Comments on: Frameworks and the Break-Even Point http://kohari.org/2008/04/01/frameworks-and-the-break-even-point/ Organized chaos Fri, 05 Sep 2008 19:19:52 +0000 http://wordpress.org/?v=2.6 By: Nate Kohari http://kohari.org/2008/04/01/frameworks-and-the-break-even-point/#comment-88 Nate Kohari Fri, 23 May 2008 17:23:48 +0000 http://kohari.org/?p=82#comment-88 None of the post was a joke. The jokesters just came to fill out the comments. :) None of the post was a joke. The jokesters just came to fill out the comments. :)

]]> By: Sameer Alibhai http://kohari.org/2008/04/01/frameworks-and-the-break-even-point/#comment-87 Sameer Alibhai Fri, 23 May 2008 17:18:19 +0000 http://kohari.org/?p=82#comment-87 I'm sorry, I'm new to DI. Can you please explain to me which part of this post is the "joke" and which part is true? I’m sorry, I’m new to DI. Can you please explain to me which part of this post is the “joke” and which part is true?

]]> By: Matt Blodgett http://kohari.org/2008/04/01/frameworks-and-the-break-even-point/#comment-86 Matt Blodgett Thu, 03 Apr 2008 19:03:34 +0000 http://kohari.org/?p=82#comment-86 Great post, Nate. "Maybe it’s just me, but those 4 lines of code (complete with IntelliSense!) are much simpler than the 20 or so lines of XML. You’re using a big powerful IDE, so why not take advantage of it to set up your DI framework?" Yes! Yes! Yes! I couldn't agree more. Great post, Nate.

“Maybe it’s just me, but those 4 lines of code (complete with IntelliSense!) are much simpler than the 20 or so lines of XML. You’re using a big powerful IDE, so why not take advantage of it to set up your DI framework?”

Yes! Yes! Yes! I couldn’t agree more.

]]> By: Not John Slater http://kohari.org/2008/04/01/frameworks-and-the-break-even-point/#comment-85 Not John Slater Thu, 03 Apr 2008 17:32:17 +0000 http://kohari.org/?p=82#comment-85 Guess again. Keep up the good work. Guess again.

Keep up the good work.

]]> By: Nate Kohari http://kohari.org/2008/04/01/frameworks-and-the-break-even-point/#comment-84 Nate Kohari Tue, 01 Apr 2008 18:58:26 +0000 http://kohari.org/?p=82#comment-84 @John: For once you post a comment with your real name! What's wrong, run out of other people to impersonate? *grin* @John: For once you post a comment with your real name! What’s wrong, run out of other people to impersonate? *grin*

]]> By: John Slater http://kohari.org/2008/04/01/frameworks-and-the-break-even-point/#comment-83 John Slater Tue, 01 Apr 2008 18:56:04 +0000 http://kohari.org/?p=82#comment-83 Good one. You had me fooled captain refactor. Good one. You had me fooled captain refactor.

]]> By: Nate Kohari http://kohari.org/2008/04/01/frameworks-and-the-break-even-point/#comment-82 Nate Kohari Tue, 01 Apr 2008 14:53:22 +0000 http://kohari.org/?p=82#comment-82 @Issac: I take it you don't agree. :) @Issac: I take it you don’t agree. :)

]]> By: Isaac Mendenhall http://kohari.org/2008/04/01/frameworks-and-the-break-even-point/#comment-81 Isaac Mendenhall Tue, 01 Apr 2008 14:50:44 +0000 http://kohari.org/?p=82#comment-81 I read your post three times before I realized it was April Fools Day. Good one! I completely fell for it. I read your post three times before I realized it was April Fools Day. Good one! I completely fell for it.

]]> By: Nate Kohari http://kohari.org/2008/04/01/frameworks-and-the-break-even-point/#comment-80 Nate Kohari Tue, 01 Apr 2008 13:39:01 +0000 http://kohari.org/?p=82#comment-80 You don't need to (and in fact you shouldn't!) create a new kernel each time you activate something. Most applications only ever need one kernel, typically created during startup. I actually have an idea in the works for automatic module installation. Basically, it would search through all loaded assemblies for types that inherit from IModule and install them into a kernel. I've also been tempted to create a static facade that would provide easier access to a kernel instance. I've avoided it because the purist in me thinks it's a bad way to go about DI -- since it couples all of your code to the static facade. I'll probably add it before 1.0, but with a warning in the documentation that There Be Dragons. :) You don’t need to (and in fact you shouldn’t!) create a new kernel each time you activate something. Most applications only ever need one kernel, typically created during startup.

I actually have an idea in the works for automatic module installation. Basically, it would search through all loaded assemblies for types that inherit from IModule and install them into a kernel.

I’ve also been tempted to create a static facade that would provide easier access to a kernel instance. I’ve avoided it because the purist in me thinks it’s a bad way to go about DI — since it couples all of your code to the static facade. I’ll probably add it before 1.0, but with a warning in the documentation that There Be Dragons. :)

]]> By: Justin Etheredge http://kohari.org/2008/04/01/frameworks-and-the-break-even-point/#comment-79 Justin Etheredge Tue, 01 Apr 2008 13:31:33 +0000 http://kohari.org/?p=82#comment-79 I started using ninject precisely because it lacks xml config. Binsor is also an option, but I thought that ninject overall was quite simple to use. Now, we just need a way to define a standard kernel and module so that ninject can have a static "Get" method so that you don't need this line "IKernel kernel = new StandardKernel(new TwitterModule());" every time you want to activate something. Then the debt would be even lower. I started using ninject precisely because it lacks xml config. Binsor is also an option, but I thought that ninject overall was quite simple to use. Now, we just need a way to define a standard kernel and module so that ninject can have a static “Get” method so that you don’t need this line “IKernel kernel = new StandardKernel(new TwitterModule());” every time you want to activate something. Then the debt would be even lower.

]]>