Laravel Roadmap and LTS (Long Term Support) , is there any?

Ashammari, we're looking at building a laravel application on RHEL7 too - did you have any trouble getting everything set up? and did you have to get hold of any non official packages for anything?

Stability would be a critical issue for us too - stability in dependencies is of course another minefield.

I have asked Taylor in twitter about LTS and the roadmap for Laravel and he said , there is nothing like this.

@AlexCallard, we are in the evaluation phase, so far we found the following:

• RHEL7 supports PHP5.4 with mcrypt extension. However, Laravel 4.2 works just fine in this environment.
• RHEL7 does not support mcrypt with PHP5.5 officially , you have to use another non-official repository such as fedora( remi repository ) or others to install mcrypt ext. and based on RHEL7 feedback, they will never ship php5.5-mcrypt extension in the near future. That's why we are not sure about Laravel , if it's in the future drops PHP5.4 support, then it will be a problem for RHEL7 users.

There will be no official LTS yet, but Taylor said in last podcast that he's thinking about a 'fix bug policy' for previous releases in a short future.

On the other hand, PHP5.4 support only goes until 2015 if you're aware of it ?

Thanks @RemiC , yes I know about php.net support , but RHEL7 (RedHat) will support php5.4 for a longer time like until 2020 or something.

@alexcallard That is very interesting (and surprising). I think this is the negative side affect of working with a framework that is basically a one man show...

If you're working for a company that requires LTS, perhaps a bleeding edge framework isn't something you should be aspiring to, although you can always freeze Laravel at an appropriate commit, or use L4.* which is already well established.

I'm sure Taylor does have longer term plans, but right now L5 is probably not on the radar for you.

This is a big concern for me too. We charge our customers a lot of money for big, custom web apps and they want to know that said app is going to maintained and secure, etc. for a good few years after the investment.

My experience with open source communities is that the newer versions get most of all of the attention. Laravel releases a new major version practically annually, which is great, but if I build an enterprise app today in v5, will v7 be out two years from now and v5 patches and updates virtually abandoned by the community? We can already pretty much see this with v3. v4 is still getting a little bit of love, but v3 is pretty much a ghost-town already, less than three years since it was released.

There are some things I really like about Laravel, and we'll likely use it for some smaller jobs. But any bigger build with a long life, I worry about the LTS issue. If the Laravel community really wants Serious Biz to build apps on the framework, there has to be some assurance that a version will at least get security patches for a full 5+ years. That seems like an eternity to us code monkeys who watch the bleeding edge, but to the people signing the checks, it doesn't seem like a very long time.

My $.02!

@ashammari it is a fair question. I've asked the same question. Also the single point of failure is Taylor and at a certain point there is a scalability issue. That said, not sure there is much out there that is as well suited for small dev shops in the PHP community and right now it works. All the more reason to embrace the SOLID, etc. dev principles taught here at Laracasts so if somewhere in the future we need to port the app is relatively modular. @JefferyWay I suspect you've thought about these questions. Any insight given your experience in the dev / PHP community? I suspect there is no clean answer, kind of it is what it is and make the decision based on what you know now?

My opinions side mostly with open source and forward moving projects.

Before my current job, I was working with 5 year old code. The best way I could describe that code base would be Windows XP. It worked, did the job but had so many patches that fixing anything would break more things than would get fixed. I realize that corporations want to limit expenditures for applications but in reality they are only asking for heartache and need to be educated better. Too many times they have been told that it will be fixed yet without the understanding that the fix isn't supplying a sound foundation. I know this sounds like a dream scenario but 5 years support from an open source project that keeps moving forward is really hard to ask. I guess .Net would actually be more fitting for enterprise level long life application.

I also think that if an agency is getting paid big money to develop something on top what they got for free yet still require legacy support I think it's only fair that the agency invests some of their revenue into said open source project.

Regarding supporting older php / *nix versions. My opinion is that the framework should support the oldest possible versions without sacrificing security and a majority of functionality. However, corporations should also learn that upgrading to newer versions of software means better security. But, yet, they also will think about the cost of updating the applications to the newer versions of server software.

@ questions I'm not involved with larvel so the following are guesses. Actually, these questions need to be directed to Taylor directly rather than some forum.

Will Laravel support php 5.4.x for the next two years at least? we use Redhat 7 with php 5.4, since Redhat does not support mcrypt on php 5.5
Will Laravel stick with Symfony 2.x for the next two years?
For how long Laravel 4.x/5.x would be supported?
Is there a Paid support for Laravel?

a. possibly, L4 originally needed 5.3.7 except for some packages that used 5.4 specific code b. How much will Symfony change from now? c. I've read that L4 will only be getting security updates. No new functionality which makes a lot of sense. d. Nothing official that I know of. There are lots of hungry devs out there who depending on your budget will help you.

If you're looking for stability over a long period of time, for a large app or otherwise, you have to weigh your options. An app that has a long shelf life may better suited in something else, keeping in mind that you won't get the latest greatest updates/features. Having access to the latest is nice but it's also overhead work to keep up with the changes.

If there is a certain case where you can't simply update the framework and tech stack, or the app will sit for a long period of time (a la Enterprise), these type of scenarios would favour an LTS. A year goes bye in a blink in a large environment. Depending on your situation an LTS could be beneficial. Business partners don't want to hear that their investment goes down the tubes and they're vulnerable to security holes because the framework shut down since one guy was running it. They are buying stability. Some are ok with the risks of using the latest, but business leaders need to understand overhead involved in continued changes and have a plan for a scenario as described above.

Here are some options that might be more suitable to LTS.

In PHP, Symfony has LTS support: http://symfony.com/roadmap Version 2.7 (being released May 2015) will be supported by bug fixes until 2018 and security fixes until 2019. Symfony is operated by http://sensiolabs.com/ - a commercial entity. They offer services supporting Symfony but if SensioLabs were to fail I don't know what would happen to Symfony. Symfony components (not the framework) are used in some of the largest open source PHP apps available today, including Laravel, so I would assume they would continue to be supported.

If we look into other languages and comparable frameworks: In Python, Django 1.8, just released this April, supports 3+ years of support (2018). https://docs.djangoproject.com/en/dev/internals/release-process/#lts-releases Django seems very mature in this space, has a Django Foundation supporting the framework, and Python is very organized with the Python Software Foundation. Well documented and considered proposals via Python Enhancement Proposals (PEPs).

To my knowledge Ruby on Rails does not offer LTS support.

I'd avoid Node.js since the environment is rapidly changing.

Java EE has some well supported LTS type releases but requires a different mindset and, in my opinion, has higher overhead than PHP or Python.

Well, it looks like there'll be an LTS version :smile:

https://laravel-news.com/2015/05/laravel-announces-v5-1-will-be-lts/