WeBWorK Main Forum

Removing OPL from server - any danger?

Re: Removing OPL from server - any danger?

by Nathan Warkentin -
Number of replies: 2
Danny and Andras, thank you both for your answers.

In response to your comments, Danny, about contributing to the OPL, it has certainly been on my mind. My work is busy, and I am still in the midst of implementing courses, so OPL contributions will not happen in the short term. However, the improvement upon my school's program since I started implementing WeBWorK in our courses has been substantial. I intend to give back to the project in whatever ways that I can.

Indeed, my main frustration with implementation has been the overall low quality of exercises in the OPL - I often come to regret using them in my problem sets. Contributing some of my work to the library seems a reasonable response.

One question as a follow-up, regarding contributions to the OPL. I have noticed that a number of problem authors in the OPL use macros hosted on their institution's servers to write problems. While this may work for very large, stable institutions. I work in a comparatively small international school, and it is not reasonable to assume that any macros hosted on one of our servers will persist for an extended period of time. I have a fairly large number of subroutines that I've developed to ease problem authoring and checking answers, and I store them in custom macros locally. What would be a best practice for publishing to the OPL in a circumstance like that?
In reply to Nathan Warkentin

Re: Removing OPL from server - any danger?

by Danny Glin -

Regarding macro files related to OPL problems, there is some work being done currently to clean this up.

One of the principles of the OPL is that submitted problems should be institution-agnostic (i.e. they should not have references to institution-specific resources), though this has not always been enforced.  Although macro files are sometimes named after an institution, the subroutines within them should be applicable outside of that institution.

Going forward the recommended practice will be something like the following:

  • Macros should not explicitly refer to anything relating to your institution.
  • All macros that are relevant to the community should be submitted to the PG repository and not the OPL.  If they are related to macros in existing files then they should be submitted as modifications to those files.  If they are a new class of macros then a new file can be created.

If you get to a point where you're ready to submit something you can open a pull request on Github, and the release team will provide feedback if things need to be moved around.

In reply to Danny Glin

Re: Removing OPL from server - any danger?

by Nathan Warkentin -

Thank you for the clarification, Danny. This is the kind of policy that I would expect - it is not sensible to have publicly available problems with components tied to a specific organization.

With respect to my own contributions, I am sure that some of my own subroutines have relatively sloppy implementations. A good example of this would be the logic that I use in checking that students have factored a polynomial correctly within a single answer blank. Although my code does its job well enough, I would be surprised if it was ready for inclusion in the repository. In this case, is it better to submit it in its extremely unpolished format or delay and find time to work on it later?