** Using WeBWorK **

I would find it very beneficial if users would save the problems they created and assembled as problem sets as pdf files that could be downloaded. That way people could see what the problem looks like without having to save and build problem. It would make it easy to scroll through a pdf file to see if particular problems would be useful. The actual code file could just accompany the pdf file(s) I have a frustrating time sifting through the code or building them just to see if I would use them.

If there is an easier way. Please let me know!

Thanks,

Gregg

<| Post or View Comments |>

This isn't what you want (yet) but it might help. Go to software download --> Problem Library You will find that you can click on the sets to open them, and then you can view both the source and the rendered version of the problem. (This is good for learning how to write the problems, it's not quite so good for scanning through a bunch of problems quickly.) You can also select problems from the set with the check boxes and have that group of problems TeXed and emailed to you as a TeX file. If you have TeX locally you can then TeX and view this conveniently. It's not perfect, but it may be better than what you are doing.

I'll eventually add an engine that will create a pdf file of the whole directory and present it. This is what you want I think. It's not hard to do, but I'll be away almost all of July, so I may not get it done until sometime in August. The problem library mirrors my copy of our CVS problem library ("rochester"). It is up to date when I'm in town. It may fall behind a bit when I'm not here to update it from CVS. I'll also add the other problem libraries sometime in August.

-- Mike

<| Post or View Comments |>

Thanks! That is exactly what I want. Would the engine you create also create pdf files for problems that other institutions have written and submitted to the repository or just Rochester's problem library?

Thanks for the quick response.

<| Post or View Comments |>

Thanks for the suggestion. See the end of this message for more general comments on your idea.

You can pretty easily get pdf output of the entire problem library (or a large subset of it). We do this at Rochester and put a printout (which is very thick) in a public area. People borrow it and xerox portions they are interested in. We also find it useful to make available pdf versions of the problems assigned in actual courses.

Here's how to do this fairly easily. Dump the whole problem library into the templates directory of some private course, e.g. "mycourse" (or make the templates Directory of mycourse a link pointing to the library). In the webworkCourse.ph file for mycourse, make sure your login is listed in $PG_environment{'PRINT_FILE_NAMES_FOR'} = ['apizer','gage']; since you will want files names listed in the output.

Now on the build problem set page, select all the sets (or the subset you are interested in) and build them --- this requires one click. Now on the page where you get pdf output, select 20 sets at a time and get the pdf output (professors can get output from multiple sets --- students can only output one set at a time). The 20 comes from $max_num_of_ps_downloads_allowed = 20; in Global.pm. If you have a fast server, you can up this limit. The limit is there so that your browser doesn't time out if you have a slow server. This is a little tedious, but you will probably only do it once or twice a year.

Your suggestion that professors do this on a course by course basis (say one pdf file covering the entire course) and make it publically avialable is a good one. This is part of a bigger question of how best to share problems and recommended lists of problems. As a start look at http://webwork-db.math.rochester.edu/cgi-bin/cvsweb.cgi/ Here you will find individual libraries used at Rochester, ASU, DCDS, Indiana, and Stony Brook. We are hoping that everyone writing problems and wishing to make them publically available will establish a repository at this site --- the ones above have been set up on an expiremental basis. There may be a lot of over lap in these sets but at least it's a start on a national library. It might be a good idea to add collections of set definition files and pdf output from actual courses so that if you respect how Prof A from University B teaches subject C, you can look at exactly what they did.

These are the type of questions we want to discuss at the WeBWorK session at MathFest on Friday, August 2 (http://www.maa.org/meetings/mf2_maas.html). Please attend if you can.

Arnie

<| Post or View Comments |>