13

Retreading old ground

This has been batted around several times before on Meta, but I think it's worthy of a periodic refresher. With that in mind, it would be counterproductive to duplicate-close this question.

What, concretely, and with reasonably immediate results, can the long-time or regular users (and mods, in their dual capacity as users) do to attract new interesting questions from new or less-regular users?

In search of novelty

I want to emphasize interesting. We get plenty of new questions every day; on that score EL&U is among the most productive non-trilogy sites in the network.

But for the most part, the kinds of questions we see are workaday questions from (mostly) non-fluent users of English who are trying to solve small, practical problems. And, for the most part, looking at it from a more abstract lens (which regulars, perforce, do), most of them are repeats, falling into a handful of higher-level patterns.

And we regulars, having seen so many of these, tend to find them terribly dull, and not only not engaging, but to a good extent off-putting¹.

Navigating the woods

There are enough questions about reducing or redirecting this stream, though, with no clear solution in sight, so instead I'd like to focus this question on improving the ratio of novel-interesting-questions : banal-questions by raising the numerator, rather than lowering the denominator.

Again, I'd like to focus on concrete, relatively immediate suggestions. Given that this has not organically occured, it suggests there's a investment of our own time and effort in order to reap the benefits of a more engaging and less disheartening experience on the site.

Drawing a treasure map

So, what can we do? Ideally, I'd like to see answers in the form of suggesting and elaborating a single idea (one idea per answer), along with the commitments we (the regular users) would have to make to realize this benefit.

Something like:

  • Ask new and interesting questions ourselves (like Mari-Lou and Josh [sorry I forget your new numeric ID] make an effort to do)
    • Is this scalable? Seems to me we need consistent infusion of fresh voices with novel, interesting questions to sustain the flow.
  • Reach out to groups with similar interests (e.g. the communities around LanguageLog, etc) and market EL&U a bit (organically, so people are attracted over the long term)
  • Relax rule X or constraint Y (e.g. permit and encourage questions of new usages, slang, or jargon which haven't been firmly established in reference works yet, in full knowledge this will make it more difficult to substantiate answers).

And so on. I'm open to anything.

But the objective is to collect concrete ideas for which there will be a relatively immediate return in the proportion of novel, interesting questions asked by newcomers, given a specific, documented investment by the regular-user community.

Self-empowerment

This section was added after the fact, inspired by Sven's first suggestion and Lawrence's recent suggestion

We've had a couple of suggestions for improvements to the site which could -- almost certainly would -- improve the number of novel, interesting questions.

These are good ideas, and I endorse them, and I've upvoted them. But I would like to ask future answerers to keep the focus on "self-actualization", i.e. changes within the power of the community (and the mods, sans diamond hats) to effect ourselves.

I say this because SE has a "standard 6-8 week waiting period" to implement new features, which in practice results in an effective delay of between 6-8 months and eternity.

In addition, they are currently focused on other projects (which are covered on the big Meta), so I'm even less hopeful such feature requests will get priority over the next couple of quarters.

So, I think it would be most helpful for answers to focus on change we can make ourselves, and in particular change that requires some sacrifice on our behalf, because it's my firm belief that you get what you pay for.


¹ Meaning this question, of course, is asked in a terribly self-interested spirit. I love EL&U and I've loved participating here. But my participation falls off week by week, inexorably. I don't want that to happen; I've extracted so much joy and learning from this site.

But the gamification system (rep and badges and votes) has failed to motivate me since, I don't know, 2K? And where once I saw every new question with fresh eyes, found some element to pique my interest (remember when you'd see a comment from @DanBron at the top of almost every question chain?), that has long since faded too. I'm as jaded as the next 25K user. I've seen it all, and all I can taste is dust.

13

As most Meta readers know, an EL&U site participant who earns a gold badge in a particular tag category gains the ability to unilaterally close duplicate questions (the "golden hammer"). But this is essentially a negative capability. I would like to see English Language & Usage introduce a complementary positive capability: a "golden shield" to protect good questions from closure.

Here's what I propose: High-rep users—people who have a gold badge in a given tag category or who have a threshold number of reputation points overall plus a silver badge in a particular tag category—should have the ability to protect from closure any question in one of their gold or silver tag areas that they consider good and interesting, regardless of whether the question "shows research" and regardless of whether it may look like a duplicate or seem primarily opinion-based to reviewers who have far less reputation or demonstrated interest in the tag area involved.

Currently, 78 EL&U site users have 20,000 or more reputation points, 45 have 30,000 or more, 31 have 40,000 or more, and 21 have 50,000 or more. On the tags side, site participants have received 32 gold badges (19 of them for single-word requests) and 218 silver badges (78 of them for single-word requests). At some point, it seems to me, we ought to be able to trust the judgment of such users when they say that a question is interesting and answer-worthy.

Consider the "pronunciation" tag: Today, four site participants—tchrist, Peter Shor, sumelic, and Araucaria—have a silver badge for this tag (no one has a gold badge yet); all four also have at least 30,000 reputation points on this site. I would completely trust the judgment of any of those four people if they said that a question tagged as "pronunciation" was interesting and worth answering. Enabling participants to protect good questions that fall within their main areas of interest might also give people who have not yet earned certain silver badges more motivation to ask and answer questions in those areas, in order to gain the power to protect relevant questions.

  • 1
    Wonderful answer. Thank you, sincerely. Can I prevail upon you to split the two seperate suggestions into two seperate answers, so the community can weigh in on them independently? I think a critical part of this effort will be community consensus (and I have plans to make further posts to elaborate on explicit plans for the answers demonstrating the highest consensus among the people we will ask to participate and invest). Separately, I love the rep- and badge-dependent nature of the first suggestion, because it may help motivate me, individually, to care about such things again. – Dan Bron Oct 9 '18 at 2:30
  • 2
    Also, I agree with you, despite being one of the heaviest-handed with the "research" close reason, that it has proven, pretty unequivocally, not to have borne fruit. No one likes saying "I was wrong", but I was. – Dan Bron Oct 9 '18 at 2:31
  • 3
    The ability to unilaterally open any question seems like it might be overpowered. The "positive" version of the dupehammer is itself, since it's possible for wielders to unilaterally reopen a question closed as a duplicate. – Laurel Oct 9 '18 at 2:43
  • Reopening (not closing) phrase/idiom/expression requests could be extended to those with the SWR dupe hammer and vice-versa. All four requests are very similar in nature, and presumably, these users are experienced enough to recognise a question that has potential and is not POB or a duplicate. – Mari-Lou A Oct 9 '18 at 8:38
  • What do you propose should happen when a golden hammer encounters a golden shield? Would the hammer or shield prevail? Or would the golden hammer remove the shield, requiring input from others in the community for closure? Your suggestion sets the stage for quite a bit of action. :) – Lawrence Oct 9 '18 at 15:58
  • @Lawrence It should work as it works now, which is a golden hammer only takes one vote to change the state. All change states are reversible. If a question is hammer closed, it would take 5 normal vtes to reopen, or a single hammer reopen, and then 5 more regular (or a single (other) golden hammer) to reclose. – Mitch Oct 9 '18 at 16:21
  • 1
    @Mitch Good point. It makes the shield more powerful than the hammer: the hammer can only close-as-dupe while the shield can protect/reopen anything. On the protection front: I think Sven's suggestion includes preemptively protecting questions. If the hammer can change a duly protected question instantly, it makes the hammer more powerful than the shield wherever the hammer can be wielded. – Lawrence Oct 9 '18 at 22:01
  • Why don't you ever offer bounties on questions, especially one that deserves greater attention or better answers? – Mari-Lou A Oct 15 '18 at 6:41
  • 1
    Reputation means nothing but longevity. Everything based on it, which is everything, is so much BS. – 9fyj'j55-8ujfr5yhjky-'tt6yhkjj Oct 17 '18 at 4:16
11

We should be more careful about labeling a question a duplicate. I don't have statistics, or even examples, but my impression is that a significant number of Qs labeled as duplicates vary enough from the "original" that they may bring forth worthwhile answers, and/or the answers to the "original" are not very good.

Once a Q is labeled a duplicate, that is usually the end of it. And there is the follow the leader syndrome, where once one person says duplicate, four more say "Yep, duplicate" possibly after agonizing analysis, but, I suspect without enough thought.

What to do: Pledge to ourselves to read the original and the accused duplicate and its answers skeptically, and write a cogent "no its not a duplicate" on the proposed duplicate, unless it is. And vote to reopen, with comment, if necessary.

Again, I wish I could give good examples, but I often think that people who vote duplicate are missing important nuances.

  • 2
    The downvote is too bad. This is exactly the kind of idea I was reaching for: something that asks a sacrifice of we regulars, despite the creases of cynicism furrowing our brows, as an experiment to see if magnanimity will result in an improved experience for everyone. – Dan Bron Oct 8 '18 at 23:41
10

I realize that many site participants view bad question inflow as the central problem facing this site; but I think that unintentionally discouraging good question inflow may be at least as serious a problem—and it's one that we have largely ignored or, at best, sought to remedy by making all questions harder to post. For several years now, we have used the "show research" requirement as a key component of our question-filtering system, and yet I see no evidence that this approach has yielded a higher proportion of good questions than we used to get under the previous system (where, at one point, "too localized" was a close reason instead of "insufficient research shown").

So my second suggestion is to abandon the "show research" close reason and restore the "too localized" close reason in its place. Why not see whether assessing the worthiness of a question on its inherent interest, rather than on its ability to pass muster under a test fundamentally unrelated to its inherent interest, leads to an increase in the numerator of good questions?

  • Thank you, Sven! – Dan Bron Oct 9 '18 at 2:40
  • 1
    The problem, though, is that while inconveniences such as bad question inflow can be identified and quantified, something like "unintentionally discouraging good question inflow" cannot. Your remedy would open the floodgates on the former while not necessarily mitigating the latter (and certainly not doing so demonstrably). – Robusto Oct 10 '18 at 1:10
  • 1
    @Robusto: My suggested remedy might open the floodgates on bad question inflow—but it might not. The crucial question is whether rejecting questions as either "general reference" or "too localized"—that is, as being of long-term interest to no one and as being of even short-term interest only to the poster—wouldn't turn out to be just as effective at filtering out bad questions as "research not shown" purports to be, while having the additional virtue of identifying genuine flaws in bad questions rather than identifying a formal shortcoming that a showing of research technically satisfies. – Sven Yargs Oct 10 '18 at 17:17
7

One idea I've had for a while (but never got around to posting, as far as I remember) is that we should hold a periodic contest. Lots of other sites have contests like this, although the rules vary from site to site. Some sites even hold multiple separate contests.

Contests like these usually only require a small amount of rep to participate (depending on the format 5 rep is usually the minimum and 125 is nice to have), although it seems pretty unlikely any new users would participate anyway.

I'm not sure what exactly the best format for a contest would be. Here are some things to think about though:

  • Would we be asking questions around a topic? A topic could be anything ranging from prepositions (tag based) to Bullokar's grammar books. This would mean we would have a featured meta question asking us to vote on topics before the actual contest runs. For non tag based topics, it also means that people need to keep a list of the questions that were posted about that topic, unless we decide to make a tag specifically for the contest. In any case, I like the idea of topics because I'd like to see more questions on subjects I like. Also, I'm interested to see what interesting things others can think of for topics.
  • Would we decide winners based on different categories? What I mean specifically is that it might be nice to have categories such as "best question asked by a new user" (Depending on how many people participate, this may not be practical to do every time.)
  • How would we decide who wins? Some sites use a hard metric like number of votes on main and others post candidates on meta and the winner is the one who gets the most upvotes on meta. Or would the goal just be to all participate together? Winning wouldn't mean much, although validation is nice to have. Some sites have answer contests where the reward is a bounty, but this makes less sense when we're looking at questions.
  • How often would we hold the contest? And how long would we be collecting submissions for it? Some contests happen only in January and look at everything that was posted in the past year, although I personally think this is much too long a time frame for this purpose. We don't have to always have a contest active either.
  • Would we want to look at anything other than questions? The obvious other thing to include would be answers, but it might be interesting to look at the more invisible forms of participation, such as edits.

However, the first question to answer is: Would the community participate? If the answer is no, this isn't something we should do.


Here's some links to contests that were held on other sites for ideas:

(Not a comprehensive list by any measure.)

  • Once this thread settles down, in a few days, and no more suggestions are inbound, I'll create a new Meta-thread for any suggestion which seems workable to me, and there we can find if the community wishes to participate (outside the upvotes you'll get on this answer, I mean). – Dan Bron Oct 8 '18 at 23:52
5

I recently tried to get a user to improve their question by referencing the help pages, but the promising "How do I ask a good question?" FAQ was so bland as to be almost useless for showing a visitor how to actually ask a good question on EL&U.

There should be examples of good and bad questions, examples of what we mean by each closure reason, and notes about why those examples are good or bad.

J.R.'s "Please, everyone... details. Please" on Meta.ELL is an excellent starting point. The mod-editable page "What topics can I ask about here?" also has some good material.

To fully implement this suggestion:

  1. Beef up the existing "How do I ask a good question?" help page.
  2. Point new users with off-topic questions to that page.

I've found that some users are open to improving their questions if given specific information while others opt to delete their own off-topic questions. Both help. This approach also reduces the 'hey, who do you think you are?' and 'it's not so bad' reactions by pointing to an objective standard.

If I had to point to a single area that would improve questions, it would be enforcing the "show your homework" requirement. This should be front and center in any guide purporting to give advice on how to ask a good EL&U question.

  • 1
    There's currently a broader discussion of this question as it applies across the network being discussed (well, linked to a duplicate which discusses) over on Meta.se. – Dan Bron Oct 9 '18 at 16:30
  • The "good" examples do exist but they're hidden away on the SWR tag info page english.stackexchange.com/tags/single-word-requests/info. How many users have even seen that list?? I think it's good that the questions listed are not the most popular, and not all of them include a sample sentence but their requests are all very clear. – Mari-Lou A Oct 10 '18 at 7:20
  • 1
    @Mari-LouA What I like to find in a good question is a documented honest attempt to solve the problem. Some questions are hard to attempt without first knowing the answer (SWRs especially), but many aren't. Even failed attempts are good because they document 'likely' solutions that the OP rejects or doesn't understand. Actually, some of your recent questions might work as examples. :) – Lawrence Oct 10 '18 at 7:32
2

But for the most part, the kinds of questions we see are workaday questions from (mostly) non-fluent users of English who are trying to solve small, practical problems.

And yet the site seems to be content with this.

I could probably post many examples, but this one will do just fine.

It's by a new user. As I'm composing this, the profile plainly shows:

  • this user is from Croatia
  • this user has been a member since today

I'll admit, the question is an interesting one, but, given that it was something encountered on an English exam, I figured someone might think to give ELL a mention. Yet no one bothered, and this seems to be par for the course.

Six years ago, RegDwight proposed a new exchange for English Language Learners, primarily because of the high influx of questions from that demographic. Three years later, the site graduated from beta. Yet when new ELU members drop in and post a learner's question here, it seems they are much more likely to get an answer or two than any kind of comment nudging them toward ELL.

That's fine, if that's what this community has decided it wants to be. But if you're looking to spice things up with novel, interesting questions, the community might start with a more concerted effort to be who they claim to be.

When I visit ELL, I expect to find questions from people struggling to learn our wonderfully vexing language. Because of that, I very much enjoy the good ones. But when I visit ELU, I, like Dan, am hoping to find something refreshingly different. Instead, I often hunt hard to find a few novel gems sprinkled among many questions written by learners who seem to have stumbled among the wrong exchange – and no one seems to notice or care.

  • Thanks for the thoughtful reply, as always. I will say the ELU is pretty liberal with the gentle comments towards ELL, and by far our most common close reason (research) has a direct link there. But there are two headwinds preventing head mechanisms from effectively redirecting the flow (despite that everyone involved’s strongest wishes for it to work: (1) sheer volume. There are 10, max 25, regular users on ELU in a given week, spread out over continents and time zones. We can’t possibly comment on or close every question. (2) OPs simply do not care. They want answers, now, not instructions – Dan Bron Oct 9 '18 at 23:43
  • I can with absolute certainty assure you we all notice AND care. It’s just that our current toolset is impotent to stem or redirect these questions to ELL. And, as I said in this question here, no workable solution has presented itself to lower or redirect the number of off-topic basic questions here. Hence the focus in this thread on raising the number of interesting questions, in absolute terms. In the hope that you, and I, don’t have to scrounge so much to find one. – Dan Bron Oct 9 '18 at 23:46
  • 1
    How does your answer help attract interesting answers on EL&U? Neither site lacks new questions, it's the quality/effort invested (spell checking and citing dictionary definitions) and some sort of thoughtfulness that is sorely missing. -1 – Mari-Lou A Oct 10 '18 at 7:45
  • @Mari-LouA - I’ll admit, my answer does little to “attract” good questions. However, so long as ELU keeps fielding learner questions in large numbers, the density of good questions will always remain low – and I think that’s been problematic. – J.R. Oct 10 '18 at 11:14
  • 3
    I think this does address attracting interesting questions. First impressions are important. @Mari-LouA Log out of SE and pretend that you've never seen EL&U or ELL before and look at the questions on the front page. Does EL&U look like it is the type of site it aspires to be? A Ferrari dealership doesn't fill up their lot with pickup trucks. Making sure that the community has similar interests is important to generating content that is "interesting" and attracting new people that enjoy that content. Actually, this might be a good question for Community Building. – ColleenV Oct 10 '18 at 12:10

You must log in to answer this question.

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