15

I have no idea whether this question is even answerable, but I have noticed we have a positive plethora of questions closed or put on hold by the community, either for being too basic, speculative, or lacking prior research.

Is there anything, within the SE format, we can do about improving the quality of questions?

6
  • 3
    Answerable it is. Whether the solutions are such ones, and work efficiently and effectively is another problem. – LаngLаngС Dec 2 '19 at 11:30
  • That's my concern! – TheHonRose Dec 2 '19 at 11:36
  • 7
    I have no answer, but I strongly support the question. – Mark C. Wallace♦ Dec 2 '19 at 12:16
  • Perhaps clarify somehow that "closed" is just an ugly symptom observed, not the problem. The problem is bad Qs primarily, and then only second lack of RO-worthy edits, but thirdly trap-door-closings (the latter perhaps aggravated if Qs would be salvagable?) – LаngLаngС Dec 2 '19 at 13:49
  • 1
    @LangLangC - I think I understand, I was using the on-hold/closed views of the site members as the symptoms of the underlying disease - ie, poor quality questions. – TheHonRose Dec 2 '19 at 14:06
  • 2
    I think we can customise the on-topic page in our Help Centre to include links to meta questions that make our expectations explicit, in much the same way as the similar page on Skeptics:SE does. – sempaiscuba Dec 2 '19 at 19:24
8

I'm going to take another swing at the dead horse. Other SE have a convention/policy/cultural norm that if the question doesn't include explicit reference to preliminary research/attempts to solve the question, the question is immediately closed (with the intent to revise and re-open).

Closing questions is easy; the complicated part will be:

  1. Build the cultural norm in such a manner that we're not perceived as unfriendly (because apparently it is not unfriendly to ask strangers to do simple basic research for you).

  2. Establish the expectation that we will assist users to revise and reopen.

  3. Accept that we're going to lose a significant number of new users. There is an inherent tension between "question quality" and "friendly to new users" - we can mitigate that tension by crafting our message but either we are welcoming and friendly and accept low quality questions or we are scientific and rigorous and raise the bar.

Two very good comments that I'll summarize by emphasizing that the challenge is to craft standard messages that communicate our expected standards/cultural norms without conveying undue offense. Someone needs to draft those messages; the consensus is that my tone is too harsh.

The messages should be short, but can reference broader discussion in history.meta.stackexchange.com

6
  • 2
    @MarkCWallace Spot on. I have always been vocal in urging "kind" treatment for new/young/naive users, but, arguing against myself, I do not want to see History:SE become a kind of YahooQ&A. If we are to be seen as a serious History forum, I think we have to exercise some "quality control" on questions, requiring at least some basic, documented research - even if it is just Wikipedia, chucking out push /counter-factual questions, and, generally, tightening up our criteria. I admit I am not sufficiently au fait with the inner workings, of the site, but think there must be a way to do this. – TheHonRose Dec 2 '19 at 12:46
  • 2
    As a broad and general marching order, I'd agree. But how to do that? Eg, concrete: How to educate the militarists here that UVing and answering every asinine milQ without much merit is unwanted? Plus recently I learned that such metaQs as this don't carry much weight to begin with (lack of representativeness/ participation). Then on principle the point 1 here is inherently impossible to achieve (also hinted in 3) if we interpret CoC as MSE does: We need to try objective, universalist standards, but a CV/DV will always be perceived by some as unfriendly (added: sometimes they even are!) – LаngLаngС Dec 2 '19 at 13:33
  • 2
    Just to express the notion: I oppose that supposed consensus on 'tone too harsh'. Eg: The 'standard comments' might need expansion or further refinement, & I don't want them to be demotivational or even harsh. But how they have worked in the past for increasing quality remains a mystery to me. We explain, IMO friendly "what's up?", we are helpful & friendly, but comments still get criticised, answerers jump ahead, Qs do not get improved and newbies are rarely turned repeat customers with increasing quality Qs. The ongoing (if seasonally varied) influx of one-off OPs is just an added problem. – LаngLаngС Dec 2 '19 at 14:54
  • 1
    The meta question was invoked because my wording was too harsh; last week I was told that I was "conducting an inquisition". I am content with the conclusion that my QC efforts are too aggressive, the core question is what level of QC do we want? – Mark C. Wallace♦ Dec 2 '19 at 15:25
  • 3
    Agreed on the core. (But I must have missed that accusation. Otherwise, when you did use one of the boiler plates for that QC, I would take that more of an illustration that either the boiler plate needs improvement or much more likely that 'some will feel offended, no matter what'. And while trying to accommodate such feelings, to a degree, we can't just discard any criticism based on sometimes/often quite unreasonable sensibilities. Constructive criticism is vital to prevent this from becoming a dumpster. That is surely not 'too agressive'. – LаngLаngС Dec 2 '19 at 15:43
  • 3
    Taking undergraduate study as a baseline (which I do not think is too high), it is usual in the final year to submit an extended essay of the student's choice. The role of the academic supervisor in formulating the question is critical here, guiding the student away from too easy/been done to death/too large a subject etc. It is a very foolish student who would ignore such advice, or take any criticism personally. Sadly, this does not seem to be the case here, and efforts to help improve questions are often seen as confrontational, prompting angry responses in return. Solution? IDK! – TheHonRose Dec 2 '19 at 19:21
3

One thing I would caution about: We are heading into one of those periodic times when we start getting a blizzard of "Please write my History essay for me." questions. This will come to an abrupt halt when the term ends at the end of the year, and won't pick back up until late spring (the other traditional time when poor students start to look at their marks and worry about the coming judgement day).

Point being, take care to separate our typical background question traffic from our "seasonal" traffic. Its good to have a policy for that kind of question (and I think we do), but lets not mistake what's currently happening for a new normal.

2
  • 2
    Eternal September? After knowing that meme I can't even listen uninfluenced to song titles anymore, Wake Me Up When September Ends – Song Meaning. But can you quantify/guesstimate the numbers for each? Some essay-based Qs could be great on-topic, good quality Qs here, and what is the expected baseline for "usual level of bad" (so far/recently)? Sheer absolute numbers seem in slow decline. – LаngLаngС Dec 2 '19 at 15:02
  • 1
    But the journal metrics are up. Our quality can’t be down! Eternal September applies to original research too now. – Samuel Russell Dec 3 '19 at 6:53
3

Two particular undesired outcomes seem to happen too easily.

  • A user jots off a question with no research. They don't know our standards or are attempting to freeload. Imagine if they were asked to check a box affirming that "This question cites prior research from at least two sources."
  • A user whose question is closed concludes that they are out of options, especially when the stated reason to close seems to miss the mark. Imagine if they then saw guidance saying "to get your question reopened, do the following...."
1
  • 1
    The first option would require a change to the SE design - not something we have any control over. It might be an idea that is worth suggesting on Meta:SE though. The second option may already be in place. See the post New Post Notices rollout on Stack Overflow for more information. The new post notices came into effect today. – sempaiscuba Dec 5 '19 at 23:04
3

Lead by example

One thing we tried on the (now discontinued) Startups SE was to try to lead by example.

We'd set a theme on meta, and then invite regular users to post questions on that topic. The moderators picked themes from a meta question to coordinate the whole thing. The experiment was fairly successful insofar as I can recollect.

One mistake the mods did was to periodically pick cringeworthy themes. These led to obviously forced questions that attracted very low quality answers. Another mistake was to cycle through themes too quickly. It was one theme per week or fortnight (can't remember); a better rhythm might have been a month or two. These mistakes seem easy enough to avoid if we try it on History. The benefit besides a regular stream of good questions it generated is that newcomers had questions that met expectations all over the site to read before asking.

The main issue I'd be concerned about if we were to try this for History is whether we're able to find a large enough pool of non-trivially searchable questions. ;-)

4
  • 1
    This very oblique to how I read the question. And very good. Thought that this was mainly about training newbies, not just posting new Qs. But your idea should indeed have beyond halo effects of mere statistical number improvement or ratios, namely: 'established users asking more questions'. – LаngLаngС Dec 7 '19 at 11:15
  • 1
    @LаngLаngС: yeah, the main benefit besides activity was that newcomers could see a good collection of well received questions on the front page at any point in time. – Denis de Bernardy Dec 7 '19 at 13:55
  • On a more general note: it might be worth a try to come up with such a 'pool' (for a contest?), but it strikes me somehow as 'artificial'. The root of it still stands though: why don't high-reps post more questions; what even explains this reluctancy? – LаngLаngС Dec 7 '19 at 16:31
  • @LаngLаngС: It's the same reason why high rep SO users seldom ask questions. Namely, they're proficient enough with the use of a search engine that they're able find answers without depending on random strangers. Speaking as a high rep user on both sites, my questions on both fall in three piles. The first is questions that I was hoping someone might know the answer to off the top of their head before I looked it up. The next is unimportant side questions that I was too lazy to look up myself. The last is questions that I wanted answered by experienced users. – Denis de Bernardy Dec 7 '19 at 17:11

You must log in to answer this question.

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