Why are there so many bastard problems?

So, I don't think this will make me seem like *less* of an ogre, but I thought I'd put down some of my reasoning behind your homework sets.

The homework is really, really difficult. I won't pretend otherwise. In fact, it is comparable to what physics majors at places like MIT would see - meaning if you are getting through it, or at least understanding the solutions after the fact, you are learning physics as well as anyone anywhere.

(Your exam problems will not be as hard. Not even close. I'll post some sample problems tomorrow to give you an idea.)

Out of every (say) three problems I give you, I usually expect one of them to be pretty easy or very similar to an example problem in the book or online. A second one usually has a painful but straightforward brute-force solution, and a more mathematically sophisticated but shorter solution. Then, there is usually the bastard problem, which is just going to suck.

So why the bastard problem? One, I can't really see how much you have learned, and whether you can apply it in unfamiliar situations, if you ace all the problems. In that regard, it is OK, and expected, if you don't complete the whole thing. On the toughest problems, you will be well-rewarded for just trying something sensible, I tend to grade the 'bastard problems' pretty leniently. If you don't get those problems, you are still going to end up just fine - you will not need to solve all of them to get an "A." With the partial credit scheme based on the homework template, even just setting up the bastard problem, without solving it completely, can get you 75% partial credit.

Two, it is still very valuable if you try the problem and get stumped. There is heavy partial credit for trying, but that is not entirely the point either. If I just solve the really tough problems in class for you, without you having tried them, they will either be hard to follow or seem deceptively easy. So the theory goes, at least.

Three, for most of you it is going to 'click' at some point, particularly once we learn some new techniques in the coming weeks. Most of this week's homework can be solved in a few lines with the benefit of hindsight we'll have at that point. It is just hard in the beginning, period.

Four, you are free to (and encouraged to) collaborate. This is not a justification for making the homework difficult, but a reminder that you should help each other get through it when you can ...

Finally, part of the point is that our students often get to 300 level physics classes and get creamed. I'd rather it be reversed, at least for honors students - by the time you get to PH301, you should manhandle it.

So how do I calibrate the homework? I solved all of this week's problems on Friday night and just made a second pass to double check. If I can't do the whole homework set in about an hour, using only the same things we covered in class, it is too much.

As an aside: part of this is about time management as well. Is it OK to spend time x and get 90% credit, or should you spend time 2x to get 95%? Multiply that by 4 or 5 courses and life outside classes ... don't spend three extra hours on physics for 5% more if you can spend it on calculus for 25% more.