Talk:The Future
From APIDesign
You mention compilers here quickly. I could actually imagine that a NetBeans plugin might be the most natural place for API design/development pointers to exist.
For example, I could imagine an "API Design >" menu item, that contains things like, "Check for public method not declared final" or "Check for removed public methods" - or, these things could come up as warnings under a "API Design Warnings" section of errors/warnings.
In fact, if you believe (which I don't think you do, based on your text) that one could create an open-source NetBeans API Deisgn & Development Plugin, this section would be the perfect place to launch such a thing and forward readers to a website with more information.
--Dmkoelle 03:07, 23 April 2008 (UTC)
Actually I have written a hint to check for one of the most common API errors - hidden class. A situation when you have a method or a field in a public class that is package private. This was relatively easy to write, and there could be more hints like this.
On the other hand, this is just an IDE support. I'd like to have "objective" one. E.g. a check during build time, on a server.
--JaroslavTulach 22:42, 25 April 2008 (UTC)
I'd have to agree with Adam. While I recognize that the epilogue is your place to expound on your own personal thoughts, this section seems a little too far removed.
--Dmkoelle 03:02, 23 April 2008 (UTC)
This entire section seems too philosophical. I might skip it all.
Saying that you hoped your book would be like Newton's Principia Mathematica is sort of like a composer saying that he hopes his work will be another Beethoven's Ninth. It might be better to leave such comparisons to those who review your book!
--AdamDingle 17:17, 20 April 2008 (UTC)
Sorry, but I also agree. This section, and ones like it, are unnecessary. It's like you're writing a book on auto repair and including chapters on fluid dynamics and metallurgy.
--Richunger 05:36, 26 April 2008 (UTC)
I can't help but think that the two "Note" call-outs in this section, so close to the end of the book, will be glossed over by most readers. Could both of these Notes be shortened?
--Dmkoelle 03:09, 23 April 2008 (UTC)
I would note that one reason people are afraid of foreign code because they're afraid they'll break something. Having good test coverage can give them the confidence they need to work on foreign code without this fear.
- Ne, ne. Not in case of students as far as I know some. Most of them does not even know what a test is. Another thing to fix in the education.
--TomWheeler Wed Apr 23 20:38:48 CDT 2008
It's not necessarily fear, either. Writing new code is more fun!
--Richunger 05:29, 26 April 2008 (UTC)