Ask HN: Do you use print() debugging?

Or puts in Ruby, or System.out.println() in Java, or echo in shell scripting and PHP, or printf() in C/C++, or console.log() or worse, alert() in JS, or the equivalent in whatever language and environment you're in?

I'm primarily a JS developer, and the first debugging tool I use is usually alert() or console.log(). I usually just have to locate a single line that is past the point something has gone wrong, and I can usually very quickly binary search between this line and the event handler to find the offending line and fix it. Of course, the binary search is guided by intuition and if I don't find it after a few alerts I will start setting breakpoints in the debugger and/or stepping through instead, but somehow I've always seen the debugger as this big, heavyweight sledgehammer I don't whip out unless I'm actually stuck.

I do the same thing in all of the above mentioned languages, and in fact have never used a debugger at all for Ruby or the Pythons, though I assume they exist (nor for shell scripting or PHP, but do those exist?). I consider myself at least a proficient coder, and in particular have substantially used all the above languages (I've also used Scheme in a class, but nothing substantial). One of the best coders I know, who mainly codes in Ruby, also often first tries puts debugging. Yet I've also been ridiculed (in a nice way) by another friend for even suggesting to someone that they try to debug what seems like a simple mistake with a few quick print statements.

In a way I agree with them--print statements are a crude way to debug. But they're quick and dirty and get the job done and aren't hard at all to clean up. So I wonder: how many others debug with print statements first and a debugger second?