|CLOSED sets confuse students||topic started 4/3/2002; 11:23:28 AM
last post 4/7/2002; 2:01:48 PM
|Michael Gage - Re: CLOSED sets confuse students
4/7/2002; 2:01:48 PM (reads: 1014, responses: 0)
The "closed" set designation has caused trouble here as well. I'll think about making this (and the other descriptions) locally configurable. Unfortunately any additional choice is one more thing that might have to be delt with at configuration time.
The best outcome would be to find a designation that most everyone will approve of so that it doesn't need to be changed.
Here's what we have:
closed -- confusing, students won't review their homework past due -- overdue -- reasonable, although it has connotations that you can still do it if you hurry
graded -- indicates that the final grade for the problem set has been recorded -- this is more or less the situation
Are there any other suggestions?
As to the ability to access homework problem set by problem set. That is very much within the realm of the current technology -- but the best way to do it is still being investigated. You (and any one else) can help with suggestions and ideas.
Here are a couple of approaches currently working. The grading and recording feature is not part of either of the current implementations. The overall idea is to separate out the problem rendering mechanism (which is pretty thoroughly perl based) from the grade recording, web page pretty printing, and instructor database access for organizing problem sets which can be handled by most Course Management Systems and web Content Management Systems.
Individual problems for the library, the latest update served from the CVS (Concurrent version server) library directory : See http://webhost.math.rochester.edu/webworkdocs/ww/listLib?command=setsOnly
Individual problems imbedded in a page.
http://webhost.math.rochester.edu/gage/slides/13/12 -- there are still some bugs in getting these problems to work
Most of the problems in the tutorial section are also being served live from the CVS library, although there are few bugs with connection in that case as well.
These all show "proof of concept" that the method of grouping and delivering the problems can be changed without changing the rendering mechanism. There is still more work to do obviously. Bill Wheeler has also been working on this sort of problem at Indiana University. He may have more to say about it.
Summer is a good time to work on projects like this.