AdefWebserver wrote
I am also a professional programmer who uses WAP 8 hours a day (I have to at work). We can all agree that WAP is great. I recommend the dynamic compilation method because it is easier. it has less steps. Every configuration step a person has to make is a potential stumbling block.
Do you believe me when I say that I have talked to over 100 ASP.NET programmers about DotNetNuke and over half of them has tried to make a DotNetNuke module and only 3 have done it before we made the changes in the past 2 years?
You can convert the entire DotNetNuke core to be one big dynamic module and it will run just the same. We could use the Visual Studio Team System and have 100 developer's working on it. The only reason we don't do it is the same reason we don't convert to C#. The end product would be just the same.
What is there to be gained by insisting the method I am advocating does not work for large projects?
Ok first I think we got a little of base here... :) If I look back at the posts, the whole discussion sounds kind of funny :) At the end people will choose what suits them best.
Now...
I believe you completely about that half of developers... I was one of them a few years back struggling with DNN and my development first steps... As mentioned above, I really am happy that you provide guidance you are providing.... believe me, it's great!
The whole point of my struggle generally is to show you and others that there are more advanced users out there, and that some focus should be given to them as well. Because as much as those half people are happy with the way you advise to develop modules, the other half are sometimes on "scared" away because they think that what you advise is all there is - and that that is the limit of DNN (which is not true).
Also, my point is not about WAP, nor dynamic compilation, nor converting DNN to be dynamic compilation based... I personally think (although I don't say it much because I don't have the concrete numbers) that standard compilation is faster for internal development... but in the end it comes down how you structure your solution more than which type of project you put in (my preference as you know is WAP + IIS for the personally fastest dev. experience).
My complete point of this conversation was not to say that your method was bad (sorry if it turned up like that)... only that for larger projects (in my personal experience and the experience of the people who are using my approach) there is a faster and more productive alternative, and that the biggest difference is often shown to people who work on bigger module solutions.
Although I am tempted to say that if you only move DNN core from the solution that we would get the similar productivity gains, only in that case you automatically start thinking (at least I do) of WAP-like project and not Website one for the module itself.
What do you think? Shall we call it a truss?