I would have to imagine that the interesting hacking comes into effect when, instead of writing the interface by hand, you abstract the interface into code that makes interfaces trivial.
In the words of Jeannette Wing, the two A's of "computational thinking" are "abstraction" and "automation"
In other words, interesting hacking occurs when you realize that a certain task can be automated (e.g. writing UIs or administering servers), you create a useful abstraction for it, and then you write the code that performs the task largely automatically.
This is why pg champions Lisp as a great language for startups: Lisp is perfect for building other languages because of its powerful macro facilities, among other things.
I guess what I'm saying is that "startup hacking" is as interesting as you make it. Sure, you can toil away writing similar code over and over again to create various facets of the UI, but the more interesting thing to do is to write the program that writes that code for you.
In the words of Jeannette Wing, the two A's of "computational thinking" are "abstraction" and "automation"
(from http://www.cs.cmu.edu/afs/cs/usr/wing/www/ct-short.pdf)
In other words, interesting hacking occurs when you realize that a certain task can be automated (e.g. writing UIs or administering servers), you create a useful abstraction for it, and then you write the code that performs the task largely automatically.
This is why pg champions Lisp as a great language for startups: Lisp is perfect for building other languages because of its powerful macro facilities, among other things.
I guess what I'm saying is that "startup hacking" is as interesting as you make it. Sure, you can toil away writing similar code over and over again to create various facets of the UI, but the more interesting thing to do is to write the program that writes that code for you.