News for hackers. Stuff that matters.
Re:What Doesn't Seem Like Work? by MarkPNeyer on Friday January 16, 2015 @06:44PM (#8902528)
i was a physics major, too. the student in our year who was best at math had trouble with the "hand wavy" stuff we'd do. "well, this term is basically zero, so we'll just remove it" and then she'd pull her hair in frustration, like "how can you just do that? how is it still valid?"

it wasn't enough to tell her "well it still gives us the valid results" - she had to be able to interpret every portion of an expression and then make intuitive sense of it.

at first i was annoyed with this habit of hers, but eventually i started doing it to. for a lot of really math-minded people, hand waviness is anathema, and physics was full of "2 + 2 = 5 for sufficiently large values of 2, so we'll just assume that to make things easier"

Re:What Doesn't Seem Like Work? by _pius on Friday January 16, 2015 @07:21PM (#8902676)
"well, this term is basically zero, so we'll just remove it" and then she'd pull her hair in frustration, like "how can you just do that? how is it still valid?"

I've tried to teach friends with that hangup and found it difficult. Those people weren't good at math, though. I feel like with mathematicians, you should just be able to explain with limits.

Re:What Doesn't Seem Like Work? by sedachv on Friday January 16, 2015 @07:45PM (#8902779)
> I feel like with mathematicians, you should just be able to explain with limits.

Assuming the limit exists without actually proving the sequence converges is going to produce total junk. Jaynes spent a good portion of Probability Theory illustrating this.

Re:What Doesn't Seem Like Work? by _pius on Friday January 16, 2015 @07:48PM (#8902794)
I'm just talking about using the concept of limits as a teaching tool, not assuming a particular limit exists.
Re:What Doesn't Seem Like Work? by sedachv on Friday January 16, 2015 @08:40PM (#8902954)
Yes, this is why the mathematicians are pulling their hair out! You can't just drop terms when you feel like it.
Re:What Doesn't Seem Like Work? by TeMPOraL on Saturday January 17, 2015 @09:31AM (#8904374)
I feel that the right approach (at least it worked for me) is to tell the matematician - "see, this term is very close to zero for all practical inputs we can see in real life (proof of which will be left as an exercise for you), so - as you can see from the structure of the equation - we can safely drop it without introducing any significant error to the end result. Yes, the result will be a little bit less accurate; no, no one will notice. And if you happen to encounter a high-energy (or whatever) problem, you can always go back and use the full equation". Explicitly accepting that each handwave introduces errors and imprecision but that sometimes those errors are below what we can measure (or care about) should be acceptable enough for a math-minded person.