The Joomla Platform and Composer

[Andrew Eddie]

A few of us have been talking about the future of the Platform and how to manage it, particular as we are approaching a massive change to namespacing. In addition, one of the goals this year is to work out deprecation policies and such. It’s also obvious, looking at the PHP world, that Composer and Packagist are settling in to stay.

To deal with all these issues, I’d like to float a proposal for how we can move forward with the Platform this year.

SUMMARY

The current Joomla Platform will be maintained for as long as people see fit to do so, but without introducing namespacing or any other new “breaking” changes. Effectively this would be used to support the Joomla CMS and the interests of those developing applications on the current version of the Platform.

Next generation work, for example namespacing, will happen in separate packages that are designed to be installed with Composer and published on Packagist. Current platform packages will be progressively moved into separate repositories to be managed by individual teams of maintainers.

THE “OLD” PLATFORM

The current Joomla Platform will continue to be maintained in the https://github.com/joomla/joomla-platform repository under the current Year.Sequence versioning convention. Code may still be deprecated on a schedule but no other breaking changes will be made (for example, like was done with the MVC changes).

Maintenance will continue to be crowd-sourced for as long as people want to maintain this repository.

Obviously the CMS will be one of the heaviest users of this old Platform, but it would also be used by other application developers until such time as they migrate to the new Platform.

THE “NEW” PLATFORM

Composer and its companion Packagist have become the defacto standard for integrating PHP libraries into your applications. The new Joomla Platform needs to fit more easily into this paradigm as well as making it easier for people building Joomla application to be able to leverage other libraries available through Composer more easily. Therefore, with all the work that is going on with namespacing, we propose that the current platform is broken up into individual packages that can be published to Composer and installed with Packagist.

We envisage that there will be one “core” platform package that will contain coding elements that are either very common or critical to most other platform packages. This would include such things as JRegistry, as well as introduce and external dependencies like PHP-FIG interfaces.

Each package would have its own set of maintainers that would manage pull requests against the package but all packages would be bound by the current coding standards or any other related Joomla policy. People would obviously be able to manage more than one repository.

Each package would manage it’s own version numbering according to semantic versioning.

There is also the opportunity to clean up any package (removing legacy code), or even dropping existing packages (like JFile, etc).

CONCLUSION

This new methodology presents a number of opportunities and challenges.

Opportunities:

1. Delegates management away from a small group of platform maintainers that can easily become a bottleneck for accepting or reviewing contributions on the platform as a whole. It is much easier for people to focus on approving code for smaller chunks of code.

2. Allows developers to pick-and-choose what they need to build applications regardless of what core library those applications are built on.

3. Still maintains the Joomla “brand”

4. Allows for packages that fall into disrepair or just hold no interest to be easily retired.

5. The Joomla leadership just has to be responsible for providing the right tools and for guiding policy.

Challenges:

1. There is some new overhead in creating repositories and maintaining access control lists.

2. We’d have to work out how to approve new packages.

3. Probably a host of others we haven't thought about yet.

Even though the Composer paradigm still has some issues to sort out, I can’t hep but see it becoming the standard way of building PHP applications in the future. If the Joomla Platform wants to be a viable offering (and let’s be honest, we are never mentioned in the framework space like we are in the CMS space), I think we need to go down this road - it’s just a question of working out the details of how.

The floor is open for feedback. Thanks in advance.

Regards,
Andrew Eddie
http://learn.theartofjoomla.com - free tutorials and videos on Joomla development

[Amy Stephen]

It is good.

[Nick Savov]

Hi Andrew et. al,

Sounds good to me, assuming we can get enough volunteers to maintain it
all and #3 of the challenges doesn't present any show-stoppers.

Kind regards,
Nick

[David Hurley]

I see this as a good step in the right direction. In order for the "New" platform to stay current and inline with other "frameworks" then the move to Composer is a necessity in my opinion. Only then can we start to build out a framework that will hold its own against others in the space.

The challenges raised are good points and there should definitely be some structure and discussion regarding approval of new packages and ACL. 

This does seem to mean there will be those that are developing the future of the "New" platform while also requiring a separate team to continue progress on the existing Joomla! platform as it relates to the CMS and other platform based applications that currently exist. Would this new platform be named something unique or how would that be handled to differentiate from the existing platform? Are we still discussing something that falls under the Joomla! project or will this split completely away?

This move is much needed for a legitimate framework and I'd like to discuss this more to begin this process. I think this will take some time to structure and make sure it's handled correctly and with as little confusion as possible. 

-

Thanks,
David Hurley
Joomla! Community Development Manager

--
You received this message because you are subscribed to the Google Groups "Joomla! Platform Development" group.
To unsubscribe from this group and stop receiving emails from it, send an email to joomla-dev-plat...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[Andrew Eddie]

On 16 February 2013 11:03, David Hurley <david....@joomla.org> wrote:

This does seem to mean there will be those that are developing the future of the "New" platform while also requiring a separate team to continue progress on the existing Joomla! platform as it relates to the CMS and other platform based applications that currently exist. Would this new platform be named something unique or how would that be handled to differentiate from the existing platform?

That's an interesting question. At this stage I don't think it wise to move away from the Joomla "brand", but there could be a subtle renaming, for example, from Joomla Platform (the old way) to Joomla Framework (Composer version).

 

Are we still discussing something that falls under the Joomla! project or will this split completely away?

I think if people want to do their own thing, there's nothing stopping them doing that now. But I don't see this moving away from the umbrella of Joomla. However, it's entirely possible that we don't try to do everything ourselves. For example, we may decide in Composer world, it's not worth maintaining JLog in the long term as there are perfectly good alternatives around for that. 

My vision for this is that you can decide to bootstrap an application with Joomla code and then add whatever you like, or you can bootstrap with Slim or Symfony but pick and choose bits from Joomla that better than other offerings (our Github package is arguably the best available).

 

This move is much needed for a legitimate framework and I'd like to discuss this more to begin this process. I think this will take some time to structure and make sure it's handled correctly and with as little confusion as possible. 

Sure.

Regards,

Andrew Eddie

[Nestor Ledon]

I feel this is a good move overall. I've been of fan of Joomla! for a while and I feel very comfortable working with it. Pushing to make it viable as a standalone framework with Composer and Packagist is just going to make Joomla! a hundred times better. I can't wait to get involved! Where do I sign!?

[Ronni KC]

Sounds like a pretty good plan to me.

It addresses quite a bit of the problems we see today and also lets diversity flow.

[Andrew Eddie]

So just a quick update on where we are at.

We are currently doing all our edits in a single tree:

https://github.com/joomla/joomla-framework

We are using the working title of "Joomla Framework" for the code that will be accessible from Composer. It seems to be how the wider PHP community talks about this kind of stuff and that's our target audience as much as the Joomla community is.

We are using get sub-split to make the individual repo's that feed into composer - these are read only.

There's still a ton of work to do. At the moment the focus includes (but is not limited to):

* Running code sniffer over all the packages. We updated the "private" rule for underscores recently (that eliminates one more difference we have with our code standard and PSR-2).

* Namespacing the unit tests.

* Removing all deprecated code. We assume that people using Composer will be ok with not using any old code.

* Removing all CMS coupling. This will include things like database driven form fields.

* Update README.md files in package to use namespacing.

* Swap out the existing Cache package for Louis's replacement.

If you are able to help on a regular basis (like, a few hours or more a week), let us know and we can set up a skype chat to keep everyone coordinated. If you have any questions, don't hesitate to ask - no secrets around here ;)

[Nick Savov]

Hi Andrew,

Would this be a good opportunity to get to over 90% unit testing or
perhaps over 80%? It might be a good idea before releasing it on composer
to get the unit testing where we'd like.

Kind regards,
Nick

[Andrew Eddie]

On 25 February 2013 11:14, Nick Savov <ni...@iowawebcompany.com> wrote:

Hi Andrew,

Would this be a good opportunity to get to over 90% unit testing or
perhaps over 80%?  It might be a good idea before releasing it on composer
to get the unit testing where we'd like.

At the moment just getting things working again would be a nice goal.

The rule is we need unit tests for brand new work and we've got a huge backlog to catch up on. We'll have a better handle on things once we get everything running again. However, it there are people that want to help increase the existing code coverage, I'm more than happy to guide them.

Regards,

Andrew Eddie 

[Nick Savov]

I agree with that assessment. Thanks Andrew!

Kind regards,
Nick

[Don]

Those are all very good and valid points Chris. I appreciate the thought you out into the post. I'm going to re-read and process it a bit more to give it the response it deserves, but let me say I agree with just about everything you said. 

Sent from my iPhone

On Feb 25, 2013, at 10:59 AM, kisswebdesign <chris.jo...@kisswebdesign.co.uk> wrote:

This sounds like the way forward if Joomla is to become a framework. I have had a look at packagist for the Joomla stuff and can see its advantages - great start :-)

From what I can see, there will be 2 development streams - 'old' platform and 'new' framework - the Joomla CMS will be built on the 'old' platform, which will become a 'poor cousin' to the 'new' framework.

Engineers/developers/programmers/marketers love working on cool new stuff, so the framework will get more attention than the platform - this is natural, normal and not a criticism.

Given the relatively small number of developers who know enough about the platform to maintain and develop it, and the fact that they will also be the ones developing and maintaining the framework, it is easy to see how the platform could get left behind - which puts the CMS in danger of running into an evolutionary dead-end.

So, is there a plan (or even a line on a roadmap) to _assist_ the CMS in moving from the platform to the framework?

Given that the CMS is the largest customer of the platform, they are somewhat tied to it, and it to them. Decoupling them from each other is an evolutionary step change that must be done if the Joomla brand is to provide both a framework and a CMS, but the CMS also needs to make this step function change (all be it at a later time) to embrace the new framework and get all the benefits it brings.

From a Joomla brand POV I think it would be necessary for the CMS to use the framework (eat your own), or you get the situation where people look at Joomla and say "If that framework is so good, why isn't the CMS using it?" and "If the old platform is stable enough for the CMS, with its huge install base, then why should we risk using this new framework?".

I see many developers here that are also CMS developers, so maybe the CMS would adopt the framework naturally at some future date. However, I would prefer to see a consolidated approach where the framework and CMS are 'singing from the same hymn sheet' - and both say that the framework is the way forward, it will overtake the platform as the foundations for the CMS after (for example) the next-plus-one LTS (CMS 4.5) is released - this gives plenty of time for:

1. The framework to be developed, and be stable enough for the CMS to use it in production;
2. The CMS developers to make the necessary code base changes required to use the framework;
3. 3rd party devs to plan for the future.

Additionally, it gives Joomla customers (be that 3rd party devs, website builders or end-user admins) a clear view of the future where Joomla is a strong brand that will provide (ever improving, robust) solutions for them now and in 5+ years time. Allowing everyone to make decisions now that have long lasting meanings for them and for Joomla.

For example...

A company is planning to move from a static HTML brochure site to a CMS. They look around and come up with a shortlist of

• Wordpress
  
• Joomla
  
• Drupal
  

One of the key factors for most businesses is money (value, price, etc), and they want a solution that they can rely on for the next 5 years. They know that once the decision is made, changing is expensive, so they want to make sure that they make a good choice based on good information.

All 3 solutions can (probably) provide a suitable solution for them right now, but what about the future? How are the releases managed, bugs fixed, security updates applied? What about future technologies and stability - we don't want to be left behind, but we want something that is stable and useable for us and our customers.

A quick search for "joomla roadmap" then "wordpress roadmap" then "drupal roadmap" doesn't help - there are lots of results in the search pages, but they are not very helpful. So they end up using whichever CMS the developer likes best at the time - and they are probably happy with it now and for the next 5 years.

Wouldn't it be better if the search for a Joomla roadmap had produced a page showing where Joomla is going and the long term plan - then the company would ask for Joomla, because it has the long term vision and a plan on how to get there.

I knocked-up a quick sample roadmap (attached) - I just made up the timescales because I don't know the real ones, but it would be simple enough to put the right ones in.

If Joomla had a page that showed the roadmap (maybe as a now-2quarters to now-plus10quarters default view, scrollable left & right, with a downloadable version) I think it would help potential customers understand where Joomla is heading, and give them confidence that Joomla is here for the long-haul. Also new and existing developers would be able to see, at a glance, what is coming and make plans accordingly. Additional notations for registered users could also show break-points where things are deprecated, and where changes to methods, classes, etc. would break backwards compatibility.

Obviously it would need to be updated - probably in line with PLT meetings, so any big changes are captured and reflected in a consistant manor.

If you are still reading, well done. These are just my thoughts on where the Joomla platform and framework are heading, and the potential impact this has on the CMS and the perception of Joomla as a brand/product family. It is not meant to offend, hurt or otherwise annoy anybody, and if I have then accept my heartfelt appology.

In summary, I think this is the right way to go - and I hope the CMS follows suit and uses the framework (even if it means a painful transition).

Chris.

PS - I am happy to help where time, and more importantly skills allow. So far I am only dipping my toes into the water ;-)

--
You received this message because you are subscribed to the Google Groups "Joomla! Platform Development" group.
To unsubscribe from this group and stop receiving emails from it, send an email to joomla-dev-plat...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

<sample-roadmap.pdf>

[Craig Phillips]

Great points.

This does cover all of the areas I've been concerned about re: the impact of  the 'old platform' vs 'new framework'...

From my point of view the CMS needs the new framework but also the new framework needs the "flagship" application that the CMS is (whether you like CMS's or not, or prefer the new fangled hotness that are web applications).  The fact is "Joomla!" to the greater masses is a CMS, not the bifurcated products we all know and love.

I'll just add that the "missing roadmap" is definitely also something hard to explain to the our normal clients (i.e. Universities, Local Government groups and State Government Departments, Universities and Legal firms).

Cheers

Craig

[Andrew Eddie]

On 26 February 2013 02:59, kisswebdesign <chris.jo...@kisswebdesign.co.uk> wrote:

Pretty cool post. It's going to take a while for it all to sink in :)

This sounds like the way forward if Joomla is to become a framework. I have had a look at packagist for the Joomla stuff and can see its advantages - great start :-)

From what I can see, there will be 2 development streams - 'old' platform and 'new' framework - the Joomla CMS will be built on the 'old' platform, which will become a 'poor cousin' to the 'new' framework.

I think that's a fair statement, but it also needs to be kept in perspective. The rate of change that an application developer can tolerate is much higher than what the CMS extension community will accept. That's going to cause a disparity.

 

Given the relatively small number of developers who know enough about the platform to maintain and develop it, and the fact that they will also be the ones developing and maintaining the framework, it is easy to see how the platform could get left behind - which puts the CMS in danger of running into an evolutionary dead-end.

I don't really see the Platform as-is hanging around for too long. I think we'd hand it back to the CMS fairly quickly and they can decide what they want to add/drop etc. On the bright side, it reduces the double handing the CMS has for fixing bugs in both the CMS repo and the Platform repo and keeping all that stuff in sync.

 

So, is there a plan (or even a line on a roadmap) to _assist_ the CMS in moving from the platform to the framework?

That's something the PLT is going to be talking about this year. My personal thoughts are that the roadmap is always going to be very high level. Like, we know when each version is coming out, that's easy. What's in each version is hard to say because it depends on who wants to do the work. I don't know how to solve that problem.

 

Given that the CMS is the largest customer of the platform, they are somewhat tied to it, and it to them. Decoupling them from each other is an evolutionary step change that must be done if the Joomla brand is to provide both a framework and a CMS, but the CMS also needs to make this step function change (all be it at a later time) to embrace the new framework and get all the benefits it brings.

The problem with the CMS is that it needs a lot of time to do this, because the extension developers have to embrace the changes in ways that preserve their sanity. I think what will happen is the Framework will race ahead, and then when the CMS gets to the point where it simply must make radical change, the Framework is waiting to be consumed in a fairly mature state.

 

From a Joomla brand POV I think it would be necessary for the CMS to use the framework (eat your own), or you get the situation where people look at Joomla and say "If that framework is so good, why isn't the CMS using it?" and "If the old platform is stable enough for the CMS, with its huge install base, then why should we risk using this new framework?".

The latter is easier to answer. The old platform is great for building the Joomla CMS, but not much else so I don't think we have that problem. The former point is harder and it's up to how fast the extension developers will support change in the CMS. Many developers have simply upgraded their extensions from 1.5 standard to "just work" in 2.5 or even 3. Not many actually embrace all of the cool extra features and such (JForm triggers for example).

 

I see many developers here that are also CMS developers, so maybe the CMS would adopt the framework naturally at some future date. However, I would prefer to see a consolidated approach where the framework and CMS are 'singing from the same hymn sheet' - and both say that the framework is the way forward, it will overtake the platform as the foundations for the CMS after (for example) the next-plus-one LTS (CMS 4.5) is released - this gives plenty of time for:

1. The framework to be developed, and be stable enough for the CMS to use it in production;
2. The CMS developers to make the necessary code base changes required to use the framework;
3. 3rd party devs to plan for the future.

Additionally, it gives Joomla customers (be that 3rd party devs, website builders or end-user admins) a clear view of the future where Joomla is a strong brand that will provide (ever improving, robust) solutions for them now and in 5+ years time. Allowing everyone to make decisions now that have long lasting meanings for them and for Joomla.

I think what will inevitably happen is that the CMS will split into two parts. One will be a legacy part that tries to ride out the current architecture for as long as it can. The other part will probably look at a new generation application that's better suited to delivering content to any device, not just a browser.

 

If Joomla had a page that showed the roadmap (maybe as a now-2quarters to now-plus10quarters default view, scrollable left & right, with a downloadable version) I think it would help potential customers understand where Joomla is heading, and give them confidence that Joomla is here for the long-haul. Also new and existing developers would be able to see, at a glance, what is coming and make plans accordingly. Additional notations for registered users could also show break-points where things are deprecated, and where changes to methods, classes, etc. would break backwards compatibility.

Interesting thought. I'd like to see some mocks of that page :)

 

If you are still reading, well done. These are just my thoughts on where the Joomla platform and framework are heading, and the potential impact this has on the CMS and the perception of Joomla as a brand/product family. It is not meant to offend, hurt or otherwise annoy anybody, and if I have then accept my heartfelt appology.

Very good and appropriate comments!

 

In summary, I think this is the right way to go - and I hope the CMS follows suit and uses the framework (even if it means a painful transition).

Well, we'll see how things go. The reality is we probably have more people that are allowed to dedicate "work time" to the Framework than the CMS and that makes a huge difference to what you can accomplish. It goes in seasons though.

[Paul Orwig]

Some of my thoughts about a roadmap are below.

On Mon, Feb 25, 2013 at 6:17 PM, Andrew Eddie <mamb...@gmail.com> wrote:

On 26 February 2013 02:59, kisswebdesign <chris.jo...@kisswebdesign.co.uk> wrote:

 

So, is there a plan (or even a line on a roadmap) to _assist_ the CMS in moving from the platform to the framework?

That's something the PLT is going to be talking about this year. My personal thoughts are that the roadmap is always going to be very high level. Like, we know when each version is coming out, that's easy. What's in each version is hard to say because it depends on who wants to do the work. I don't know how to solve that problem.

I think it would be great if leadership announced something along the lines of "We feel the following X features/enhancements are important, and if people are interested in working in those areas, we have some standing working groups that we encourage you to join".

That type of roadmap doesn't force anyone to work on anything, doesn't include any deadlines, and doesn't exclude anyone from contributing in other areas, and it doesn't lower any standards for accepting code.

After the project creates that type of high level roadmap, then it should be up to other parts of the project to work together to promote them: blogs, JCM articles, tweets, sessions at events, etc.

Finally, it is up to leadership to actively support the working groups by providing some structure (a project lead, a leadership liaison, a project manager, etc.) and support.

Those are some of my ideas for how we could implement a high level roadmap.

Thanks,

paul

[Amy Stephen]

Andrew - excellent comments, vision, direction.

"The old platform is great for building the Joomla CMS, but not much else."

In time, people will see why that is true and how that cannot continue to be true, for Joomla to be a viable option for developers today.

You're putting the bang back on Joomla.

[Ian]

It's also worth noting that since the new work being done is name spaced it avoids conflict almost entirely with the existing platform.  What this means is that parts of the new work can be brought in as needed and you can move subsystems over in a more incremental manner.  It might be worthwhile for the CMS to work on a way to leverage composer to manage what packages extensions require and to bring them in as needed.

Ian

[Andrew Eddie]

On 28 February 2013 21:51, kisswebdesign <chris.jo...@kisswebdesign.co.uk> wrote:

Agreed, the CMS re-absorbs the platform - reversing the split. Platform could iterate faster than the CMS, but in reality it will fall in line with the CMS release schedule.

Back to 2 products - CMS (including platform) and Framework.

Yes.

 

Framework doesn't appear in the 2013 goals from the PLT, but there are 3 platform releases. Is this an oversight, or is the framework not on the PLT radar yet?

We wrote the goals before we decided to take this detour.

 

Also, why are these new features not being embraced - lack of publicity about them, poor documentation, solving a non-existent problem, development time required does not have any payback, etc. If the reasons can be understood, then positive actions can be taken (now and in the future) to give people what they need, want, adds real value, etc.

I'm really not sure. I suspect it's to do with the custom "layers" the developers have built over the years to isolate themselves from unexpected change. But I don't really know. Lack of documentation and awareness would be a part of it though. We are trying to catch up in that area but it's slow going.

 

This sounds really bad for Joomla and end users, 2 CMS's with the Joomla brand. There should only be one Joomla CMS, there could be a extended LTS end-of-life version that only gets security updates, but the latest version is recommended (and is preferable to devs and end-users due to features, usability, flexibility, etc)

I would argue that the only true CMS is a web services platform (technically speaking). What we call the CMS together is really a Web Site Building Tool. But that's just me :) I don't see the current "CMS" dying any time soon, but it's certainly past its used by date.

 

Having 2 CMS's would only lead to confusion in the marketplace, and turn people off of choosing Joomla.

The "other" thing, whatever it is (it's so secret nobody knows what it is), will most certainly be a different beast. It will probably be a suite of things you "implement". It would be like the current CMS that you can download and have up and running in an hour. Probably something like a pure Angular frontend with this rich UCM-based services layer behind it. It could even a number of different sub-projects that are all lightly coupled but interact with each other seamlessly.

But nobody really knows. All we know is there are practical limits to how far you can take the current CMS given the rate of change the extension developers will tolerate (without ownership of the changes that is - if we can garner that, then we can spin the flywheels up really fast).

Regards,

Andrew Eddie

[kisswebdesign]

 

Framework doesn't appear in the 2013 goals from the PLT, but there are 3 platform releases. Is this an oversight, or is the framework not on the PLT radar yet?

We wrote the goals before we decided to take this detour.

 

OK, thanks for the clarification.

 

Also, why are these new features not being embraced - lack of publicity about them, poor documentation, solving a non-existent problem, development time required does not have any payback, etc. If the reasons can be understood, then positive actions can be taken (now and in the future) to give people what they need, want, adds real value, etc.

I'm really not sure. I suspect it's to do with the custom "layers" the developers have built over the years to isolate themselves from unexpected change. But I don't really know. Lack of documentation and awareness would be a part of it though. We are trying to catch up in that area but it's slow going.

Thats what I assumed to - but you know what they say about assumption. Has anyone ever asked these developers?

 

 

This sounds really bad for Joomla and end users, 2 CMS's with the Joomla brand. There should only be one Joomla CMS, there could be a extended LTS end-of-life version that only gets security updates, but the latest version is recommended (and is preferable to devs and end-users due to features, usability, flexibility, etc)

I would argue that the only true CMS is a web services platform (technically speaking). What we call the CMS together is really a Web Site Building Tool. But that's just me :) I don't see the current "CMS" dying any time soon, but it's certainly past its used by date. 

Semantics. What developers see as a CMS is not what the public sees as a CMS - we can thank Wordpress for that. Trying to change public perception/naming of things (even when they are strictly speaking wrong) is a waste of energy most of the time. Just mention 'hacker' to a non-tech / non-coder and see what they think a hacker is - never going to change this perception / definition, it has too much momentum and public penetration, helped by the media of course.

I also think that we developers sometimes need to take a step back and look at things from a 'member of the general public' [MGP] viewpoint - because we spend a lot of time immersed in technologies that are changing quickly and are constantly looking forward that we see as outdated what MGP sees as current, stable and desirable.

I agree that the current CMS will not die anytime soon, but I'm not sure it is out of date yet. Much as I love to be on the bleeding edge, most people hate being there and are happier using older technologies that they feel comfortable with. 

 

Having 2 CMS's would only lead to confusion in the marketplace, and turn people off of choosing Joomla.

The "other" thing, whatever it is (it's so secret nobody knows what it is), will most certainly be a different beast. It will probably be a suite of things you "implement". It would be like the current CMS that you can download and have up and running in an hour. Probably something like a pure Angular frontend with this rich UCM-based services layer behind it. It could even a number of different sub-projects that are all lightly coupled but interact with each other seamlessly.

But nobody really knows. All we know is there are practical limits to how far you can take the current CMS given the rate of change the extension developers will tolerate (without ownership of the changes that is - if we can garner that, then we can spin the flywheels up really fast).

Mmmm, new sexy hotness, want, now!

I agree, the current CMS is at its practical limit. However, lots of end-users still want this and are happy with it - so how to manage the change responsibly?

Maybe leave the CMS alone and keep calling it the Joomla CMS, just as it is now, and whatever the new thing is have a new name that conveys what it is, without looking like a direct competitor to the CMS. Maybe Joomla! NSH (New Sexy Hotness) lol.

Actually, something like you guess at above could easily be described/named as Joomla! SPA  (Single Page Application http://en.wikipedia.org/wiki/Single-page_application) This would provide a good technical naming of the 'thing', while not causing a problem for the CMS in its current form. They sound like - and are - different products, different technologies, but can be used to create similar end-user solutions (but the SPA is not limited to only this type of product).

Again, this is just me thinking out loud - or rather thinking in type.

Thanks again for your thoughts,

Chris.

[Andrew Eddie]

On 28 February 2013 23:28, kisswebdesign <chris.jo...@kisswebdesign.co.uk> wrote:

 I'm really not sure. I suspect it's to do with the custom "layers" the developers have built over the years to isolate themselves from unexpected change. But I don't really know. Lack of documentation and awareness would be a part of it though. We are trying to catch up in that area but it's slow going.

Thats what I assumed to - but you know what they say about assumption. Has anyone ever asked these developers?

Probably not formally. You don't have to go far to find a thread complaining about developer documentation, but those that know how to write it well are few and far between. It's also the case that there is no guarantee that if you had good docs and good promotion (this kind of thing is promoted at every Joomla event) that developers would change.

 

I agree, the current CMS is at its practical limit. However, lots of end-users still want this and are happy with it - so how to manage the change responsibly?

Nothing changes too quickly. If we even had a prototype of an idea on paper about what the ethereal "Joomla Next" could be, we'd be doing well.

 

Maybe leave the CMS alone and keep calling it the Joomla CMS, just as it is now, and whatever the new thing is have a new name that conveys what it is, without looking like a direct competitor to the CMS. Maybe Joomla! NSH (New Sexy Hotness) lol.

Yes, that seems to be the consensus on how to approach it. The current CMS would retain that designation and whatever the new thing is would fly under a different flag.

 

Regards,

Andrew Eddie