>>125
I don't know what all that follows from, but if you have to demonstrate everything from the ground up every time you want to discuss a topic you're probably not talking to the right people. For example, it's more productive to have a discussion about programming with experienced programmers than with goldfish.
If you're introducing something new or even a new combination of features in a PL, then sure, by all means you'd better demonstrate it. But if there's an established language or technique which improves quality and productivity for certain types of problems then demanding a personal demonstration is just willful ignorance. Besides, if I write the code, you will not learn much at all. The blub argument isn't merely about ones ability to judge a language or technique, it's about ones inability to assess it due to ignorance,
having failed to fully understand the demonstration.
Blub isn't really a serious problem, except that it keeps people writing terrible PHP, Ruby and Java code. I see a lot of this and think "if they'd only try something that made them think differently they'd come back and write the best blub of their lives!" Getting back to the demonstration, it usually looks like more work even if just conceptually. The hill is a little steep but not as high as it seems, and the valley on the other side is quite deep. But you may never know. I can show you pictures, but you need to travel there yourself to understand its relationship with your side of the hill.