1
$\begingroup$

A colleague of mine has a project where he is gathering "good problems" from many calculus topics. He wants to share them all online and continue to accept new submissions, but we know it is not a good idea to try to administer a webserver.

Would it be reasonable to host the entire operation on ME.SE?

If so, every day for the next few months, we could post a new question that is something like "What are the best problems [in topic X]?" Then we could community-wiki-post the problems that have already been gathered on the question, and let the question run its course getting new input and votes from other users. We would then publicize the questions in various external places, and probably make a tag for the project?

Anyway, I see a lot of potential benefits:

  • Publicity for the site, which has pretty low activity level
  • New users for the site

and a lot of potential downsides:

  • The low activity means we could easily overwhelm the site and completely take it over
  • The remaining regular users of the site could be annoyed / chased off by the project
  • There won't be "accepted answers" to these questions (Does this mean I should mark them Community Wiki? Does that even fix the problem?)

There are certainly other issues.

Do you think this would be a good use of the site? Do you have any suggestions of specific details that would make it work better?

Help. Thanks so much.

$\endgroup$
  • 3
    $\begingroup$ Interesting. Not sure yet. One thing, please do not say "best" in the question, but "good" would be good. :-) (I don't insist on 'good' but 'best' just asks for trouble.) This detail aside, could you give an example how specific/narrow you envision the questions? I think the question should be relatively fine grained. $\endgroup$ – quid Apr 18 '18 at 19:25
  • $\begingroup$ Also I want to emphasize that I think the details of implementation are critical. matheducators.stackexchange.com/help/dont-ask warns "avoid asking subjective questions where every answer is equally valid." We can fix this a little bit by demanding that answers include a brief discussion of why the problem is good. But I hope others will point out possible issues and how to solve them with implementation details, or reasons to not do this at all. $\endgroup$ – Chris Cunningham Apr 18 '18 at 21:04
  • $\begingroup$ @quid I envisioned very specific, for example "What is a good implicit differentiation question to ask students?" with the text explaining that you have to explain why you like your question. This would make for a very precise topic but also allow for different kinds of answers: some exercises, some conceptual questions. $\endgroup$ – Chris Cunningham Apr 18 '18 at 21:07
  • 1
    $\begingroup$ I think with the instruction in your first comment this could work well. Asking for something quite specfic and insisting the answers explain what is good/instructive about the particular example or batch of examples could work well. $\endgroup$ – quid Apr 18 '18 at 22:47
  • $\begingroup$ No, that's already happened on math.se, and this site can't come anywhere close to the repository of calc problems that math.se already has. I literally feel bad for educators who want to reinvent the wheel, as though it hasn't already been invented. and want to waste time doing it. $\endgroup$ – Namaste Nov 22 '19 at 22:07
1
$\begingroup$

...we know it is not a good idea to try to administer a webserver.

FWIW there are other places you could host it without having to administer a webserver. In particular, I think it could be implemented as a project on GitHub, which has its own pros and cons.

If so, every day for the next few months, we could post a new question that is something like "What are the best problems [in topic X]?"

No matter how specific topic X is, I think this risks being too broad as a question. A good problem for homework isn't necessarily a good problem for guided investigation in the classroom and vice versa. Suitability might depend on the order in which topics are taught in different syllabi, the ability level of the students, and their learning styles. I would suggest that you narrow down at least some of these parameters.

Anyway, I see a lot of potential benefits:

  • Publicity for the site, which has pretty low activity level
  • New users for the site

This might be easier for other people to assess if you were a bit more specific about where you plan to advertise the project.

and a lot of potential downsides:

  • The low activity means we could easily overwhelm the site and completely take it over
  • The remaining regular users of the site could be annoyed / chased off by the project

These are really one issue, IMO, and easy to avoid unless other people try to jump on the bandwagon (and bandwagon-jumping isn't something I've seen much on this stack). It just requires the discipline to go slow and not post new questions when more than N of your questions are active. If you decide to go ahead I would suggest that you start by posting just one question, to see what the reaction is (not everyone reads meta, and in fact a comment linking back to this discussion would also be a good idea); if positive, post a couple more after three days; and then wait to post new questions until at most two of the project's questions are in the top half of the front page. (When I engage in retagging projects on other stacks I limit myself to five bumped questions, but I expect that even people who find it irritating will tolerate retagging following discussion on meta as a necessary evil).

  • There won't be "accepted answers" to these questions (Does this mean I should mark them Community Wiki? Does that even fix the problem?)

Not having accepted answers is not a big problem: marking the questions community wiki would be a big problem. An accepted answer is only 1.5 upvotes, but if the question is CW then so are all answers, which means that people don't get any rep however many upvotes they get.

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .