Slashdot Mirror
Wolfram Alpha Rekindles Campus Math Tool Debate
An anonymous reader sends in a story about how Wolfram Alpha is becoming the latest tool students are using to help with their schoolwork, and why some professors are worried it will interfere with the learning process. Quoting:
"The goal of WolframAlpha is to bring high-level mathematics to the masses, by letting users type in problems in plain English and delivering instant results. As a result, some professors say the service poses tough questions for their classroom policies. 'I think this is going to reignite a math war,' said Maria H. Andersen, a mathematics instructor at Muskegon Community College, referring to past debates over the role of graphing calculators in math education. 'Given that there are still pockets of instructors and departments in the US where graphing calculators are still not allowed, some instructors will likely react with resistance (i.e. we still don't change anything) or possibly even with the charge that using WA is cheating.'"
I'm worried about all these highfallutin complex math equations enabling this thing to evolve into skynet, and these guys are worried that it's going to help people with their homework!?! *adds another layer to tinfoil hat*
This space intentionally left blank
Exactly. This thing isn't going to help you pass an exam. As it is, you can use an engineering calculator to solve equations the same as using WolframAlpha. Neither are going to be allowed on exams and I never had any homework in my engineering courses where it was okay to omit each step used to reach the solution.
Wolfram Alpha also allows you to see all the steps that lead to a result, so your solution is not going to work.
Since graduating, never in my career have I encountered a situation where I had to solve 25 simple yet unrelated problems in under an hour without the use of references or collaboration.
So you shouldn't have to know how to solve a given problem yourself, in a vacuum, because in the "real world" we have reference books and other people to collaborate with.
Now, apply that logic to the whole population of potential collaborators / reference book writers.
Everyone now assumes there's someone else to collaborate with. But who? Since anyone you might collaborate with also believes the above, they won't know how to solve the problem either. Who would write the reference books? Same problem.
At some point the buck stops at the individual. You need to know how to solve the given problems, by yourself. That's why we do tests, and that's why you (generally) can't collaborate or consult references.
Oh, you poor, naive person.
Let me introduce myself. I'm from the real world. Let me explain how things happen here.
We have to deal with tricky problems. Sometimes, a function has more than one formal integral, and some forms are more appropriate than others in different situations. Good luck coaxing your CAS into giving you exactly what you want.
We have to deal with deadlines. If you can solve a problem in two minutes on paper, that's usually quicker than loading up most software packages and trying to get your equation into the syntax of the system. (Naturally, no two systems use the same syntax.)
Even worse, we have to deal with software licensing. Mathematica and Matlab ain't cheap. Software vendors try to argue that you're a commercial institution, not a research institution, so they can gouge you for licence fees. Cross your fingers and hope that there is a small enough number of people using the software concurrently so that you can get in. Otherwise, you're screwed.
sub f{($f)=@_;print"$f(q{$f});";}f(q{sub f{($f)=@_;print"$f(q{$f});";}f});
Try a query for "electricity production". Alpha doesn't know what to do with that, either, because Alpha doesn't have any data on it... yet.
In the feedback form at the bottom, request that somebody go out and collect it. It'll happen.
-- Somebody who actually reads the feedback.
It is difficult to determine who is cheating in course work and who is supplying the most input with team work. At least with an exam there is a test of knowledge and understanding.
Yes, I already said that, which is why I said that I had no better alternative, and was simply pointing out that a typical exam isn't just testing your knowledge and understanding of the subject, it's also testing your exam-taking ability.
Come on Chris tell the truth. It's your friend who's good at exams and you who understand everything but can't, no matter how much you try, pass the damn things.
Truthfully, I'm great at taking exams. I could even pass ones when I didn't really understand the material that well. That's not bragging, because that ability is basically useless in the real world.
It is no wonder the middle of the road conscientious but not too bright are always in support of course work and ever ready to damn exams.
Be honest -- you're good at taking exams, but are too arrogant to admit that this doesn't necessarily mean you're the greatest at the subject matter, and too self-centered to consider how this affects anyone but yourself.
Besides, if you actually pay attention and read what I say I'm not damning exams. If this was a test in reading comprehension... So, go get a point then come back.
The enemies of Democracy are
I couldn't disagree more. I do this every day. Of course, I don't do 25 problems in one hour, but I do 25 problems/hour, i.e. solving a simple problem in a couple minutes many times per day.
Yes. Occasionally throughout your day you encounter a simple problem that can be solved in a few minutes. Of course you don't do them all back to back, of course they aren't all isolated and artificial, and of course if you go slightly slower on one such that it took you a cumulative hour and one minute to finish the 25 problems, nobody shouts "time!" and forces you to stop such that the problem remains unsolved. Not on a one-hour timescale, anyway.
In other words, what you do is nothing like taking an exam.
If you need to take ten minutes, possibly digging through reference materials, to solve a simple problem
Having artificially simple problems (such that 25 can be completed in an hour long exam) is exactly one of the things I was saying was unrealistic about exams. In real life, problems are complex, sometimes but not always decomposable into simple problems suitable for an exam, but in any case more like a class project than an exam.
And if you never encounter a simple problem that requires a reference, as in you can contain all the knowledge ever required for your field in your head at one time (example of simple reference-requiring problem: What's the opcode for a MOVDQA), then the job itself is pretty simple.
The enemies of Democracy are
Not only this, it's also testing the ability of your professor or whoever to create a valid and reliable exam in this format. Not everyone can do it, and for a lot of people, the temptation to include trick questions is very high.
When the axe came to the forest, the trees said, "Look out - the handle was once one of us."