2
\$\begingroup\$

It's pretty clear that writing a good question for PPCG is pretty hard - we close a lot of questions, and we have a low rate of questions asked/user. What are some methods we could employ to attract more good questions?

Ideas:

  • A standardized question-making form, for inexperienced question askers. Could be customized by question type, e.g. , , , .

  • More points/badges associated with question asking (Note: I don't know if this is possible.)

  • Minimum time before questions can be put on hold of a few hours, except for questions that should be on another site. In general, there's no advantage in closing a question extra quickly. We should allow time for both the OP to respond to comments, as well as for other users to make changes.

What do you think?

\$\endgroup\$
  • 6
    \$\begingroup\$ The first idea already exists. The second one would only be possible if the SE dev team wanted to implement new badges network-wide (which did happen recently). The third doesn't make much sense - the whole point of putting something on hold is to let the OP to respond to comments and let other users make changes. \$\endgroup\$ – Doorknob Aug 19 '14 at 0:25
  • 2
    \$\begingroup\$ Our low rate of questions comes from the difficulty of coming up with the idea \$\endgroup\$ – Nathan Merrill Aug 19 '14 at 0:28
  • \$\begingroup\$ If you'd like, you can join us to discuss this further in chat. \$\endgroup\$ – Doorknob Aug 19 '14 at 0:43
7
\$\begingroup\$

As an experienced Stack Exchange user, I have to strongly disagree with your suggestion to have a "minimum time before questions can be put on hold". The whole purpose of "on hold" is to let the OP fix the question for 2 days (after which the question becomes "closed" if people don't vote to reopen it).

In theory, if a question has successfully gone through the sandbox (i.e., many upvotes on the challenge, and all substantial comments resolved), people shouldn't be voting to close it, so the fact that a question is "on hold" means that this sandbox process wasn't followed through.

The thing about "on hold" is to stop people from posting answers that will soon be rendered invalid, once the question is fixed.

\$\endgroup\$
6
\$\begingroup\$

TL;DR: This is a non-issue, and our question rate is fine.

As it stands now, we have:

  • around 2k questions (not counting closed or deleted)
  • around 5 questions per day
  • around 10 answer ratio (answers:questions)
  • around 6k visitors per day
  • probably no more than 200 active users (high-end estimate from observation)

Think about that, there's five new questions per day with only a couple hundred active users. Considering that each new challenge is supposed to be based on a unique problem, I think that's pretty damn good. Some people can churn out a problem every couple days, but many are here more to answer. That's okay, we need more answers than questions.

The thing is, there are tons of questions if people want to answer. Questions only get an average of 10 answers. Obviously some get more and some less, but they're still there, waiting for more answers. It isn't the lack of good questions that is failing to attract users.

A better question in my mind is: What can we do to get more attention to older challenges? We have tons of older questions that people can jump on, but it doesn't happen that often. Sure, a question or two might get bumped up every couple days by a new answer, but there are tons that nobody seems to look at.

I'm sure the "shiny" factor is the main cause, but it's more of an issue with this type of site then a regular Q/A site. On a Q/A site, questions get answered, full stop. Once a correct, decently written answer comes along, there's little incentive to add to it (unless something changes later).

It doesn't really work that way here (with a few exceptions), because people still vote on new answers to old questions. See an old with a great answer? You can still golf down a different language (bonus for a different category of language) and get some votes. A that hasn't been touched in a while? Give it a shot! You might come up with something better. Many times people will appreciate just seeing a different method of solving a problem, even if it doesn't strictly "win".


As far as "we close a lot of questions" is concerned: Good! Frankly, we get a lot of questions worth closing. What's even better? We also reopen a lot of questions. When I see a question get closed here, it normally boils down one of:

  • Duplicate
  • Clearly off-topic
  • Broad/Imperfect spec

There's simply not much to do about the first two categories, every site has them and they just need closing.

Questions with a spec problem are where this community really shines IMO. People generally want to see questions reopened so that they have a chance to answer. We often see long comment threads about what problems the spec has and how to fix it. Sometimes the OPs respond/fix, and it gets reopened. Sometimes somebody else fixes it and it gets reopened. Sometimes it can't be helped and just dies. This works as advertised, and better than on many SE sites from what I've seen.

\$\endgroup\$
  • 4
    \$\begingroup\$ I hesitate to answer older questions out of a worry that my answer will ignored. Also, old questions tend to be underspecified by modern standards. Could we make a weekly event where we revive an old question that didn't get the attention it deserved? \$\endgroup\$ – xnor Aug 20 '14 at 0:07
  • 1
    \$\begingroup\$ @xnor "Revive an old question that didn't get the attention it deserved" sounds like a job for bounties. Easier to implement than an official weekly event, and can be done by anyone who sees an interesting old question. \$\endgroup\$ – Geobits Aug 20 '14 at 12:45
2
\$\begingroup\$

Responses to your ideas:

  • A standardized question-making form, for inexperienced question askers.

    That already exists.

  • More points/badges associated with question asking

    This would only be possible if the SE dev team wanted to implement new badges network-wide (which did happen recently). I really don't see why it's necessary though.

  • Minimum time before questions can be put on hold of a few hours

    Wait, what? Why?!

    We should allow time for both the OP to respond to comments, as well as for other users to make changes.

    This doesn't make much sense - that is the whole point of putting things on hold: to allow time for both the OP to respond to comments, as well as for other users to make changes.


It's pretty clear that writing a good question for PPCG is pretty hard ... What are some methods we could employ to attract more good questions?

Do you have any evidence that this is a problem? Naturally, coming up with questions for PPCG is going to be much harder than coming up with a question for, say, Stack Overflow.

The question you should be asking is: How can we attract higher quality questions? More questions is not necessarily better for the site; a general increase in challenge quality would be much more beneficial in the long term.

So how do we do this, you ask? There's no magical silver bullet; there's not one specific thing you can do that will improve overall challenge quality. The best thing you could do for this site is to continue to contribute to the community: Post good challenges, post solutions to challenges, help other users via the voting and commenting systems, participate on meta, encourage others to do the same, and so on. By being an active member of the community, you and every other user will each help this site succeed.

\$\endgroup\$
1
\$\begingroup\$

Standardized question-making form

I think this would be really useful, if we could make it visible enough. New users who want to, for instance, ask a question, rarely know that the should include input and output specifications, example cases, and scoring guidelines, but these are often expected in code golf questions. Similar cases exist in other kinds of questions.

As for how it would be structured, one way would be a code block with markdown ready to be pasted into the answer box with <input specification goes here> and whatnot.

\$\endgroup\$
  • 1
    \$\begingroup\$ meta.codegolf.stackexchange.com/q/1511/3808 \$\endgroup\$ – Doorknob Aug 19 '14 at 0:24
  • \$\begingroup\$ @Doorknob I doubt a new asker would realistically see that template. \$\endgroup\$ – xnor Aug 19 '14 at 0:52
  • \$\begingroup\$ @Doorknob, perhaps that could be featured instead of the retired sandbox XIII? \$\endgroup\$ – Peter Taylor Aug 19 '14 at 6:56
0
\$\begingroup\$

Yes, it is important that we get more good questions. Both because that will increase the choice for all of us who enjoy answering, but also because that will attract new users to the site at a greater rate.

I think what's important to realise is that putting people off the site by closing their questions is not as bad as it seems. Generally the people being put off in this way have minimal rep, because they have not answered (m)any questions. The kind of people who ask good code golf questions are the kind of people who write good code golf answers. The same goes for code challenges and king of the hill contests.

  • the inate skills required for asking good questions also make someone good at writing answers.
  • the experience and insight gained from working hard on answers helps avoid the many pitfalls associated with writing a question.

Someone who has a history of writing good answers but writes a question with a flaw is less likely to be put off by it being closed, but will understand that this is how the community does things, and will simply learn from it and go on to write even better questions.

For this reason I think that although we do need more good questions, what we need to focus on is attracting more people who will write good answers. If we can do that, questions will follow naturally, and they will be much better questions than if we try to do things the other way around.

If you really want to see more good questions, raise a meta question about how to get more good answers.

\$\endgroup\$
  • 3
    \$\begingroup\$ "the inate skills required for asking good questions also make someone good at writing answers" I beg to differ ;). I've written many well received questions but seldom write answers. I believe that the lure of writing questions is more powerful than the lure of writing answers for new users. A newbie is unlikely to beat the top golfscript answers for a code-golf so why even bother? But a question has limitless possibilities and can more easily become famous. So I don't necessarily agree that better answers means more good questions. (I do agree that closing questions scares new users off.) \$\endgroup\$ – Calvin's Hobbies Aug 19 '14 at 3:35
  • \$\begingroup\$ @Calvin'sHobbies I definitely agree that asking questions is more appealing for new users than answering questions. I still stand by my point though. Without experience of the site, most new users with no experience of answers ask questions we have no choice but to close. I'd rather put a rep limit on asking questions so that the people asking questions are those who stuck around long enough to gain rep from answers/edits and know what is required of a question. \$\endgroup\$ – trichoplax Aug 19 '14 at 13:47
  • 2
    \$\begingroup\$ @Calvin'sHobbies yes you are an obvious exception to all of this, as yet unexplained by science. Some say it was a terrible lab accident. Some say you are secretly a team of 12. Cyborgs. From the future. Whatever the case, I don't think you are a typical example that we can use to base decisions on... \$\endgroup\$ – trichoplax Aug 19 '14 at 13:50
  • \$\begingroup\$ For the problem of new users being wary of answering tough golf questions, I personally found that softer questions were the route into answering: Pure popularity contest -> Code length restricted popularity contest -> code golf. I do think encouraging a wide range of different questions helps new answerers take the first step. \$\endgroup\$ – trichoplax Aug 19 '14 at 13:55
  • 1
    \$\begingroup\$ I definitely agree that knowing how the site works helps people write good questions. Its also true that softer questions can help boost new users. But I still think not being able to ask a question from the start will discourage some users. The community can help revise poorly written but potentially good questions. (And yes, perhaps I am a bit of an anomaly :P ) \$\endgroup\$ – Calvin's Hobbies Aug 19 '14 at 14:33
  • \$\begingroup\$ @Calvin'sHobbies yes I'm in two minds about a rep limit on questions - it would definitely only be about 5 or 10, but I agree even that might lose us potentially great question writers. \$\endgroup\$ – trichoplax Aug 19 '14 at 14:45

You must log in to answer this question.

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