5 minutes ago, DeathBy​VisualStudio wrote

@Ian2: ROTF! That rack wasn't placed near the "Coming Soon! Windows 8 RT tablets"  display was it?

It's just another way to be on the bleeding edge.

1 hour ago, Maddus Mattus wrote

@DeathByVisualStudio: just make em take the test in a Faraday cage, problem solved.

Right, and next year Tesco just adds bolt cutters to their  'back to school'  range.

Because of this, except some subject that requires specialized calculators, most subjects in universities here also require to use HKEA approved ones.

18 hours ago, Ian2 wrote

*snip*

Right, and next year Tesco just adds bolt cutters to their  'back to school'  range.

Getting up in the exam, and cutting your way through the wires that are embedded in the walls to breach the faraday cage is more likely to be noticed by an invigilator than sending a SMS message under the table.

Take the books with you, text with your friends,..

This is how I solve problems now, why should it be any different at school?

Why should I solve the problems on my own, when I can rely on the experience of my peers and they on mine?

For example, it a programmer can't even remember basic syntax, imagine how much time they should spend on referencing books when coding.

There's also lots of fundamental equations / laws on Mathematics or other subjects. If you can't remember them, while it's true that you can always find them from books, you're doomed to be an underperfomer.

That said, some subjects (most in civil engineering discipline) are changing to "openbook exam" now. The new tests are more difficult because you can no longer gain points by reciting paragraphs (and it is good).

Do you hire a programmer because he knows the syntax of c#, vb.net or javascript or do you hire him because he can solve your problems?

Does it really matter if he uses C#. VB.Net or javascript to solve the issues?

The syntax is not important, the way you solve problems is. And the only one who can teach you how to solve problems is yourself. Your teacher can help you discover these problem solving abilities, but you will have to find your own unique way. Otherwise you are just a copycat and never be as good as your teacher.

2 hours ago, Maddus Mattus wrote

Why not allow them to cheat?

Take the books with you, text with your friends,..

This is how I solve problems now, why should it be any different at school?

Why should I solve the problems on my own, when I can rely on the experience of my peers and they on mine?

The problems you are trying to solve are mostly about finding a solution to the problem. Exams are mostly about trying to find an objective measure by which the pupil can be judged.

If you come up with twenty exam-level questions and give them to two million students who are allowed to collaborate, chances are, you'll get twenty answers that are correct. But what you won't have is a way of being able to distinguish who was able to work out the answers from others who merely copied the answers from a forum dedicated to helping with the site.

Worse still, you'd get parents "helping" their kids during the exam if they're allowed the cheat. Clever parents would just help the kids directly. Rich parents would help their kids by outsourcing to clever professionals. And poor/stupid parents - well, their kids would probably fail the exam.

The only way to really rectify this is rather than letting the students "cheat", you let them have passive access to online reference material such as wikipedia.  But the problem here is that the Internet is always a two-way street. The very act of performing a search, for example, is an act that is sending information to the Internet. This means that effectively locking down the exam so that pupils can use the Internet but can't send the questions off to be answered by their rich/clever parents is impractical.

I agree.

I put it a bit to black and white.

13 minutes ago, Maddus Mattus wrote

@cheong:

Do you hire a programmer because he knows the syntax of c#, vb.net or javascript or do you hire him because he can solve your problems?

Does it really matter if he uses C#. VB.Net or javascript to solve the issues?

If the projects in my company are all in C#, would it be desirable to hire someone who only know unmanaged C++ (remember, rules like constructor/destructor execution sequence and multiple inheritance differs)? How good do you think the one who knows C++ but doesn't understand javascript and DOM debug a web application?

And a more difficult one, if the projects in my companys are only using MFC, would it be desirable to hire someone who only know C#/VB.NET?

13 minutes ago, Maddus Mattus wrote

@cheong:

The syntax is not important, the way you solve problems is. And the only one who can teach you how to solve problems is yourself. Your teacher can help you discover these problem solving abilities, but you will have to find your own unique way. Otherwise you are just a copycat and never be as good as your teacher.

If you don't know the syntax, sometimes you can't read/debug code.

Just imagine the shock when I see an one-liner that's a mix of LINQ plus anonymous functions. Books won't show you every example, and you can read it if the you can't understand the syntax, especially if some of the types involved in that query is coded as dynamic type. Possibly even Intellisense can't help you in the case.

Now try add extension methods to the mix.

Btw, the aim of traditional exams are to assess 1) how much you can remember from the lessons; 2) how well you can apply what you learnt. Calculators (or I should say, computers) that carries these information kind of defects the first aim.

Unfortunately, those skills are much harder to test than basic knowledge, and for that reason most exams still test only knowledge. I remember I had to construct an LALR parsing table by hand in a Compiler Construction exam, which tests maybe 10% how well you understand the process and 90% how well you were able to memorize the algorithm from the book.

I've always been in favour of open book exams, but good questions for such an exam are much harder and more time consuming to create (and often also to grade), so most teachers don't bother.

Of course, allowing unlimited access to the Internet is still not a good idea because you want to test understanding of the subject matter, not the student's ability to find a pre-existing solution to the problem (which is perfectly fine in many real-world scenarios, but not in an exam setting).

It's how they did it 25(ish) years ago when I was at high school and I'm pretty sure that's how they do it still.

Herbie

Why is it acceptable in real world scenarios but not in schools?

Isn't a school supposed to prepare you for the real world?

And if it's okay in a real world, shouldnt it then be ok in school?

Bit of a paradox.

In school, there are no new bits. Everything you're solving has been solved by someone before. The reason you're solving it now is to test your understanding of the underlying principles and your ability to solve those kinds of problems.

The ability to find the solution to a problem is not the same as the ability to solve it. Would you hire someone whose only skill is finding other people's solutions, but who is incapable of doing any creative thinking of his/her own?

How would you test someone's ability to solve a problem if you can't restrict the conditions of the test such that they can't ask someone else to solve it for them?

If you are asked to solve a problem during a job interview, is it acceptable to go to Google, and copy/paste an existing solution as "your" answer?

If you ask me "what is 5+4" and I paste that in WolframAlpha and tell you the answer is 9, have you learned anything about my understanding of arithmetic?

So unless it's designed to be open book exam, calculators that can be used to cheat should not be allowed.

That's where I don't agree with you.

Schools should (and are) be places where new stuff gets made up. Google was a project at a university, so was Facebook,.

I was to harsh with my original statement. Exact courses should be tested exactly. But the higher in education you go, the less exact and the more creative you become.

It's the analyzing the problem, breaking it down and creating solutions based on proven practices that we should teach in the form of projects at schools.

Anyway, this thread was about knives and guns in schools no?

13 minutes ago, Maddus Mattus wrote

@Sven Groot:

Schools should (and are) be places where new stuff gets made up. Google was a project at a university, so was Facebook,.

University, yes. But even then only really after the Bachelor's level. You need to learn to walk before you can run. The problem of exams and problem-solving skills applies to high school and elementary schools just as much as it does to universities.

You can't "be creative" in a subject without understanding the basics. And universities and schools still need a way to test if you understand the basics.

19 minutes ago, Maddus Mattus wrote

@Sven Groot:

That's where I don't agree with you.

Schools should (and are) be places where new stuff gets made up. Google was a project at a university, so was Facebook,.

But neither were made at primary schools or by A-level students.

I was to harsh with my original statement. Exact courses should be tested exactly. But the higher in education you go, the less exact and the more creative you become.

That is already the case, and is reflected by how courses are taught at different levels.

Primary school: What is 19 + 26 (exact answer required).

Secondary school (junior): If 2x + 4 = 8, what is x? (exact answer required with maybe one discretionary point)

Secondary school (senior): Solve the definite integral between 0 and infinity of (2/x^x)x dx (exact answer required, but with discretionary points for partial success)

University: Show that the cyclic subgroup C<P> is of length P-1 if and only if P is prime,  (multiple possible answers - most points are discresionary).

PHD: Make your own question and solve it, whilst demonstrating that the question was an interesting one that hadn't already been solved (all points are discretionary - indeed, even the number of points is discretionary).

But even then only really after the Bachelor's level

I was creative on the computer before I even graduated primairy school. So I don't think you should start so late. I think it should increase gradually with age.

Start off very exact and get more creative when you progress.

It's the natural cycle. First you want to learn how stuff is ment to be used, then when you have enough experience, you know better and you chuck the rules.

You can't be creative with the shackles of the 'basic understanding'. You have to let it all go, Neo. Fear, doubt, and disbelief. Free your mind.

If they are so creative that they can out smart your test, you should out smart them by creating better tests, no by dumbing them down.

22 minutes ago, Maddus Mattus wrote

@evildictaitor: I still say, let them cheat,..

If they are so creative that they can out smart your test, you should out smart them by creating better tests, no by dumbing them down.

Suppose you employ the ten greatest minds on earth to conjure the greatest and best exam question for primary school mathematics imaginable. Let's call that question Q.

Suppose A is a valid answer to Q.

Either nobody is able to determine what A is (e.g. Q is please factor this big number into primes and give them an 2048-bit RSA modulus), in which case all primary school children fail the exam.

Alternatively, one person in the connected graph of all people that all primary school children sitting the exams know and who is willing to cheat discovers A during the exam. That person makes the answer available on a forum.

Now Q can be answered by any child typing "answer to Q" into Google. So you've turned an exam about "can you solve Q" into an exam about "can you type 'answer to Q' into Google and copy the first page onto this piece of paper".

These are assessing quite different things - the first gives you some insight into the pupil's ability to do work. The second does not.

I think the best possible exam system is one where you don't examine the children - you merely ask their teachers how bright the child is. Unfortunately this is too subjective and subject to corruption to be made official, but it's the best way of getting a picture of the child about how good that person is at doing work, by allowing the teacher to take into account other behaviours, obstacles and so on that the child encounters.

But as long as you have written exams, where all children are asked to complete the same questions, you're going to need to prevent cheating, because for any question Q, no matter how ingenious or clever you are, you're probably not want to just be testing the child's legendary ability at typing the question into google and copying the result onto paper.

At that point, why bother teaching maths, geography, English, science, etc. when all you apparently need to pass any of the exams is just one skill: the ability to cheat.

And you don't even need to be creative to cheat. All you need is for one kid to figure out how to use Google and then tell his friends.

The ultimate question, would not be multiple choice.

The means by getting the answer, the calculation or reasoning, are just as much part of the answer as is the acutal outcome of the calculation. We used to get points for approach and bonus points for the correct answer (it took 45 mins just to write down the calculation).

You can only test factual knowlidge with written tests,.

You can take an English test by writing an essay.

Math tests with solving equasians, no googling is going to give you the complete calculation for solving maths. MAPLE will not give you the way it solved it, merely an answer. If you don't understand the matter, you would be stuck googling the entire test time and not completing the exam.

40 minutes ago, Maddus Mattus wrote

@evildictaitor: @Sven Groot:

The ultimate question, would not be multiple choice.

Who said anything about multiple choice? If Q is a deterministic question asked in an exam, and X is an complete answer to it scoring 100% (whether this be a piece of music composition, an answer with full annotated working to a mathematical problem, or a four page dissertation on the religious changes that happened under Henry VIII's rule in England), then all of the children in the class can submit X as their answer and receive full marks.

Hence, if anyone posts "The answer to Q is X" on the Internet, then typing "The answer to Q" into Google yields X - an answer that gives full marks in the exam.

Hence, it becomes possible to receive full marks in the exam by knowing how to Google, regardless on the complexity of the question.

10 minutes ago, evildictait​or wrote

Hence, if anyone posts "The answer to Q is X" on the Internet, then typing "The answer to Q" into Google yields X - an answer that gives full marks in the exam.

That would only be true if the question was a multiple choice one.

Just because you have the answer, doesnt mean you get the credits.

The approach to coming to an answer is demonstrates if you have learned and understand the material, not simply propping up the x = 3. 

Take programming assignments, for example. All the people who grade the assignment see is the final program. They don't get to see you working on it (sometimes they do because the people who grade often also help out during lab hours, but they won't observe necessarily every student).

How can I, as the grader, tell whether this is a program they wrote or copied from someone else. The only means of detection available to me are the blatant case where multiple people hand in the exact same program, or if I recognize a particular solution from somewhere. Another hint can be that the work is way above the student's usual level (which is not an immediate indication of cheating, but does warrant further investigation).

This isn't hypothetical, by the way. I have graded papers, programming assignments and exams during my time as a Student Assistant at Leiden University.

Another real world example. With some papers I was grading, I noticed that one had much better grammar than that student (a Chinese exchange student) usually used. I put some bits of the paper into google, and found it was copied from another paper on the same subject.

Are these acceptable scenarios, to you? In all these case the students have learned nothing about the subject but still hand in correct answers to the assignments, despite not being multiple choice. There aren't always such telltale signs of foul play, either.

If you say "cheating is allowed", then there is no way to avoid this. How do you tell the difference between someone doing the work and someone copying the work if you're not there to watch them do it?

The "show your work" thing is nice when you want to avoid students just solving the equation using their graphing calculator. It does nothing to prevent them from writing someone's else's answer, which shows the work, it's just not their work.

2 hours ago, Maddus Mattus wrote

*snip*

That would only be true if the question was a multiple choice one.

No it doesn't. If the question is "Write a Haiku about C9" then a valid answer is

Maddus on C9
Is often controversial
But fun none-the-less

Now suppose that haiku gets 100%. Now one of the students writes the answer on their blog. Another student looks at their exam paper: "Write a haiku about C9" and they type it into Google:

Ah! I've found one! And copy the above haiku onto their exam paper.

The teacher receives 30 copies of the same haiku as the answer to the exam question. Now since the exam is deterministic, all of the children get the same mark, and even though it's obvious that they all cheated, we're specifically allowing cheating, so we can't deduct marks for that.

So of the 30 students, one wrote a good answer to the question. 29 just googled the question, wrote it down and got full marks.

BONUS: http://channel9.msdn.com/Forums/Coffeehouse/165505-c9-free-form-haiku-

All the written tests I've taken in my life, didnt actually test anything. Including the Microsoft ones.

@evildictaitor: If two students come up with the same Haiku, they can share credit,..

Let's see how that is an incentive to come up with creative work.

49 minutes ago, Maddus Mattus wrote

@Sven Groot: I've never found complete examples to problems I faced at my education, only parts of a solution.

Really? You must not have looked very hard. Even back when I went to high school (90s) it was possible to find complete book reports for just about any book online.

And some of the programming courses at my University used the same assignments every year. Finding last year's answers was usually trivial.

Then there's the fact that nearly all programming assignments allowed you to work in pairs. This usually meant that one person did the work for one class, and the other for another class. This means that there's one subject that each of the two never actually did assignments for, and thus wasn't tested in.

If two students come up with the same Haiku, they can share credit,..

And what if they both claim to have come up with it independently? You have to give them both full credit, otherwise you're punishing them for cheating, which you want to explicitly allow so you can't punish them for it.

18 hours ago, Sven Groot wrote

How can I, as the grader, tell whether this is a program they wrote or copied from someone else. The only means of detection available to me are the blatant case where multiple people hand in the exact same program, or if I recognize a particular solution from somewhere. Another hint can be that the work is way above the student's usual level (which is not an immediate indication of cheating, but does warrant further investigation).

This isn't hypothetical, by the way. I have graded papers, programming assignments and exams during my time as a Student Assistant at Leiden University.

That's why we always had demonestration session before grading on FYP. The students in the group tell the lecturer which part did they contributed in, and the lecturer ask key questions that "people who done that part should know". If the student don't know the answer, either he/she explain what "other method" they used to achieve that result, or cheating is assumed.

4 minutes ago, Sven Groot wrote

Really? You must not have looked very hard. Even back when I went to high school (90s) it was possible to find complete book reports for just about any book online.

I've never done a book report, we we're made to tell the story in the book face to face or infront of class.

But I can imagine a teacher knows these type of sites, one or two google searches would reveal if the report is copied or not.

And still, what if the student copied it? What's the harm to the teacher? None,. The student only hurts himself with it.

And some of the programming courses at my University used the same assignments every year. Finding last year's answers was usually trivial.

Then the test makers are lazy, maybe that counts as a form of teacher cheating.

Then there's the fact that nearly all programming assignments allowed you to work in pairs. This usually meant that one person did the work for one class, and the other for another class. This means that there's one subject that each of the two never actually did assignments for, and thus wasn't tested in.

If they all hand in the same program, they can share credit.

I've done the assignments you talk about and I did not experience quite what you describe.

We've helped each other understand the assignment and the solution. Sure, there were students who piggy backed on some of the others, but in other courses it was exactly reversed. We all grew from each others experience. I've done projects where I did little, and I've done projects where I did nearly all. But learned from them all.

And what if they both claim to have come up with it independently? You have to give them both full credit, otherwise you're punishing them for cheating, which you want to explicitly allow so you can't punish them for it.

I'm not punishing at all, I'm giving credit to each individual piece of work. If two students come up with the same solution, they can share the credit.

Back to my original point I was trying to make;

Cheating mostly hurts the cheator

27 minutes ago, Maddus Mattus wrote

Cheating mostly hurts the cheator

Yes, and of course everyone of school age is wise enough to know that already (or believe it when told). Who cares if people who could otherwise have had promising careers screw themselves over because no one cared enough to prevent them?

Who cares if half the people get out of school knowing absolutely nothing because they could just cheat the whole way.

Mentality of students is a different problem.

It a young age when, in Holland at least, you are forced to go to school, the teacher is the customer and the student the producerer. The intent to cheat is there, because the student may not want to be there and cheat his way out.

When at a later age you will see the mentality change, the students become the customer and the teacher the producer. Now the student actually wants to learn, if he wants to cheat, then it mostly hurts himself.

When you leave school with your degree and you know nothing, the businesses will start to value the degree less. So the students and the school has the incentive to make sure when they hand out their degree's, the student actually know what he is talking about.

So, design tests that have a smalelr chance of being cheat at and help your fellow students become the best students they possibly can.

I like how the Dutch DMV solved it. Multiple choice, all at the same time, limited time per question. No chance in hell you can cheat on those.

35 minutes ago, Maddus Mattus wrote

@Sven Groot:

So, design tests that have a smalelr chance of being cheat at and help your fellow students become the best students they possibly can.

Sure, I agree with that, but that's easier said than done. Especially when you consider that elementary and high schools are often vastly understaffed (and the teachers are underpaid), and that at higher levels of education the teachers are often researchers who only teach because they have to, not because they want to.

I like how the Dutch DMV solved it. Multiple choice, all at the same time, limited time per question. No chance in hell you can cheat on those.

That's not a good example, because this only tests your ability to memorize (it's not as bad as some other multiple choice tests, because at least you're given real scenarios and asked to interpret them based on traffic regulations, rather than just regurgitate the rules). It's possibly the method of testing that least allows the students to use their creativity.

Plus, weren't you arguing that we should allow cheating because that's how the real world works? Why have you suddenly shifted to designing tests that prevent teaching?

(This is also an example of why "you can look stuff up in the real world" doesn't always apply, because you can't exactly Google road signs while you're driving)

The DMV is a good example because there is no interpretation to traffic regulations. If you come from the right, you have right of way. You need to memorize that. And that way of testing is excellent.

The DMV test was an example of a test you cant cheat at. So you wouldnt need to prevent it. Sure you can look over your shoulder at the other guy, but because of the time constraint you would run behind hoplessly (seen it happen).

Let me adjust my point then to something we can agree on, I often formulate my opinion a bit to harsh, I'm working on that;

Instead of trying to prevent cheating, put that energy into designing tests so that is in the students best interest to learn and master the skill being tested.

A written exam is easy to make and grade, it's also easy to cheat at.

3 hours ago, Maddus Mattus wrote

I've never done a book report, we we're made to tell the story in the book face to face or infront of class.

Examining someone face-to-face is vastly more expensive than giving them a written paper to do.

But I can imagine a teacher knows these type of sites, one or two google searches would reveal if the report is copied or not.

But you're explicitly allowing cheating because "it's more like the real world".

If you have to implement a Binary Sort, does your boss punish you for copying a working sample from StackOverflow? Then why should an examiner (under your system)?

And still, what if the student copied it? What's the harm to the teacher? None,. The student only hurts himself with it.

Having a 1st class degree in Computer Science will benefit the cheater more than the lack of knowledge of how to build a LALR table will damage his career.