COMMENTS AFTER ASSESSING A PREVIOUS YEAR'S ASSIGNMENT 2

As this is the end of this year's 773 course, there isn't a lot of point in listing all my possible comments, so I've only written down things which might be of broader applicability, or which reflect on your marks or grades in some way.

MAIN POINT.

Remember what the course is about, and write down relevant material in your report. I think that this implies a good proportion of the comments which follow. What I want is a report which demonstrates that you understand the course topic and can apply it to specific examples. I am happy to know that you can design and construct complicated electronic or mechanical or software artefacts, but if it isn't fairly directly associated with the topic I can't give you marks for it. ( That's why I restrict myself to 10 marks for presentation, though I think it's very important. )

Read the instructions.

I suppose that an alternative would be "Stick to the instructions", but I'm not sure that I'm ready for the idea that graduates will read instructions and then just ignore them. It's quite stupid enough not to read the instructions in the first place.

When I say "number the pages", I mean "number the pages", NOT "don't number the pages". I don't believe that so many of you are using word processing software which can't manage automatic page numbering. Section numbers are NOT a satisfactory substitute; they turn up in different places on different pages, so you can't find them quickly.

When I state explicitly that I expect you to look for other sources ( more a matter for the presentation than the assignment, but it applies here too ), I don't expect to find no other sources unless you explain why.

When I say that people working in groups must submit individual reports, I don't mean that a joint report will do.

And did anybody so much as look at "SORT OF CHECKLIST" ?

And so on. You are not the first students to behave in such a ridiculous way, and I don't suppose you'll be the last. I suppose it is mildly instructive to find that distributing the instructions by WWW has exactly the same effect as distributing them on paper - and it's cheaper for the department.

Think visibly.

It's no use your performing prodigies of intellectual activity if I can't see any evidence. Just saying "I decided that ..." isn't much use to me; if you don't show me how you decided, then for all I know you decided by tossing coins, for which you get no marks.

This is particularly important if I think that your conclusion is silly. If you can show me that you got there by a sensible argument, or even by an argument in which only one step was questionable and you didn't know it, I can give you a lot of credit; but if you just give the conclusion, I go back to the coin-tossing hypothesis.

Think about everything. Think about what you're going to do, and tell me why you made your decisions. Think about what happened, and tell me why you reached your conclusions - or why you couldn't get any further.

Think about the course material, and use it where it applies. Don't go out of you way to drag in course material, but point it out when it's there. It is by no means uncommon for people to reinvent wheels which have been described in lectures or course notes ( or textbook ), without giving any hint that they've heard of the wheels before. This is very unimpressive, especially when they reinvent the wheels wrong.

I would much rather have evidence of lots of thinking ( provided that it's sensible thinking ) than pages of experimental results. If you need the results to think about, all well and good, but if you don't show me that you've been thinking about the results, they're not much use.

References.

When I lay out for you a full description of a reasonable convention for giving references, I expect you to use a reasonable convention for giving references. It needn't be the same one, but should have all the conventional parts, so that someone else can find the reference easily if so desired.

The point of identifying reference works is to make it possible for people to refer to them. If you just give a bibliography list, how does anyone else know where to find the items you've used ? Read right through all of them until it turns up ?

When you use some information from one of your references, identify the source precisely in the text. If it's an article, give a full reference in an acceptably standard form ( see the "CHECKLIST" ); if it's a book, identify the book in a standard form, and give the page number if it's likely to help.

Alan Creak,
June, 1997.

Go to the 773 course page;
Go to me;
Go to Computer Science.