But it's really about what the community wants. YAF is a community project.

Anyone have any thoughts? Are people already using MVC instead of Webforms?

We're not --

It could make YAF more light-weight and also make TDD a breeze.

Do you suggest to have a MVC version of YAF along with a Webforms version? or switch to MVC completely?

Users don't care at all is it MVC or not. They don't care what is inside at all.

Is an application a developer toy or an application for people to use?. That is a real question.

The main programmer's principle - don't touch what works fine.

The most conservative part of any application is the data layer. It's the last thing to touch for a professional programmer.

MVC is an option for new applications and a supplimentary option for existing ones.

MVC requires times more time than a WebForms application. In some cases it's useful - i.e. e-commerce and if I need write such an applicaton my choice will be surely MVC..

Let's imagine you have a WebForms application.

Your users need new features, but instead of it you waste your time rewriting it to MVC...

I sure that if you make things like this - your users will finally pay cold feet to it, because competitors make it more and more attractive for users, and you can offer them only new bugs in a previously working functionality .

I feel more comfortable with Webforms right now and there's no doubt rewriting to MVC takes a hell of a lot work and will probably be buggy for some time.

But you know, It's like there has been an MVC paradigm in most of open source projects and I think at least it's necessary to discuss about it.

But as I said, performance and functionality should be the first things to consider.

But YAF code is and will stay Webforms for now... But I'll be chipping away at all the dependencies and making the code much more abstract and testable. Basically, moving to MVP (Model/View/Presentation) which will allow testability. WCF will be a much bigger part of the next version of YAF, as well.

The goal is to get the code out of the webforms. It's a mess right now.

Oh yes, already :roll:

I have 3 datalayers and I have no plans to rewrite them. And I have some features which I need to implement. Currently YAF has incompatibilities with my datalayers, I need to modify a lot of code and it's much easier for me to make a fork which will work with my datalayers out of box and will have features needed for me. So don't take me into account.

MVC is not a bad thing, but requires too much time for development, but if behind of paradigm will be perfomance and moderate system resources it's quite a deal 😉