first time ever we had a major breakage in the PHP code without the buffer of a legacy layer.  That's still 6 years of relative peace if you wanted it.  Since then, there have been relatively minor backward compatibility issues arise and most are either fixed, or developers just make the appropriate changes.  For most of them, it's not a big deal.  However, if it is a big deal, the best way to handle it is explain the problem with "code".  Complaining non-specifically about changes is not helpful.

 

On the contrary, many Joomla
> functions require extensive reverse engineering in order to use them,

I feel your pain, and in such cases I take some extra time to document what I've learned so the next poor sucker doesn't have such a hard time about it.  Pages such as the following were born out of similar frustration that I felt:

https://github.com/joomla/joomla-platform/blob/staging/docs/manual/en-US/chapters/packages/http.md

https://github.com/joomla/joomla-platform/blob/staging/docs/manual/en-US/chapters/packages/input.md

https://github.com/joomla/joomla-platform/blob/staging/docs/manual/en-US/chapters/packages/log.md

https://github.com/joomla/joomla-platform/blob/staging/docs/manual/en-US/chapters/packages/profiler.md

I would welcome the help to document more API packages.

 

> and are of unknown longevity.

I'm not quite sure what you are expecting (a tag that says "no changes for 10 years?") but we take care to document deprecated well.  We miss a few things and maybe make the wrong call every now and then (who is perfect after all) but overall you know when something is going bye-bye.

 

In future I will just write my own code
> and keep the interface to Joomla as minimal as I can.

I understand where you are coming from, however my personal practice has been to jump and and improve what's there so everyone can benefit, not just me, or my employer, or my customers.  I've always sought to "extend" Joomla, not replace it with bespoke code or other frameworks (Zend, PEAR, Nooku, Symfony, whatever).  There are also lots of ways you can buffer changes in the core, but not throw the baby out with the bathwater (a lot of people have done a lot of work to the autoloader just to make this happen).

To conclude, if you have suggestions about how to improve the Platform (if you don't know the distinction between what is and isn't the platform, I'm happy to explain), jump on this list:

https://groups.google.com/forum/?fromgroups#!forum/joomla-dev-platform

If you want to talk about how to improve (or fix) the CMS, try this one:

https://groups.google.com/forum/?fromgroups#!forum/joomla-dev-cms

And if you just have a general problem with extensions and using Joomla (like how do I deal with so many versions), just on this one:

https://groups.google.com/forum/?fromgroups#!forum/joomla-dev-general

The reality is you have to take time to engage in the community to get something back.  I know that's tough as a lone developer and I know what it takes to develop very large and complicated extensions (and to be brutally honest, I've probably dealt with a lot more change in Joomla over the years than most).  But if all you do is run with the hand you are dealt, there's not much anyone can do for you (if you are a regular contribution, my apologise in advance - your handle is just not familiar to me).

Regards,

Andrew Eddie

[Rigó Tamás]

Unfortunately, the biggest problem is really the Joomla documentation continued backwardness. It would be important in the documentation and the code simultaneously published.

Important even the demo code publishing. If you want to understand how to working in Jommla  now everything should be reverse engeenering in Joomla Library.

 

Now you can prepare to be a v3.0, but the v2.5 module not running J3.0. There is little information on what is changing and why! This is a big problem during the development of the individual modules.

I was surprised to see that my J2.5 module does not run on the new system. These functions change the client does not pay. Upgrade to Joomla the changes to prepare for free work.

 

Why is it good? What is the goal? Joomla is a small Microsoft who dreamed one world, but the world does not need it so fast?

 

Tamas

 

From: joomla-de...@googlegroups.com [mailto:joomla-de...@googlegroups.com] On Behalf Of Andrew Eddie
Sent: Sunday, November 25, 2012 1:00 PM
To: joomla-de...@googlegroups.com
Subject: Re: [jgen] All these changes in Joomla are killing me!

 

On Sunday, 25 November 2012 06:58:21 UTC+10, Mr Goose wrote:

--

You received this message because you are subscribed to the Google Groups "Joomla! General Development" group.

To view this discussion on the web, visit https://groups.google.com/d/msg/joomla-dev-general/-/FYZ6rnUxK7UJ.

[Bakual]

Am Sonntag, 25. November 2012 14:54:04 UTC+1 schrieb ChrisG:

Maintaining upward compatibility for 10 years would be a minimum for me. I'd prefer more. 

 

Just a thought about this. In informatics, 10 years is like infinity. The world wide web is about 20 years old, and you're talking about 10 years backward compatibility? Knowbody can tell you today what will be in 10 years, how the internet will look. What I can tell you is that in 10 years, knowbody will use a component which was developed today, it will be outdated by at least 8 years. So 10 years? Not gonna happen.

[Bakual]

Actually, the documentation about backward compatibility issues between Joomla 3.0 and 2.5 is quite good, and was there when Joomla 3.0 released.

http://docs.joomla.org/Potential_backward_compatibility_issues_in_Joomla_3.0_and_Joomla_Platform_12.1

It was even linked in the release notes. So this shouldn't be an issue.

Also Joomla 3.0 is only a short term release meant for developers to adjust their extensions. We still have a year till the next long term will be released. Ample time I think. Also it's possible to write code that runs in Joomla 2.5.5 and Joomla 3.0.

To post to this group, send an email to joomla-d...@googlegroups.com.
To unsubscribe from this group, send email to joomla-dev-general+unsub...@googlegroups.com.

[Andrew Eddie]

On 25 November 2012 23:54, ChrisG <chris....@gmail.com> wrote:

Andrew, thank you so much for taking the time. Gosh, I shall have to watch my words more carefully here in future.

Avoid hyperbole.  It reduces confusion (I just got slapped for doing just that in another thread myself, hehe) :)

 

Perhaps "constant" wasn't a good choice, but extension development is not a full-time occupation for me, and in my busy life, these API changes come around more quickly than I would like. I personally do not welcome the 6 month treadmill. I live in dread of what the next release will bring. I really miss the stability of Joomla 1.5.

I understand.  The unfortunate side effect of that is that there was absolutely no innovation allowed in the CMS over the life of 1.5 and that caused a lot of good developers in our community to loose interest.

 

About speaking up, yes, I know that as one who doesn't contribute very much I shouldn't criticise.

Not quite so.  Speaking up is contributing.  Constructive criticism is always welcome.  

 

I don't have time to follow the groups.

Ok, I hear this a lot - from small operations probably like yourself, and from well-known, multi-million dollar businesses.  As a business person that translates into "I don't have time to invest myself in the thing that makes my business possible".  If you were in the transport business, it's akin to saying I can't afford to take my trucks off the road to have them serviced.  If you rely on Joomla for your income (or anything else for that matter), then it is always worth your time to, at the very least, keep up with how it is changing.

 

I don't know how you do it.

Here, I think, is the take home for me.  We have a lot of information floating around if you know where to look.  I'd love to work with people to see if we can made things easier and reduce the number of channels you have to monitor.

 

But I suspect I speak for many in my position who also don't have enough time, and who find the rapid pace of change unwelcome. My sympathy really lies with the "real users" - the rather non-technical masses who use Joomla as a tool to run a small business.

It would be a safe bet that most change in the Joomla CMS is driven by user needs.  Remember the debates about how weak the ACL was in 1.5 and how people were walking away from Joomla in droves because of it?

 

By upgrading Joomla levels, I mean any upgrade that requires manual intervention and/or hunting for new versions of extensions (which may or may not exist). Small business owners have a million other priorities and for many of them it might be quite a technical challenge that requires them to set aside a whole day or even more. So they put it off for as long as they possibly can, which only makes it even harder.

Unfortunately that's life dealing with software, and most things mechanical for that matter.  People also have this blind notion that once a web site is done, it shouldn't cost them any money to maintain.

 

Joomla makes it quite easy for non-technical users to get started with a new site, and it's quite easy for the site to evolve and become quite sophisticated. But upgrading it to a new Joomla level, especially if they need to jump a few levels, can be much more challenging for non-technical users who are just not familiar with all the techniques that you and I take for granted. 

That's a fair comment.  But the web is also getting more complicated to use.  It's a bit like saying "the traffic wasn't nearly as bad around here when we started business, now it's a pain to get to work".  Who's fault is that?

 

Yes, I should drop support for 1.6 and 1.7. No argument there, although I know there will be whinging and complaining so I've been guilty of putting that off too…

Sucks I know.  Just give your customers a bit of warning.  My axiom is that the customers that walk away because of things like that I never really needed in the first place (high maintenance customers cost me too much money).

 

I take your point about documentation and again I shouldn't criticise if I don't contribute. But where I come from developers need to produce some kind of documentation. Churning out undocumented code is really not playing fair. Couldn't there be some kind of oversight that just won't accept that as a practice?

There is ... now, at least on the platform.  We ensure the DocBlocks that go to making the automated API reference are complete and in good condition.  In fact, you can't even merge contributions without them (our code sniffer takes care of that).  For new packages or new chunks of work we general require some level of "how does one use this" type of information, even if it's only in the pull request description.  So going forward, we are in a better place.  But ... we have an enormous backlog that needs to be caught up somehow, and this is the main problem.  In addition, there is a veritable wealth of information buried in docs.joomla.org, but it is just that - buried.  It's difficult to find.  I'd love to have a small team of people volunteer to scape the best pages from the wiki and put that material in our developer manual in the repo.

 

After all, there are programming standards for Joomla. Should the project be accepting undocumented contributions? And I mean proper descriptive documentation, like yours, not a generated list of functions and parameters.

Documentation like mine is a bit of an art form.  Not everyone can write tutorials, and I am one of those crazy folks that actually loves documenting the code I write :)  Anyone can help with this effort though.  It's as simple as finding a wiki page that is up to date and has value and letting us know about it.  I'll probably personally take the time to scape it and put it in the correct place.  What I don't have time to do is going to find these diamonds in the sand.

 

Some Joomla API's have been around for a while, but some that were introduced in 1.6 are already gone in 3.0. It's the not knowing that makes me reluctant to trust any of them. Maintaining upward compatibility for 10 years would be a minimum for me. I'd prefer more. 

There are many reason why code does change and needs to change.  Some times it's because of PHP issues, other times it's because we need to break something to move forward.  1.6 is a good example of the version we had to break to move forward, and it has been very successful at doing that.  The Platform is going to continue to get rid of a lot of cruft.  Another way of looking that is that some very talented and experienced developers are making changes so maybe it's worth learning the reason why and improving your own skills.  Keeping up with incremental changes in Joomla will save you time and pain compared to leaving it to the last minute - I'm sorry, but that's not something you can blame the volunteers for (accidental mistakes aside).

 

Thanks for the links, despite much Googling I hadn't found your documentation pages, and I will try to follow and contribute to some of the groups. I've been around since Mambo, just not very vocal.

We'll keep working on how to make the information we know is out there more accessible.

Thanks.

Regards,

Andrew Eddie 

[Andrew Eddie]

On 26 November 2012 03:31, Bakual <werbe...@bakual.ch> wrote:

Actually, the documentation about backward compatibility issues between Joomla 3.0 and 2.5 is quite good, and was there when Joomla 3.0 released.

http://docs.joomla.org/Potential_backward_compatibility_issues_in_Joomla_3.0_and_Joomla_Platform_12.1

It was even linked in the release notes. So this shouldn't be an issue.

This is a good reference but the developers that needed that information still seemed to miss it.  If people have ideas on how this information can be publicised let me/us know.

 

Also Joomla 3.0 is only a short term release meant for developers to adjust their extensions. We still have a year till the next long term will be released. Ample time I think. Also it's possible to write code that runs in Joomla 2.5.5 and Joomla 3.0.

If I was still in business my strategy would be to downgrade my support for 1.5 to security only.  Let your customers know that new features will only be added to 2.5+3.x versions of your extensions.  If you absolutely have to support new features in 1.5, charge them for it - charge them a lot to make it worth your while.  You may even want to require PHP 5.3 as a minimum (I would).

Regards,

Andrew Eddie 

[Michael Babker]

We've been trying our best to share API changes as they were coming.  Back at JAB in May, I had a fairly packed room to (weakly, I'm not much of a public speaker ;-) ) discuss some of the larger API changes between Platform 11.4 and 12.1, still weeks before we had a 3.0 Alpha.  I know the link to those slides and the video of the talk has been on their site since then, and that link to the docs site was passed around on Twitter that same weekend (I contributed to it, and I forgot to put it in slides).  So, all I can add here is everyone needs to make a little effort; we can make all the resources in the world available, but they're no good if folks refuse to take 10 minutes every week to just look for them and see if there's anything relevant to them.  I certainly don't read every message from the mailing groups, but I do skim to make sure I don't miss something important to me.

From: Andrew Eddie <mamb...@gmail.com>
Reply-To: <joomla-de...@googlegroups.com>
Date: Sunday, November 25, 2012 3:31 PM
To: "joomla-de...@googlegroups.com" <joomla-de...@googlegroups.com>
Subject: Re: [jgen] All these changes in Joomla are killing me!

[Paul Orwig]

A lot of people keep saying there is not enough information out there or it's too hard to find, and a lot of other people keep saying the information is out there, but you just have to know where to look and make the effort to keep up.

What if we had a prominent web page (or short group of web pages) that was a "project news/get involved/developer resources" type of portal where we aggregated links to some of the content that currently is served from a number of different sites: announcements (www site), blogs, events (community site), feature stories, help wanted, etc.(JCM), and developer resources (mailing lists, Wiki, developer site). Some of the links on this portal would be relatively static, and some would be regularly updated. Maybe we could set up targeted opt-in newsletters for different types of users, maybe more RSS feeds, etc.

Is there interest in this type of concept? It has been talked about before but it hasn't ever happened. I don't think it would be that hard to pull off, and if there's still interest I will do what I can to move it forward in 2013. In my mind, the biggest pieces we'd need are a prominent place for the portal pages (redesigned community site home page?), some guidelines for how/where different types of content should get published (so the portal knows where to link to), and a cross-functional team to decide what content belongs on the portal pages, and to manage those portal pages.

What do you all think about this concept?

Thanks,

paul

--
You received this message because you are subscribed to the Google Groups "Joomla! General Development" group.

To post to this group, send an email to joomla-de...@googlegroups.com.
To unsubscribe from this group, send email to joomla-dev-gene...@googlegroups.com.

[Andrew Eddie]

We have an existing developer portal (http://developer.joomla.org). The landing page is quite informative and there is a lot of information on that site that is easy to find (that is, of course, if you actually visit it).  The only thing that is a little out of date is the Platform Manual link because we've changed to a new system in the last month and are still wiring it up.  The styling on the "read more" link is ghastly, but whatevs - that's not important.

The one place I think we can improve is possibly have more news (http://developer.joomla.org/news.html).  At the moment it's really the bare minimum, which is not bad, but could be better.  I'm open to suggestions to how to make that news better, or any other part of the site for that matter.  I'm open to people volunteering to make that site better (talking about content, not design).

One thing that comes to mind after Michael's comments is people being able to give us a wave if they've talked about development and put some slides up.  I see no problem with advertising that kind of thing.  We don't have any sort of "submit some news" mechanism.  That, I think, would be useful subject to having the right tool to collect the information, and a small team to triage the submissions.

The bigger problem I have, as I've mentioned, is that there is a lot of information in the wiki that isn't getting found.  I'd prefer to see that in the Platform repository forming part of the "Platform Manual".  People also have this hangup about "official" stuff and the Platform Manual looks more official (even though it isn't, but let's keep that a secret between you and me).

Quite a few places people can help out.  Contact me if you are interested.

Regards,

Andrew Eddie

Regards,
Andrew Eddie
http://learn.theartofjoomla.com - training videos for Joomla developers

[Michael Babker]

+1 on making better use of the developer portal.  We have the ability already to consolidate this information and make it easier for folks to find, we just aren't doing a good job at it.  It's hard for me to say "hey, if you want some of my notes about the API changes made from January to May in the Platform code, you can search for my slides on SlideShare, find the video of the talk on YouTube, or use resource X" and expect folks to find what they want with ease.  A consolidated list would certainly make life easier, and one that is open for user contribution.  If it means adding one or two folks to the dev site to work on those resources, count me in.

And we should be nudging one another to document things as they are changed.  Silly me just added notes about changes I made to Smart Search to improve the multi-database support to the B/C page on the docs site.  I'll say my 50 Hail Mary's this evening.

[Rigó Tamás]

Which side do you think I should watch that tracks the changes? I'm watching the developer.joomla.org but I really can not find it when you changed it to only get a taste of the list

 

Tamas

[Andrew Eddie]

On Monday, 26 November 2012, Rigó Tamás wrote:

Which side do you think I should watch that tracks the changes? I'm watching the developer.joomla.org but I really can not find it when you changed it to only get a taste of the list

 

+1 on making better use of the developer portal.  We have the ability already to consolidate this information and make it easier for folks to find, we just aren't doing a good job at it.  It's hard for me to say "hey, if you want some of my notes about the API changes made from January to May in the Platform code, you can search for my slides on SlideShare, find the video of the talk on YouTube, or use resource X" and expect folks to find what they want with ease.  A consolidated list would certainly make life easier, and one that is open for user contribution.  If it means adding one or two folks to the dev site to work on those resources, count me in.

 

And we should be nudging one another to document things as they are changed.  Silly me just added notes about changes I made to Smart Search to improve the multi-database support to the B/C page on the docs site.  I'll say my 50 Hail Mary's this evening.

 

From: Andrew Eddie <mamb...@gmail.com>
Reply-To: <joomla-de...@googlegroups.com>
Date: Sunday, November 25, 2012 5:02 PM
To: "joomla-de...@googlegroups.com" <joomla-de...@googlegroups.com>
Subject: Re: [jgen] All these changes in Joomla are killing me!

 

We have an existing developer portal (http://developer.joomla.org). The landing page is quite informative and there is a lot of information on that site that is easy to find (that is, of course, if you actually visit it).  The only thing that is a little out of date is the Platform Manual link because we've changed to a new system in the last month and are still wiring it up.  The styling on the "read more" link is ghastly, but whatevs - that's not important.

 

The one place I think we can improve is possibly have more news (http://developer.joomla.org/news.html).  At the moment it's really the bare minimum, which is not bad, but could be better.  I'm open to suggestions to how to make that news better, or any other part of the site for that matter.  I'm open to people volunteering to make that site better (talking about content, not design).

Hey. Have a look at:

http://developer.joomla.org/code.html

And look at the links for the latest commits and the latest pull requests. Otherwise look in the news for the change logs of new releases of the platform. 

Hope that helps. 

Regards

Andrew Eddie 

--

[Allon Moritz]

I don't know why people always refer to the changes in the API. The change in the UI was much bigger and caused a rewrite of the whole UI of my extensions. The API changes were adapted within minutes. I hope somebody will focus on the migration path for the next time when the UI completely changes.

my two cents...

[Naouak]

There is a lot of problems regarding migration on Joomla 3.x :

- New markup for Admin template

- Changing from mootools to jquery (some will says that mootools is still supported, but let's face it, you don't want both on the same page)

- A lot of API change (class renaming, function renaming, stuff removed, stuff completely different).

All of them are long to process and not just a find replace like some like to say.

The major problem with the approach of joomla is that these changes are not a one-time change but something that happens every major version.

If you could break backward compatibility once and not every version you would be a lot smarter for developpers. I prefer doing a major rewrite once and not doing a lot of small rewriting every 6 months.

Naouak, Grade 2 de Kobal.
Site web: http://www.naouak.net

On Mon, Nov 26, 2012 at 8:58 AM, ChrisG <chris....@gmail.com> wrote:

. I managed to get JHtmlSliders to work, but it took a while to figure out how to fix the css. I still can't figure out how to control the icon on a custom toolbar button, so for now some of my buttons have no icon at all. I haven't had any customer feedback on my 3.0 versions yet, but I don't think they are going to be very happy bunnies.

[Andrew Eddie]

On 26 November 2012 16:39, Allon Moritz <allon....@gmail.com> wrote:

I don't know why people always refer to the changes in the API. The change in the UI was much bigger and caused a rewrite of the whole UI of my extensions. The API changes were adapted within minutes. I hope somebody will focus on the migration path for the next time when the UI completely changes.

my two cents...

/me does a little twinkle.  However, the process involved a lot of community consultation - more than anything we've ever done as a project and I think the end result is worth the pain.  The downside is the lack of "Bootstrap for developers that can't CSS to save themselves" ... but don't look at me to write that one - I need that for me :)

Regards,

Andrew Eddie 

[Andrew Eddie]

On 26 November 2012 19:31, Naouak <tar...@gmail.com> wrote:

There is a lot of problems regarding migration on Joomla 3.x :

- New markup for Admin template

- Changing from mootools to jquery (some will says that mootools is still supported, but let's face it, you don't want both on the same page)

- A lot of API change (class renaming, function renaming, stuff removed, stuff completely different).

All of them are long to process and not just a find replace like some like to say.

That totally depends on the sate of --your-- code.  As I said to Chris, and as we've told people for virtually every new version (this is NOT a new conversation), if you keep up with the incremental changes, the job is not so big.  If you write extensions with your own conventions and such, you are on your own and have only yourself to blame if it goes all wrong.  If you are wondering why your code is so hard to upgrade, post a few gist's - I'm sure you'll get some pointers.

Regards,

Andrew Eddie

[brian teeman]

On Monday, 26 November 2012 09:41:08 UTC, Andrew Eddie wrote:

The downside is the lack of "Bootstrap for developers that can't CSS to save themselves" ... but don't look at me to write that one - I need that for me :)

Have you looked at the slides from Andrea's presentation at the Joomla World Conference http://www.slideshare.net/andytarr/bootstrap-for-extension-developers-jwc-2012 "Bootstrap for developers" 

[Naouak]

The problem is not the difficulty, it's the time it take.

My company have bet on joomla 3 years ago on 1.5.

We made all of our websites with it.

As each of our websites have customs needs, we made a lot of extensions( something like 50+).

Some of these extensions have to work at any cost (payment methods, payed services, etc.) and when you have to modify it that means going through any possible test to ensure that it works fine.

If you have one extension, okay it's manageable but when you are 3 and have a lot of extensions to manage it's just a big investment.

My point is that breaking backward compatibility is something that have heavy consequences for some of your users (I don't think we are alone in working like that) and when it's repeated every 6 months it's becoming to expensive for us to follow.

Naouak, Grade 2 de Kobal.
Site web: http://www.naouak.net

--

You received this message because you are subscribed to the Google Groups "Joomla! General Development" group.

[Andrew Eddie]

On 26 November 2012 20:45, Naouak <tar...@gmail.com> wrote:

My point is that breaking backward compatibility is something that have heavy consequences for some of your users (I don't think we are alone in working like that) and when it's repeated every 6 months it's becoming to expensive for us to follow.

Breaking of backward compatibility is not repeated every six months.  Currently it IS allowed every 18 months but there is a proposal to extend this to every 2 years (the "major" increment).

I do sympathise - 50 extensions is a lot for three people, but it's also your decision to make that many. What that says is you need to be extremely smart about how you design and maintain those extensions.  You would want to invest heavily in unit testing.  You can also review what you are doing an see if there is anything repetitive, and then consider where you can contribute code to the Joomla core to offload the long term maintenance.  To that end, there is a discussion about a commerce package on the Platform list at the moment.  You might want to get involved.

https://github.com/joomla/joomla-platform/pull/1706

Regards,

Andrew Eddie

[Naouak]

I wanted to do some stuff for joomla. At first, before github was used, on the website you were supposed to sign a paper which I couldn't found anywhere.

When github was introduced, I did a pull request (10 months ago) and it is still waiting for someone to do something with it ( https://github.com/joomla/joomla-cms/pull/90#commit-ref-4c9c45e ).

Now, we almost don't use joomla core components and so our stuff is not really relevant for core. I want to share some of our code with communities to help maintain it but it's company property and it's hard to make them understand what we can gain from open sourcing some stuff.

For the 6 months, it was mainly for 1.6,1.7 and then 2.5 : We decided to use 1.6 when it was out and then the LTS was introduced and we got a bit screwed by the backward compatibility breaking between these versions (some of our sites were in 1.7, some in 1.6 and we didn't have time yet to do the migration from 1.6 to 1.7 or 2.5).

We don't even know if we will change from 2.5 to 3.x any day as it would break again a lot of stuff.

I barely have time to read some mails from this mailling list, I can't follow github repositories and other mailling list.

If you need a way to communicate with your developpers and user, take a look at how concrete 5 is doing it, they are doing very well and people don't have to seek information.

Naouak, Grade 2 de Kobal.
Site web: http://www.naouak.net

--

[Rigó Tamás]

I look at the link and very happy. In the relation to it is a big mistake that is difficult to reach out to your available materials. I watch a lot of presentations (slideshow)  never heard of. How these will be published? Where can I get information about the events of what was said there?

For me it a little gas in the middle of Europe (Hungary Joomla User Group does not work not even respond to their side of the broken 2 days ago - J1.5 was used – http://joomla.org.hu )

 

Tamas

 

From: joomla-de...@googlegroups.com [mailto:joomla-de...@googlegroups.com] On Behalf Of brian teeman
Sent: Monday, November 26, 2012 11:30 AM
To: joomla-de...@googlegroups.com
Subject: Re: [jgen] All these changes in Joomla are killing me!

 

On Monday, 26 November 2012 09:41:08 UTC, Andrew Eddie wrote:

The downside is the lack of "Bootstrap for developers that can't CSS to save themselves" ... but don't look at me to write that one - I need that for me :)

 

 

Have you looked at the slides from Andrea's presentation at the Joomla World Conference http://www.slideshare.net/andytarr/bootstrap-for-extension-developers-jwc-2012 "Bootstrap for developers" 

--

You received this message because you are subscribed to the Google Groups "Joomla! General Development" group.

To view this discussion on the web, visit https://groups.google.com/d/msg/joomla-dev-general/-/V0fmDbxwCq8J.

[Donald Gilbert]

Andrew - the link to the platform repository on that page ( http://developer.joomla.org/code.html ) is not correct. Having the .git. on the end of the url breaks.

[Adam Stephen Docherty]

ChrisG,

I see this more of a transition/growing up period really and I think once UCM and new MVC are finalized things should stabilise. Things could not remain the way they were with Joomla, well they could - but it wouldn't really have been taken seriously as a platform and changes in tech would have killed it. You see developing something like Joomla is not all about pleasing people now, it's about reading future trends and having the ability to quickly adapt to changes. It's a delicate balancing act this, technology is moving at lightspeed and the current standard can become defunct in a very short time, as can the standard that replaces it. 

So in light of this, an inability to react quickly can bring about the downfall of a project - things will probably chug a along for a few years, but then all of a sudden something will come along that just does things better. After this there will be a scramble to get up to speed, but if the leg work has not been done early it will likely be a hack and by the time it is realized the tech has probably moved on again! With the code not being agile the scramble will be repeated each time until people get bored and look for something more organized.

I, for one, believe the project is moving in the right direction. I have seen many of the core devs come of age over the last decade or so - we have some VERY talented people working on J! (CMS and Platform), people that have cut their teeth on Joomla, people that know the system and the requirements much better than you or I probably ever will. There is still more work to be done, but I am sure those that make the calls have the best interests of 3PD in mind - well with the BIG PICTURE anyway. As a 3PD myself I know it sucks having to refactor things constantly, but I think the rewards outweigh the inconvenience due to the fact the Joomla will continue to be credible with the general user base for many years to come. Important! because general users are the bread and butter :)

--

Adam

[Terry Arthur]

I think it is the ending punctuation . not the .git

On Mon, Nov 26, 2012 at 9:58 AM, Donald Gilbert <dilber...@gmail.com> wrote:

Andrew - the link to the platform repository on that page ( http://developer.joomla.org/code.html ) is not correct. Having the .git. on the end of the url breaks.

--

[Donald Gilbert]

Terry, I think you're right.

Also, I sent you a message on twitter - but I find rather hilarious that you work for "eclecticeasel" because I work for "electriceasel" and we both implement Joomla for our clients. heheh

[Terry Arthur]

Hey Don,

I got that note. It's a small world after all. Great business name by the way!

[Brad Gies]

Andrew, I thought I'd just point out this little gem in your post again for anyone that missed it :

>That more or less happened 2 months ago.  The horse has bolted and sky has not fallen (can't wait to see Skyfall >though).  However, there is an opportunity for either volunteer led or commercial entities to pick up extended >Joomla 1.5 on their own, and probably charge a premium for it.

It seems to me that if all this concern about extending the support for 1.5 is justified then the number of websites needing/wanting it create a great opportunity for enterprising developer/company.

In my opinion, it's actually a great way for the Joomla! community to go. The active volunteers that are interested in moving Joomla! forward can continue developing, and those developers/companies that need J 1.5 support can do it.

Maybe the Joomla! community can encourage this effort by asking interested people to make their intentions known, and agreeing to put some kind of notice on the official Joomla! website something like "The Joomla! Open Source project has discontinued support for Joomla! 1.5, however extended support is being provided by the following : " and then give a list of links to the various (competing) companies and individuals providing extended 1.5 support. I think if the Joomla! community made a point of organizing it a little it could be quite successful.

That should cover everyone's concerns.

Brad.

-- 
Sincerely,

Brad Gies
----------------------------------------------
bgies.com              maxhomevalue.com 
idailythought.com      greenfarminvest.com
---------------------------------------------- 

[TownWebsites]

Can you describe the management structure of the Joomla project?  How are decisions for changes in the platforms, how are the changes vetted, who manages the volunteers etc.

 

Thanks,

 

Charlie (prospective volunteer, if there is a management structure capable of steering)

[Andrew Eddie]

On 26 November 2012 21:32, Naouak <tar...@gmail.com> wrote:

I wanted to do some stuff for joomla. At first, before github was used, on the website you were supposed to sign a paper which I couldn't found anywhere.

When github was introduced, I did a pull request (10 months ago) and it is still waiting for someone to do something with it ( https://github.com/joomla/joomla-cms/pull/90#commit-ref-4c9c45e ).

That one, regrettably, is out of my hands.  I would most certainly prefer it if the CMS embraced pull requests more, but that's their call and they must have their reasons.

 

Now, we almost don't use joomla core components and so our stuff is not really relevant for core. I want to share some of our code with communities to help maintain it but it's company property and it's hard to make them understand what we can gain from open sourcing some stuff.

It's hard when the company you work for only uses Joomla to save money and doesn't see that contributing back is a good investment.

 

I barely have time to read some mails from this mailling list, I can't follow github repositories and other mailling list.

If you need a way to communicate with your developpers and user, take a look at how concrete 5 is doing it, they are doing very well and people don't have to seek information.

Can you point me to some examples of where they are doing it right?

Thanks in advance.

Regards,

Andrew Eddie 

[Naouak]

One of the simplest yet best way to communicate to users :

Every day when you connect to concrete5 as an admin, you got a nice lightbox with a condensed resume of latest useful information on community : last news on concrete5, last extensions, last documentations added... etc.

I personnally plan to add that to my companies jomla when I have time to facilitate communication with our end users about what's new or new changes to extensions.

Naouak, Grade 2 de Kobal.
Site web: http://www.naouak.net

--

[Andrew Eddie]

On 27 November 2012 02:35, Brad Gies <rbg...@gmail.com> wrote:

Maybe the Joomla! community can encourage this effort by asking interested people to make their intentions known, and agreeing to put some kind of notice on the official Joomla! website something like "The Joomla! Open Source project has discontinued support for Joomla! 1.5, however extended support is being provided by the following : " and then give a list of links to the various (competing) companies and individuals providing extended 1.5 support. I think if the Joomla! community made a point of organizing it a little it could be quite successful. 

I think the key here is that those that are interested in this effort need to drive it (in other words, it's not the PLT's job to organise this).

Regards,

Andrew Eddie

[Andrew Eddie]

On 27 November 2012 07:56, Naouak <tar...@gmail.com> wrote:

One of the simplest yet best way to communicate to users :

Every day when you connect to concrete5 as an admin, you got a nice lightbox with a condensed resume of latest useful information on community : last news on concrete5, last extensions, last documentations added... etc.

Ok, that's not what I thought you meant.  As an aside, I'm not a huge fan of news like that coming through the CMS itself.  Few developers know how to handle dealing with sites behind a firewall.

However, I've set up Pulse to do all that on my iPhone from the feeds that are freely available.  There's nothing stopping you from doing your own mashup with any number of tools that are available (Google Reader, etc).  All the major sites have a feed or two.

Regards,

Andrew Eddie

[Donald Gilbert]

Someone could put together a feedstitch of all J! Dev news and make that available. Nothing like doing the work for 'em.

--

[Andrew Eddie]

On 26 November 2012 20:30, brian teeman <joom...@googlemail.com> wrote:

Have you looked at the slides from Andrea's presentation at the Joomla World Conference http://www.slideshare.net/andytarr/bootstrap-for-extension-developers-jwc-2012 "Bootstrap for developers" 

No.  Is there a collation of slides and videos anywhere?

Regards,

Andrew Eddie

[brian teeman]

The videos are being uploaded now and will be linked to from the session details at http://conference.joomla.org

[Andrew Eddie]

On 27 November 2012 08:42, brian teeman <joom...@googlemail.com> wrote:

The videos are being uploaded now and will be linked to from the session details at http://conference.joomla.org

Thanks.

Regards,

Andrew Eddie 

[Andrew Eddie]

On 27 November 2012 01:54, Terry Arthur <eclect...@gmail.com> wrote:

I think it is the ending punctuation . not the .git

On Mon, Nov 26, 2012 at 9:58 AM, Donald Gilbert <dilber...@gmail.com> wrote:

Andrew - the link to the platform repository on that page ( http://developer.joomla.org/code.html ) is not correct. Having the .git. on the end of the url breaks.

Fixed.  Thanks.

Regards,

Andrew Eddie 

[Terry Arthur]

@Don

Here's a link to the 30 something feeds I found that relate to Joomla!. I put them in a bundle for sharing.

https://www.google.com/reader/bundle/user%2F00257525959543242187%2Fbundle%2FJoomla!%20Feeds 

Anyone can edit to their needs or preferences and let me know what I missed.

Terry

[Andrew Eddie]

On 27 November 2012 11:12, Terry Arthur <eclect...@gmail.com> wrote:

@Don

Here's a link to the 30 something feeds I found that relate to Joomla!. I put them in a bundle for sharing.

https://www.google.com/reader/bundle/user%2F00257525959543242187%2Fbundle%2FJoomla!%20Feeds 

Anyone can edit to their needs or preferences and let me know what I missed.

Nice work, but it highlights the signal-to-noise problem :)  A good master reference nonetheless (couple of ring ins though, hehe).

Regards,

Andrew Eddie 

[Niv Froehlich]

Been following these threads and a couple of thoughts:

(I mostly follow the dev thread 'cuz I'm fairly new and it helps me to keep a pulse on what's happening, what the issues are, where Joomla!'s headed, etc.)

I've probably read all the 100s of posts, even if many were just cursory - in the 'Joomla! Sky is Falling Category'

That said:

1) I wonder how much of the 'exodus' and 'killing me' reactions are a result of the rapid change and improvements to technology that is taking place.  In the world of I.T., if you don't move forward, you get left behind (I took a 3 year hiatus from development and jumped back in 6 months ago - advancement are 'mind-boggling'  I'm the first guy who wants to say - "Hey...can we just stop development for a few months so I can get caught up?").

I remember when the term 'web years' was coined to mean '6 months' - and these days there is not one developer that I know that  doesn't get the sense they are falling behind on at least something (i.e. CSS advances 2.1 -> 3 and the modular vs. monolithic specs, responsive design, jQuery and other JavaScript libraries, the rapid adoption and business use of social media, and so many other technologies and uses etc. ) - Yeah, it would be great to hit the pause button - but moving forward and trying to keep up while hanging on to your seat - is perhaps the new 'reality.'  Love it or leave it.

- and -

2) My gut feeling tells me that things are just about to get better than ever for the developers who stick with all changes with  an 'aggressive forward looking' approach and attitude - the pain will be worth it - and for those who want to 'jump ship' - you may be doing so just as OS and Joomla! and really about 'rock and roll' 

So yeah - the team may be 'overdeveloping while going mach 10 with their hair on fire' - but truthfully, I'm loving every second of it and learning a hell of lot - with so many people from the Joomla! community willing to help!

So thank you to all the devs!  One day I hope to contribute something really positive the code base, something really useful to everybody, and I can't wait until I hear somebody (or even a whole bunch of people) whine and bitch and moan about that change.

N 

On Sat, Nov 24, 2012 at 11:32 PM, ChrisG <chris....@gmail.com> wrote:

Nick, sorry to create another topic but "Joomla Exodus" is huge and rambling and I really wanted a topic to focus on the needs of small commercial developers. I'm afraid I don't agree about the changes being minimal. For a trivial extension, yes, but not for a large extension.

Gary, many thanks for your input, I will definitely create an abstraction layer as you suggest. Thanks for the info about Akeeba, I was thinking along those lines but hadn't got the details yet. I also like what you are doing with namespaces and I will definitely use some of those ideas. I suspect most commercial developers will be forced down a similar route, so one could actually envisage a common commercial API layer serving the needs of commercial developers, which are (perhaps not surprisingly) rather different from the priorities of the core developers.

--
You received this message because you are subscribed to the Google Groups "Joomla! General Development" group.

To view this discussion on the web, visit https://groups.google.com/d/msg/joomla-dev-general/-/GkE4evgT1bAJ.

[Donald Gilbert]

Niv - you've said everything I meant but didn't know how to say.

[Nick Savov]

Hi Naouak,

Please submit that communication feature to the Joomla core, by adding
your code in the following tracker:
http://joomlacode.org/gf/project/joomla/tracker/?action=TrackerItemBrowse&tracker_id=8549

We welcome your contribution! Joomla is community project run entirely by
volunteers. It's an Open Source project and you can contribute code to
the core, so if you have an improvement that you want to see, we hope you
add it to the core!

If you need any help along the way, please let me know by email or Skype
(nick.savov).

Kind regards,
Nick

[Nick Savov]

+1

[Glenn (GMail)]

Niv - +1 here also, thinking it, just not able to put it so succinctly as you have just done.  Cheers.

[Nick Savov]

Hi Charlie,

Glad to hear you're considering volunteering!

You can see the leadership teams on the following pages:
http://www.joomla.org/about-joomla/the-project/leadership-team.html
http://opensourcematters.org/osm-board.html

The changes to the CMS are basically vetted on the CMS mailing list, the
JBS (Joomla Bug Squad) mailing list, the platform mailing list, and the
Joomla General mailing list (google or ask if you need the links). It
just depends which one is relevant for the respective issue. Often issues
are cross-posted.

Here's an example of one proposed change:
https://groups.google.com/forum/?fromgroups=#!topic/joomla-dev-general/APlA3cjbV_k

As you can see, it was posted two days ago and no developer has responded
yet. Just like so many of these "I'm upset and I want you to know it"
discussions :) , the conversation about changes occur red(usually) long
before the changes occurred. So it's better to speak up before something
happens, rather than be upset after the fact. It has to be a total team
effort and everyone communicating.

The biggest organized group of volunteers that works directly on the code
is likely the JBS. It's very easy to get involved in JBS if you're
interested:
http://docs.joomla.org/Bug_Squad

Hope this helps!

Kind regards,
Nick

> Can you describe the management structure of the Joomla project? How are
> decisions for changes in the platforms, how are the changes vetted, who
> manages the volunteers etc.
>
> Thanks,
>
> Charlie (prospective volunteer, if there is a management structure capable
> of steering)
>
> On Sunday, November 25, 2012 4:25:26 PM UTC-5, Andrew Eddie wrote:
>
>> On 25 November 2012 23:54, ChrisG <chris....@gmail.com

>> <javascript:>>wrote:

>>
>>> Andrew, thank you so much for taking the time. Gosh, I shall have to
>>> watch my words more carefully here in future.
>>>
>>
>> Avoid hyperbole. It reduces confusion (I just got slapped for doing
>> just
>> that in another thread myself, hehe) :)
>>
>>
>>>
>>>
>>
>

> --
> You received this message because you are subscribed to the Google Groups
> "Joomla! General Development" group.
> To view this discussion on the web, visit

> https://groups.google.com/d/msg/joomla-dev-general/-/2l51yNHL95EJ.

[Naouak]

As I mainly want to use this feature for my company and that we will stay for 2.5 for a long time, I will develop it for 2.5 won't it be problematic for adding it in the core ?

@niv

I think you're missing one of the main point when speaking about web dev. Every new stuff in css  or HTML doesn't break old stuff, a code written 10 years ago can still be rendered the same today. Even though you can learn a lot of stuff always to improve general aesthetic of your site, you can stay with old knowledge without any problem. Backwards incompatibility is a serious bug for browser vendors.

[Andrew Eddie]

On 27 November 2012 16:18, Naouak <tar...@gmail.com> wrote:

I think you're missing one of the main point when speaking about web dev. Every new stuff in css  or HTML doesn't break old stuff, a code written 10 years ago can still be rendered the same today. Even though you can learn a lot of stuff always to improve general aesthetic of your site, you can stay with old knowledge without any problem. Backwards incompatibility is a serious bug for browser vendors.

That's totally unrealistic - the would be like saying we need to support hosts still on PHP 3, IE 2 and Netscape Navigator 4. The ONLY way around that (no backward incompatibility) is to write your own software stack.  The development strategy allows for backward incompatible changes at well defined points in time.  If this is unacceptable, Joomla is not for you.

Regards,

Andrew Eddie

[Naouak]

I was just saying that then example is a really bad one. I'm not saying you should never break backward compatibility just that you should always avoid it if you can (e.g.: renaming stuff is one thing that shouldn't happen once it's been released). Even more when a lot of people depends of your code.

Like I said previously, if you really need to break backward compatibility just break once and try to not do it again for a long time.

[Andrew Eddie]

On 27 November 2012 16:34, Naouak <tar...@gmail.com> wrote:

I was just saying that then example is a really bad one. I'm not saying you should never break backward compatibility just that you should always avoid it if you can (e.g.: renaming stuff is one thing that shouldn't happen once it's been released). Even more when a lot of people depends of your code.

Like I said previously, if you really need to break backward compatibility just break once and try to not do it again for a long time.

Right, so for the CMS that happens with you do a major release (2.x to 3.x).  The API that's in 3.0 is locked into the 3.x series until 4.0, that is, in terms of backward compatibility.  New API can obviously be added during this time.

Now, it doesn't always work "perfectly" but that is the idea. And in hindsight, Joomla 1.6 should have been called 2.0, but hey, it is what it is.

Regards,

Andrew Eddie

[Naouak]

IMO, you should not plan when you can break compatibility, you should do it only if you cannot do otherwise. The longer the same code can work the best it is for developers and end users.

--

[Niv Froehlich]

@ Naouak - your point is well taken to what you wrote below, and re-iterated by by me in one of these countless threads - we are not in disagreement, but I certainly wouldn't begrudge a browser engine for dropping support of  <font> tags.  At some point in time, you just gotta move on. That said, I think as a positive result of all these threads that a) developers will be more inclined and proactive to keep current; and b) there will be a greater emphasis from the core on backwards compatibility (a 'softer stick' so to speak...)

[Radek Suski]

IMO, you should not plan when you can break compatibility, you should do it only if you cannot do otherwise. The longer the same code can work the best it is for developers and en

Sorry, but this is an extremely ignorant statement for a developer.

If everybody would think like this, we will still be working on Windows 95 because it was the only one OS compatible with DOS

Sent from my iPad

[Nick Savov]

Only Joomla 3 is accepting new features. Joomla 2.5 is in LTS (Long Term
Support) with no new features added unless they are essential. So you
could implement it in Joomla 3 as a way to give back to the community and
to use for future projects later down the road ;)

Kind regards,
Nick

[Andrew Eddie]

On 27 November 2012 16:43, Naouak <tar...@gmail.com> wrote:

IMO, you should not plan when you can break compatibility, you should do it only if you cannot do otherwise. The longer the same code can work the best it is for developers and end users.

I agree we should not do it if we don't have to, but I also believe we should not be maintaining "old" code for excessive periods of time.  That's a maintenance problem.  I disagree that the longer you leave it in, the better it is for everyone.  We know for a fact that the legacy layer in 1.5 allowed developers to smoothly transition from 1.0 to 1.5.  However, it also made many developers complacent and they never got around to updating the code - until they were forced to ...

So, the rule of thumb for the CMS is that you should be able to support 2 major versions of Joomla with the same codebase (for example, 2.5+3.x).  That means a developer need only review their code when the next major release comes around.  I believe that will be every 2 years.

So, after you've gone through the compatibility list and got your extensions working for 2.5+3.x, you should not necessarily have to touch your code until 4.0 starts knocking at the door in mid-2014.  That is completely reasonable in my view and there is ample time to factor in extension upgrades into your business plans.

The bottom line is the project is doing a pretty good job at managing change.  The problem is that we are in that part of the release cycle where you are forced do some work as an extension developer, probably catching up on things you've been putting off for ages.  That's life in this game unfortunately.

Regards,

Andrew Eddie

[Naouak]

Dosbox wasn't created for nothing, win 7 had a win xp mode because it was necessary.
Then again I don't,t said you shouldn't break backward compatibility, you should avoid it at all cost.

Major frameworks have preferred doing a complete rewrite when they need to break compatibility like for instance symfony. Doing it step by step like joomla is doing it is in my opinion not a good way to do it.

[Andrew Eddie]

On 27 November 2012 17:00, Naouak <tar...@gmail.com> wrote:

Major frameworks have preferred doing a complete rewrite when they need to break compatibility like for instance symfony.

I understand that was not well received in the Symfony community.

Regards,

Andrew Eddie 

[Radek Suski]

Dosbox? Seriously?

On Tuesday, 27 November 2012 08:00:20 UTC+1, Naouak wrote:

Then again I don't,t said you shouldn't break backward compatibility, you should avoid it at all cost.

I can agree on one thing for sure, we have sometimes small issues with deprecating methods and dropping the compatibility a bit too fast. It's my personal opinion BTW. But your statement above is soo wrong that it's just hard to believe.

Nevertheless, because I don't see any sense in this whole discussion, Joomla! is an Open Source project. It's being developed openly and while this development you are always free to download the current development version and check if it still work with your extension.

You can prepare for changes soon enough, and even more, you can report these issues before a version is being released.  

I'm telling this (again) because I noticed that most of this kind of posts are mostly created by people which do not care about the fact, do not participate (at least reading) in discussions and then whining how bad Joomla! is after a stable version has been released.

 

On Tuesday, 27 November 2012 08:00:20 UTC+1, Naouak wrote:

Dosbox wasn't created for nothing, win 7 had a win xp mode because it was necessary.
Then again I don't,t said you shouldn't break backward compatibility, you should avoid it at all cost.

Major frameworks have preferred doing a complete rewrite when they need to break compatibility like for instance symfony. Doing it step by step like joomla is doing it is in my opinion not a good way to do it.

On Nov 27, 2012 7:46 AM, "Radek Suski" <suski...@gmail.com> wrote:

IMO, you should not plan when you can break compatibility, you should do it only if you cannot do otherwise. The longer the same code can work the best it is for developers and en

Sorry, but this is an extremely ignorant statement for a developer.

If everybody would think like this, we will still be working on Windows 95 because it was the only one OS compatible with DOS

Sent from my iPad

On 27.11.2012, at 07:43, Naouak <tar...@gmail.com> wrote:

IMO, you should not plan when you can break compatibility, you should do it only if you cannot do otherwise. The longer the same code can work the best it is for developers and end users.

On Nov 27, 2012 7:39 AM, "Andrew Eddie" <mamb...@gmail.com> wrote:

On 27 November 2012 16:34, Naouak <tar...@gmail.com> wrote:

I was just saying that then example is a really bad one. I'm not saying you should never break backward compatibility just that you should always avoid it if you can (e.g.: renaming stuff is one thing that shouldn't happen once it's been released). Even more when a lot of people depends of your code.

Like I said previously, if you really need to break backward compatibility just break once and try to not do it again for a long time.

Right, so for the CMS that happens with you do a major release (2.x to 3.x).  The API that's in 3.0 is locked into the 3.x series until 4.0, that is, in terms of backward compatibility.  New API can obviously be added during this time.

Now, it doesn't always work "perfectly" but that is the idea. And in hindsight, Joomla 1.6 should have been called 2.0, but hey, it is what it is.

Regards,

Andrew Eddie

--
You received this message because you are subscribed to the Google Groups "Joomla! General Development" group.
To post to this group, send an email to joomla-de...@googlegroups.com.

To unsubscribe from this group, send email to joomla-dev-general+unsub...@googlegroups.com.

For more options, visit this group at http://groups.google.com/group/joomla-dev-general?hl=en-GB.

--
You received this message because you are subscribed to the Google Groups "Joomla! General Development" group.
To post to this group, send an email to joomla-de...@googlegroups.com.

To unsubscribe from this group, send email to joomla-dev-general+unsub...@googlegroups.com.

For more options, visit this group at http://groups.google.com/group/joomla-dev-general?hl=en-GB.

--
You received this message because you are subscribed to the Google Groups "Joomla! General Development" group.
To post to this group, send an email to joomla-de...@googlegroups.com.

To unsubscribe from this group, send email to joomla-dev-general+unsub...@googlegroups.com.

[Radek Suski]

Yes, the web is moving fast, but (correct me if I'm wrong) a web page written in 1997 will still render perfectly in today's browsers. The W3C understand backward compatibility. All large software vendors understand it. And IBM learned to their cost (with OS2) that you can't just create a new API and expect the world to jump on.

I think you're comparing it the wrong way, the question is if any important website would still render nicely in Internet Explorer 4

On Nov 27, 2012, at 09:35, ChrisG <chris....@gmail.com> wrote:

Of course I accept that Joomla has to move forward or it will die, but I believe that backward compatibility could be maintained with relatively little overhead. Was there really that much benefit in deleting old functions that merely called newer functions? Was there really that much benefit in renaming JDatabase::nameQuote() to JDatabase::quoteName()? I just don’t believe the benefits (if any) justified the work created. I had to edit hundreds of sql queries, which necessitated days of testing. Why was it necessary to remove those old functions? What harm were they really doing?

Yes, the web is moving fast, but (correct me if I'm wrong) a web page written in 1997 will still render perfectly in today's browsers. The W3C understand backward compatibility. All large software vendors understand it. And IBM learned to their cost (with OS2) that you can't just create a new API and expect the world to jump on.

I have tried to get my point across that it's the extension developers that bear the brunt of the work created when Joomla breaks backward compatibility. We are the ones who get the grief from the site owners. I just think we should have a say in this.

--
You received this message because you are subscribed to the Google Groups "Joomla! General Development" group.

To view this discussion on the web, visit https://groups.google.com/d/msg/joomla-dev-general/-/Pws2AG9qFhkJ.

To post to this group, send an email to joomla-de...@googlegroups.com.

To unsubscribe from this group, send email to joomla-dev-gene...@googlegroups.com.

[Naouak]

Radek please just elaborate and don't say just "it's bad", "you're kidding me" and other stuff like that. I won't answer these kind of answers.

I think we can agree on no true way to manage backward compatibility exists. I think I have made more than clear that every stuff that I say in this conversation is merely my opinion.

If you have read the thread, you would have understood that the problem for me for example is that I got a big number of extensions and not much time to do migrations. Some times ago I have made some calculations and just migrating extensions would take at least a week of work for my team. When it seems just changing some code with another, to me it's checking that everything works well, that nothing is broken and that's also explaining all that to management so we can justify the loss of a week.

We are a small team and we cannot afford to put too much in just maintaining stuff.

Naouak, Grade 2 de Kobal.
Site web: http://www.naouak.net

To view this discussion on the web, visit https://groups.google.com/d/msg/joomla-dev-general/-/yBhP4iXX5KMJ.

To post to this group, send an email to joomla-de...@googlegroups.com.

To unsubscribe from this group, send email to joomla-dev-gene...@googlegroups.com.

[Bakual]

Am Dienstag, 27. November 2012 09:35:09 UTC+1 schrieb ChrisG:

Yes, the web is moving fast, but (correct me if I'm wrong) a web page written in 1997 will still render perfectly in today's browsers. The W3C understand backward compatibility. All large software vendors understand it. And IBM learned to their cost (with OS2) that you can't just create a new API and expect the world to jump on.

If it's written in plain HTML, it will probably still render, yes. But it will look very bad in todays eyes and nobody would visit it anymore.

If it's written in any programmic language (like PHP, Perl, whatever), chances are high that it doesn't work anymore, especially if it's a bit more complex code.

[Radek Suski]

If it's written in plain HTML, it will probably still render, yes. But it will look very bad in todays eyes and nobody would visit it anymore.

If it's written in any programmic language (like PHP, Perl, whatever), chances are high that it doesn't work anymore, especially if it's a bit more complex code.

Exactly

--
You received this message because you are subscribed to the Google Groups "Joomla! General Development" group.

To view this discussion on the web, visit https://groups.google.com/d/msg/joomla-dev-general/-/2yr6wG-sF9UJ.

[Nick Savov]

Hi Chris,

Many of the people that contribute to the core are extension developers.
So, it's definitely not a Core Developer vs Extension Developer scenario.

You *do* have a say in it and it was two years ago, one year ago, nine
months ago, six months ago, etc ago, however speaking up 2 months or so
after Joomla 3 has arrived is way too late.

If you really want to make a positive change in Joomla, get involved in
the project that's given you so much already (might I add for free). Give
back to the community that's given to you. You've been around since
Mambo, correct? I think it's time for you to consider contributing back a
little :)

Kind regards,
Nick

[Andrew Eddie]

On 27 November 2012 18:35, ChrisG <chris....@gmail.com> wrote:

Was there really that much benefit in renaming JDatabase::nameQuote() to JDatabase::quoteName()? I just don’t believe the benefits (if any) justified the work created. I had to edit hundreds of sql queries, which necessitated days of testing. Why was it necessary to remove those old functions? What harm were they really doing?

The benefit was in changing over to using the query builder class, for example $query = JFactory::getDbo()->getQuery(true).  This was also just one change in a massive refactoring of the database package to better accommodate other database engines.  There were quite a few of us that worked on that patch, and we took the time to rationalise some things that have been annoying us for years (I put the bulk of that code into Mambo 4.5 in 2003, so it's not unreasonable to consider it needed a refresh after seven years).

That one is on the knife edge of personal preference but it seemed more logical to teach people to "quote" values, and to "quoteName[s]" (as opposed to "naming quotes", huh?).  The getEscape method was a little more logical - it wasn't a real "getter".  Using the verb "escape", as used JView was seen to be more consistent.  

Likewise, "getTableFields" moving to "getTableColumns" was aligning better with database terminology (to be honest, it should probably be in a schema helper, but whatever).  "loadResultArray" was always the clumsy step-child of "loadResult"; "loadColumn" is a far more logical replacement.  The "query" method name was kind of ok for selects, but for a delete? The "execute" method more correctly describes what you are doing (executing a query/SQL).  Set the query - execute.  Simple.

DS also attracted much criticism.  You just don't need to deal with it a PHP 5 world.  Just as I don't understand cat people (I'm a dog person), I just can't fathom why all developer don't breath a sign of relief when they realise they can just use forward-slash nearly all the time.  Maybe I'm just strange in that it doesn't bother me to spend an hour doing cathartic search and replacing if it makes the code more beautiful and looking consistent.

The reasoning for the new MVC takes a while to get your head around, until you realise how simple it is and how much baggage was there to work around deficiencies in PHP 4.  I would have preferred the CMS added JModelBase in lieu of JModelLegacy but it wasn't my call and that's what the consensus came to.  In hindsight, not having "legacy" in the class name may have reduced some panic - who knows.  It's still going to take a while for extension developers, even the CMS, to work out how to deal with them, but I'm sure we'll all get there as we play with pet projects.

Other changes are more logical, even welcomed so we obviously got some things right (or we picked messing with things that virtually nobody used anyway).

Now, you don't have to agree with those reasons and you don't even have to like them and I'm not even suggesting the reasoning is perfect; I'm just responding to the justification for the work, for better or worse as the case may be.  With the words of Karen McGrane in mind, hopefully the efforts of all the contributors have made the code suck less.

Regards,

Andrew Eddie

[Mathew Lenning]

Listening to this debate and the exodus and both of them seem to be running in the same circle. 3rd party developers don't like the changes and the developers making the changes don't like the bi*ching of non-contributors that didn't speak up when all these changes were decided.

While it is true that change is unavoidable and moving forward is part of staying in the game, I thought it might be interesting to hear everyone's thoughts about the most basic principle of good OOP design. That is "build to an interface not an implementation."

As for the 3rd party developers having a hard time keeping up, I suggest you read "Domain Driven Design" by Eric Evans, it has some good concepts that are very relevant to the current state of the Joomla project.

[Sri Atm]

Changes are required ...but drastically changes may spoil the essence of joomla???

--
You received this message because you are subscribed to the Google Groups "Joomla! General Development" group.

To view this discussion on the web, visit https://groups.google.com/d/msg/joomla-dev-general/-/eQPHt6Z0LFsJ.

[Andrew Eddie]

On 27 November 2012 19:56, Mathew Lenning <mathew....@gmail.com> wrote:

While it is true that change is unavoidable and moving forward is part of staying in the game, I thought it might be interesting to hear everyone's thoughts about the most basic principle of good OOP design. That is "build to an interface not an implementation."

Moving forward, absolutely.  You'll see we've started down that path with the new MVC.  Name spacing discussions have also started.  The downside of that is that concepts like that aren't going to help extension developers that are just "getting by".

Regards,

Andrew Eddie

[Naouak]

That seems to me, renaming just for the sake of it. You could just keep aliases and not depreciate them then delete them. I believe it's common to have functions with a lot of aliases in php.

Sure the name aren't always intuitive but do we need to break extension just because the name didn't sound right ? It was even more ridiculous when it was about UK english vs US english.

Naouak, Grade 2 de Kobal.
Site web: http://www.naouak.net

--

You received this message because you are subscribed to the Google Groups "Joomla! General Development" group.

[Mathew Lenning]

I don't think we should use Microsoft as an example of proper development practices. After all IE is the poster child for how to screw everyone. I'm sure I've spent at least half my life trying to figure out why the hell it doesn't work right in IE.

I'm not defending the changes that have happened in Joomla, but you have to admit the contributors are making some kick ass features available for us to use (once we figure out how to use them)

[Swapnil Shah]

Sorry if this is trolling but Niv you are my hero! LoL!

Regards, 

Neil

Sent from my iPhone

On Nov 26, 2012, at 10:18 PM, Niv Froehlich <nivs...@gmail.com> wrote:

Been following these threads and a couple of thoughts:

(I mostly follow the dev thread 'cuz I'm fairly new and it helps me to keep a pulse on what's happening, what the issues are, where Joomla!'s headed, etc.)

I've probably read all the 100s of posts, even if many were just cursory - in the 'Joomla! Sky is Falling Category'

That said:

1) I wonder how much of the 'exodus' and 'killing me' reactions are a result of the rapid change and improvements to technology that is taking place.  In the world of I.T., if you don't move forward, you get left behind (I took a 3 year hiatus from development and jumped back in 6 months ago - advancement are 'mind-boggling'  I'm the first guy who wants to say - "Hey...can we just stop development for a few months so I can get caught up?").

I remember when the term 'web years' was coined to mean '6 months' - and these days there is not one developer that I know that  doesn't get the sense they are falling behind on at least something (i.e. CSS advances 2.1 -> 3 and the modular vs. monolithic specs, responsive design, jQuery and other JavaScript libraries, the rapid adoption and business use of social media, and so many other technologies and uses etc. ) - Yeah, it would be great to hit the pause button - but moving forward and trying to keep up while hanging on to your seat - is perhaps the new 'reality.'  Love it or leave it.

- and -

2) My gut feeling tells me that things are just about to get better than ever for the developers who stick with all changes with  an 'aggressive forward looking' approach and attitude - the pain will be worth it - and for those who want to 'jump ship' - you may be doing so just as OS and Joomla! and really about 'rock and roll' 

So yeah - the team may be 'overdeveloping while going mach 10 with their hair on fire' - but truthfully, I'm loving every second of it and learning a hell of lot - with so many people from the Joomla! community willing to help!

So thank you to all the devs!  One day I hope to contribute something really positive the code base, something really useful to everybody, and I can't wait until I hear somebody (or even a whole bunch of people) whine and bitch and moan about that change.

N 

On Sat, Nov 24, 2012 at 11:32 PM, ChrisG <chris....@gmail.com> wrote:

Nick, sorry to create another topic but "Joomla Exodus" is huge and rambling and I really wanted a topic to focus on the needs of small commercial developers. I'm afraid I don't agree about the changes being minimal. For a trivial extension, yes, but not for a large extension.

Gary, many thanks for your input, I will definitely create an abstraction layer as you suggest. Thanks for the info about Akeeba, I was thinking along those lines but hadn't got the details yet. I also like what you are doing with namespaces and I will definitely use some of those ideas. I suspect most commercial developers will be forced down a similar route, so one could actually envisage a common commercial API layer serving the needs of commercial developers, which are (perhaps not surprisingly) rather different from the priorities of the core developers.

--
You received this message because you are subscribed to the Google Groups "Joomla! General Development" group.

To view this discussion on the web, visit https://groups.google.com/d/msg/joomla-dev-general/-/GkE4evgT1bAJ.

To post to this group, send an email to joomla-de...@googlegroups.com.
To unsubscribe from this group, send email to joomla-dev-gene...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/joomla-dev-general?hl=en-GB.

[Donald Gilbert]

So - I want Joomla to (someday) move to native namespaces. I can't just whine about it and hope it happens. I have to do what I can to help it happen. So I did what I could now, and submitted a pull request - https://github.com/joomla/joomla-platform/pull/1721

This is how OSS works.

Meaning, if you want new features or you want old features to remain, you need to do your part.

[Naouak]

Donald Gilbert : 

If you arrive and say "I oppose to that pull request because it breaks backward compatibility", I don't think it will be enough. Open Source is not a docraty, it should be more like a democraty.

Going against a change in Open Source Software is generally not appreciated even more if maintainers don't know you.

Naouak, Grade 2 de Kobal.
Site web: http://www.naouak.net

--

[Michael Babker]

If you give reasons why something shouldn't be changed/removed, a compromise can be sought.  There's more than one way to skin a cat.

-Michael

Please pardon any errors, this message was sent from my iPhone.

[Andrew Eddie]

On 28 November 2012 04:54, Michael Babker <michael...@gmail.com> wrote:

If you give reasons why something shouldn't be changed/removed, a compromise can be sought.  There's more than one way to skin a cat.

Indeed.  Look, the bottom line is software development is about accepting the good with the bad.  I don't agree with everything and not everyone agrees with me, but when we come to a consensus, we all stick with in.

The Platform has made a lot of changes that make sense for the Platform.  The CMS then takes those changes and can either buffer them, or pass backward compatibility onto you, the developer.  The CMS also has a development strategy that allows them to make backward compatible changes during a new major release. Maybe it's bad timing for some, but hey, it's that time of the cycle where change is required.  Some developers have been well prepared for 3.0, some have not out of sheer bad luck or circumstances, others are in a world of pain of their own making.

The lesson here is that if your livelihood depends on Joomla, you need to invest in making yourself aware of what is going on -before- it's too late for you to make a difference (and you CAN make a difference).  If you don't, you have to be prepared to accept what comes.  We are going to have this discussion again in 2014 when 4.0 comes out; make every effort to be on the 'other' side of the fence.  Pay attention, if for no other reason than to stay one step ahead of your competitors :)

Some 'bare minimum' tips for CMS extension developers:

1. Put developer.joomla.org/news.html in your feed reader.

1a. Also put phpdeveloper.org in there as well.  It's a good mashup of news PHP.

1b. For developers that want to lift their expertise by watching what our best developers are talking about, watch the Platform mailing list as well.

2. Subscribe to the General development mailing list and put it in daily-digest mode.  Skim that often for announcements and topics that may affect you.  It doesn't take long to do that (certainly not as long as replying to, say, this thread).

3. Put the CMS release schedule in your calendar.  It's time based for a reason - so you know when it's going to happen.  Allow time to test your extensions on the alpha's and beta's that come out prior to new releases.  Don't leave it for your clients to "test" and then write you nasty emails.

4. Test installation and running of your extensions with maximum error reporting, preferably also with PHP Strict checking on as well.

Sound like a plan?  Awesome.  It's time to get back to the business of talking about and writing awesome code.

Regards,

Andrew Eddie

[Niv Froehlich]

@ Andrew - this 'checklist' is great.  Thank you!  

You wrote:

Some 'bare minimum' tips for CMS extension developers:

1. Put developer.joomla.org/news.html in your feed reader.

1a. Also put phpdeveloper.org in there as well.  It's a good mashup of news PHP.

1b. For developers that want to lift their expertise by watching what our best developers are talking about, watch the Platform mailing list as well.

2. Subscribe to the General development mailing list and put it in daily-digest mode.  Skim that often for announcements and topics that may affect you.  It doesn't take long to do that (certainly not as long as replying to, say, this thread).

3. Put the CMS release schedule in your calendar.  It's time based for a reason - so you know when it's going to happen.  Allow time to test your extensions on the alpha's and beta's that come out prior to new releases.  Don't leave it for your clients to "test" and then write you nasty emails.

4. Test installation and running of your extensions with maximum error reporting, preferably also with PHP Strict checking on as well.

--

[Donald Gilbert]

Just read this today in a blog post about a different CMS that is upgrading the platform it's built on. I could see Joomla written all over the post. Below is a perfect excerpt: (I've replaced the CMS name with Joomla)

`````

Joomla is slowly transitioning itself to the new world of PHP - which it would have to do at some point. We can do it now and stay at the front of the pack, or be forced to in a few years after we’ve stagnated. If we were to lock things down completely and not upgrade our codebase then you'd grow bored or frustrated and move to a totally new CMS, and that is going to be a LOT more difficult than the upgrade process we have planned.

`````

[editor]

Which blog post? Please could you tell me the URL so I may read the
excerpt in context?

Many thanks.

On 29/11/12 04:19, Donald Gilbert wrote:
> Just read this today in a blog post about a different CMS that is
> upgrading the platform it's built on. I could see Joomla written all
> over the post. Below is a perfect excerpt: (I've replaced the CMS name
> with Joomla)
>
> `````
> Joomla is slowly transitioning itself to the new world of PHP - which it
> would have to do at some point. We can do it now and stay at the front

> of the pack, or be forced to in a few years after we�ve stagnated. If we

> were to lock things down completely and not upgrade our codebase then
> you'd grow bored or frustrated and move to a totally new CMS, and that
> is going to be a LOT more difficult than the upgrade process we have
> planned.
> `````
>

[Nick Savov]

Looks like it's:
https://www.pyrocms.com/blog/2012/11/foundations-for-our-future

> Which blog post? Please could you tell me the URL so I may read the
> excerpt in context?
>
> Many thanks.
>
> On 29/11/12 04:19, Donald Gilbert wrote:
>> Just read this today in a blog post about a different CMS that is
>> upgrading the platform it's built on. I could see Joomla written all
>> over the post. Below is a perfect excerpt: (I've replaced the CMS name
>> with Joomla)
>>
>> `````
>> Joomla is slowly transitioning itself to the new world of PHP - which it
>> would have to do at some point. We can do it now and stay at the front

>> of the pack, or be forced to in a few years after we�ve stagnated. If we

[Donald Gilbert]

That is it Nick - thanks.

[Alonzo Turner]

We've replaced your Drupal crystals with Joomla, let's see what happens. ;p

[Naouak]

Just saying it, but they say that's the first time they will break compatibility.

They asked actively(this word is important here) to add-on dev to see if they are agreeing with the change.

On their site there is 150 extensions. A lot less than joomla.

I don't think it is good comparison.

Naouak, Grade 2 de Kobal.
Site web: http://www.naouak.net

[Nick Savov]

Hi Naouak,

In that blog post by PyroCMS, I don't see anywhere where they asked add-on
developers if they agree with the change; in fact, the opposite is true.
They basically said, they're moving on since it's best, explained why, and
said they'll the devs move along too. If I'm mistaken, please quote
specifically where they asked.

Joomla on the other hand did actively ask. For example, check out the
following leadership blog on Joomla.org:
http://community.joomla.org/blogs/leadership/1296-rfc-joomla-development-strategy.html

Joomla is open as it gets and anyone can participate in the discussion.

Hope this helps!

Kind regards,
Nick

[Nick Savov]

Typo correction "they'll help the devs move along too"

[Niv Froehlich]

1) I don't think it's productive to compare the Joomla! Project advancements and changes with other CMS systems in this way.  Other platforms may be doing it right, they may not be, and for example, with 150 developers compared to the thousands for Joomla!, the logistics and challenges Joomla! faces may be worlds apart;

2) If you don't have to break it, don't! - I think this argument has been hammered home;

3) Forward Progress:  Perhaps one of the most productive contributions I've seen is a checklist on how to stay current.   I can't seem to find any such checklist (although I'm sure one exists - or similar resources exist) on the Joomla.org web site - but nothing in a way that is intuitive to find (if I'm wrong please correct me).

This last point is important, because as somebody vesting their time in Joomla!, banking on the Joomla! platform to help put bread on the table, and understanding that forward progress is not only rapid these days, but necessary - I'd like to know the best-practices and resources for keeping pace (and keeping current).

This is where we need the help of those who are advanced 'Joomla! Gurus' - so the checklist was great! Brilliant!

What are the current resources available for people like me (i.e. jumping in late in the game, on a steep learning curve, wanting to better understand and utilize best-practices and keeping current)?

I would certainly be most appreciative of any helpful tips and advice - which I'll certainly pass along to those developers I know who are also somewhat new to the 'World of Joomla'?

Thanks

N

>>> >> of the pack, or be forced to in a few years after we致e stagnated.

[Donald Gilbert]

And this is why I posted what I did, rather than a link to the post. I really only wanted to share that excerpt. I knew people would think I was comparing the two, but I'm not. Oh well. :p

[editor]

Thank you. It is a very interesting article.

Many thanks and best wishes.

On 29/11/12 15:30, Nick Savov wrote:
> Looks like it's:
> https://www.pyrocms.com/blog/2012/11/foundations-for-our-future
>
>
>> Which blog post? Please could you tell me the URL so I may read the
>> excerpt in context?
>>
>> Many thanks.
>>
>> On 29/11/12 04:19, Donald Gilbert wrote:
>>> Just read this today in a blog post about a different CMS that is
>>> upgrading the platform it's built on. I could see Joomla written all
>>> over the post. Below is a perfect excerpt: (I've replaced the CMS name
>>> with Joomla)
>>>
>>> `````
>>> Joomla is slowly transitioning itself to the new world of PHP - which it
>>> would have to do at some point. We can do it now and stay at the front

>>> of the pack, or be forced to in a few years after we�ve stagnated. If we

[Andrew Eddie]

On 30 November 2012 03:03, Naouak <tar...@gmail.com> wrote:

Just saying it, but they say that's the first time they will break compatibility.

They asked actively(this word is important here) to add-on dev to see if they are agreeing with the change.

On their site there is 150 extensions. A lot less than joomla.

I don't think it is good comparison.

Oh, I don't know about that.

http://community.joomla.org/blogs/leadership/1296-rfc-joomla-development-strategy.html

Regards,

Andrew Eddie 

[Nicholas Brick]

100% agree well said

On Tuesday, 27 November 2012 03:18:37 UTC, Niv Froehlich wrote:

Been following these threads and a couple of thoughts:

(I mostly follow the dev thread 'cuz I'm fairly new and it helps me to keep a pulse on what's happening, what the issues are, where Joomla!'s headed, etc.)

I've probably read all the 100s of posts, even if many were just cursory - in the 'Joomla! Sky is Falling Category'

That said:

1) I wonder how much of the 'exodus' and 'killing me' reactions are a result of the rapid change and improvements to technology that is taking place.  In the world of I.T., if you don't move forward, you get left behind (I took a 3 year hiatus from development and jumped back in 6 months ago - advancement are 'mind-boggling'  I'm the first guy who wants to say - "Hey...can we just stop development for a few months so I can get caught up?").

I remember when the term 'web years' was coined to mean '6 months' - and these days there is not one developer that I know that  doesn't get the sense they are falling behind on at least something (i.e. CSS advances 2.1 -> 3 and the modular vs. monolithic specs, responsive design, jQuery and other JavaScript libraries, the rapid adoption and business use of social media, and so many other technologies and uses etc. ) - Yeah, it would be great to hit the pause button - but moving forward and trying to keep up while hanging on to your seat - is perhaps the new 'reality.'  Love it or leave it.

- and -

2) My gut feeling tells me that things are just about to get better than ever for the developers who stick with all changes with  an 'aggressive forward looking' approach and attitude - the pain will be worth it - and for those who want to 'jump ship' - you may be doing so just as OS and Joomla! and really about 'rock and roll' 

So yeah - the team may be 'overdeveloping while going mach 10 with their hair on fire' - but truthfully, I'm loving every second of it and learning a hell of lot - with so many people from the Joomla! community willing to help!

So thank you to all the devs!  One day I hope to contribute something really positive the code base, something really useful to everybody, and I can't wait until I hear somebody (or even a whole bunch of people) whine and bitch and moan about that change.

N 

On Sat, Nov 24, 2012 at 11:32 PM, ChrisG <chris....@gmail.com> wrote:

Nick, sorry to create another topic but "Joomla Exodus" is huge and rambling and I really wanted a topic to focus on the needs of small commercial developers. I'm afraid I don't agree about the changes being minimal. For a trivial extension, yes, but not for a large extension.

Gary, many thanks for your input, I will definitely create an abstraction layer as you suggest. Thanks for the info about Akeeba, I was thinking along those lines but hadn't got the details yet. I also like what you are doing with namespaces and I will definitely use some of those ideas. I suspect most commercial developers will be forced down a similar route, so one could actually envisage a common commercial API layer serving the needs of commercial developers, which are (perhaps not surprisingly) rather different from the priorities of the core developers.

--
You received this message because you are subscribed to the Google Groups "Joomla! General Development" group.

To view this discussion on the web, visit https://groups.google.com/d/msg/joomla-dev-general/-/GkE4evgT1bAJ.

To post to this group, send an email to joomla-de...@googlegroups.com.

To unsubscribe from this group, send email to joomla-dev-general+unsub...@googlegroups.com.