MyBB Community Forums

Full Version: choice of MVCand no MVC for plugins?
You're currently viewing a stripped down version of our content. View the full version with proper formatting.
Pages: 1 2 3 4 5
Given the plugin system is the biggest entry point for developers and for me at least rapid development often outweighs the need for exact code (I usually get a feature up then overhaul it later)

I have found that MVC can be a hindrance in this and many cases. For an example I have been writing a plugin to integrate Joomla and myBB but decided to scrap the plugin for the mere fact that when I started writing the Joomla side of things and user sync from Joomla, I found it overly complex and a waste of time. I want to move away from Joomla so see no need in persisting with the code for now.

Essentially just to write the sync page I had to create, about 4 or 5 files, controller, model, and the various view pages. All I wanted to do was a small script with minimal HTML output just a list of conflicts and the amount merged, this is something that could be so easily accomplished with the current myBB system without having to create a multitude of pages.

I am therefore hoping that although myBB is moving to an MVC model, that this is not enforced on plugin writers. I would hate to have to go through the hoops you often have to go through with Joomla just to get something simple up and running, there is often a use case for simplicity and speed over technically clean code.

I want to bring my entire site over to a myBB based one, but am hesitant to start writing additional plugins if it means I am going to have to completely rewrite them to fit into a rigid MVC setup.

The current myBB setup in my view is actually really good and easily has the best plugin system of the forums I have used, I am wondering if the change toward 2.0 is change for changes sake, when such a good platform already exists.

For mine, the most important things are a fast forum, and one that is easy to modify, myBB currently meets those requirements.

/rant

cheers
Dan
We haven't planned the Plugins system yet but when we do, we will take that into consideration.
Exactly what I fear about framework for plugin authors.

Quote:The current myBB setup in my view is actually really good and easily has the best plugin system of the forums I have used,

Yup...unfortunately MyBB is definitely going to use Yii for 2.0. They get rapid development and the community gets to learn a framework. Sad
(2012-08-06, 11:51 PM)labrocca Wrote: [ -> ]Exactly what I fear about framework for plugin authors.

Quote:The current myBB setup in my view is actually really good and easily has the best plugin system of the forums I have used,

Yup...unfortunately MyBB is definitely going to use Yii for 2.0. They get rapid development and the community gets to learn a framework. Sad

Yes this is my great fear. However I can still see an MVC framework driven application still allowing plugins in their current format, because there is no need to be so strictly granular.

I do question the wisdom of basing something on a 3rd party framework, it assumes support for and continued development of that framework and going forward if the yii framework is upgraded and changed, it could force large scale development of the mybb project outside of the usual release scope and timeline.

the other issue I have is that by relying on a 3rd party framework, you are throwing yourselves at the mercy of that development team and their license. So if there is a major bug or security issue found within Yii you will need to wait for the developers to patch that and release a new version of the framework before you then have to develop and release a version of myBB with the new framework version. Whilst I can appreciate Yii is pretty good and I have played with it a little. I just think a project the size of myBB and to continue to be a player, using a 3rd party framework may not be the ideal solution.

The saving grace is that the mybb team have done such a great job so far, there is no reason to believe that their implementation of mybb may be just as smart and therefore shield us from both the learning curve and the aforementioned potential issues
You're stating everything I think as well. I think MyBB is taking a risk but it's their call and we can only hope our trust in them is not wasted.
We have not 100% decided on Yii, we have been looking into other options. However, priority has been given to 1.8 ATM.
Would be interested to see a public poll by team asking the following.

For 2.0 development which would you prefer:

1. Start with fresh code
2. Use a framework
3. Rewrite portions of current base
(2012-08-07, 04:09 AM)labrocca Wrote: [ -> ]Would be interested to see a public poll by team asking the following.

For 2.0 development which would you prefer:

1. Start with fresh code
2. Use a framework
3. Rewrite portions of current base
Myself, I think they should pull an IPS and create just a base that pretty much any MyBB version will use (a skeleton of the software, if you will, with crucial stuff like user functions, posting methods, viewing methods, etc), and then build off of that with each release. They can always improve upon that core, but new features could be added on as versions move forward and tap into that base functionality.
(2012-08-07, 04:32 AM)Josh H. Wrote: [ -> ]
(2012-08-07, 04:09 AM)labrocca Wrote: [ -> ]Would be interested to see a public poll by team asking the following.

For 2.0 development which would you prefer:

1. Start with fresh code
2. Use a framework
3. Rewrite portions of current base
Myself, I think they should pull an IPS and create just a base that pretty much any MyBB version will use (a skeleton of the software, if you will, with crucial stuff like user functions, posting methods, viewing methods, etc), and then build off of that with each release. They can always improve upon that core, but new features could be added on as versions move forward and tap into that base functionality.

Agreed.
I think that is a good idea and where software in general is headed.
The company I work for is slowly moving toward that model, the core of the system is there that is basically a funnel for handling plugged in logic.

Then everything else is pluggable, that way it allows the user base to easily latch on what they need, and say a bug is found in one part of the software, it can fixed in that part without affecting any other area of the software, because it's all based on a plugin.

In the past the software has suffered because it required large scale fixes in the same areas or parts that depended on each other.

So in a sense, Threads, forum display etc could all be pluggable components
Pages: 1 2 3 4 5