9

TL;DR: how to address low-quality posts by new-users without being a jerk or otherwise scaring them away?

Introduction

I'm writing this question after reading this Stack Overflow blog on the need to be more welcoming. This blog focuses primarily on implicit bias (links to accessible BBC article), which I have not perceived as a problem here on ELU (note that I'm not saying it isn't a problem here, I just haven't consciously experienced it as such).

The actual reason I am writing this is not the SO blog, it is because of some run-ins I've had with new users. The problem I experienced is as follows:

I see a new user posting a well-intentioned answer, however, it doesn't meet the standards because it doesn't answer the question or it lacks sources (or both).

In the past I have pointed out my concerns, asking the user to add sources or point out that their answer is interesting but doesn't answer the question. I see how this may come across as somewhat passive-aggressive: in their view they have taken the time to sign-up, write an answer and then I barge in saying I (we, the community) expect them to have higher standards. Obviously I can take a bit more time and sweet talk it a bit, but that doesn't change my message: you (the new user I am addressing) have to put in more effort.

Example

An example of expressing my concerns is the following comment I made:

Please edit your comment in your answer. Also add sources to support your answer if possible. link

The comment is concise, not impolite, but also stone cold, it comes across as ordering the new user around. The new user doesn't feel very welcome and I come across as the bad guy.

Then some other community members add comments, I will cite those as well:

Hi Adam, welcome to the site. It's good that you're enthusiastic (15 posts already!) but I'm afraid they're not being well received as they don't match our site's standards. This site aims for questions to be answered "the way an expert would". This means explaining your answers and justifying them with dictionary definitions and links to them. See e.g. Nigel J's answer for a good example of how to do this.link

Your answers are appearing in the Low Quality review queue and are at risk of being deleted if they're not improved.link

Welcome to EL&U. As AndyT said, there are rules for the site and the whole SE network. You can check these out at the help center or take our tour and visit the English Language & Usage Meta. Thanks! link

Those two users are polite, but I don't think it comes across to the new user as very welcoming. Putting myself in their place I could read into the three comments as follows (in order of the first to the third; playing devil's advocate): 'nice try, but we expect (even) higher standards', 'they are even appearing in the low quality queue now, at risk of deletion', 'here are some links: go educate yourself'.

We could easily vote down, delete or otherwise deter those users, but that's not very welcoming. Furthermore, it counteracts the many calls for attracting more users who are willing to participate.

Defining the problem

Putting it briefly, the problem, I think, is how we (myself included) approach these new users. We (tend to) come across as bossy, ordering them around and saying they aren't doing well enough.

I am not blaming myself or any of the other community members because it really is difficult to say you expect higher standards and not come across as bossy.

Tackling this problem

I am proposing to open a discussion to look into ways of telling new users that their answer doesn't meet community standards without scaring them away.

To do so, I think we need one or two good sentences conveying this and a more extensive explanation in the help centre or on this Meta. Since there are many of these well-intentioned answers we can easily copy these two sentences and a link to the extensive explanation to those new users.

PS: if you think I am reading too much into things and you don't perceive this as a problem then don't hesitate to add a comment (or even an answer if the mods will allow it).

Edit

I just found this post on Meta Repository of polite responses. It provides many good template comments, however, I still think it might be a good idea to have some Meta or help centre post to link to (to make sure the new user doesn't feel scolded). There is How do I write a good answer?, but that might need to be extended.

For example, it doesn't mention (low) quality at all. Searching the help centre for 'quality' doesn't yield results explaining 'low quality answers' (there is an entry for low quality questions).

  • 11
    I am a bit frustrated by the chiding tone of that SO blog post, directed essentially unilaterally against the userbase that is so loyal to SO, has made it was it is, and has contributed and curated 100% of its content. By contrast, I am very pleased with how you’ve framed this Meta post, it’s time, and how it takes a fair and even handed view, looking at it from both sides and offering constructive criticism (and, refreshingly, not “directed at everyone but he author of the critique”, as is the usual), and a clear goal with helpful suggestions. A big fat +1 from me. – Dan Bron Apr 27 '18 at 21:33
  • 5
    Another blog post is very relevant here. As that post says, it's the condescension that becomes a really big problem. (I say that knowing full well that I'm also guilty of this; unfortunately it's something that I've known about for some time but have been unable to fix.) Also, I think the whole "race/gender" issue outlined in the SO blog is actually about the statistics from the Developer Survey, as I elaborate here. – Laurel Apr 28 '18 at 4:42
  • I think the biggest benefit will come from question templates, currently being trialed on other sites. – curiousdannii Apr 29 '18 at 13:14
  • 3
    @DanBron I read that blog post much more as a mea culpa from the owners/programmers of the site, acknowledging that they haven't taken steps that the userbase has asked for (e.g. having a different landing page for brand new users who are about to post) and also for the fact that the basic SE model may be especially off-putting to some demographics. – 1006a Apr 29 '18 at 16:36
  • 1
    The scientific validity of "implicit bias" has been debunked (WSJ and others)... – shoover May 4 '18 at 18:32
  • But relevant to this question, how can we possibly reply in a welcoming manner to this question, which, since it will probably be deleted, consists of "i have an annoying brother what should i do? i have no clue. anyway thank you in advance. i have already tried being nice it has not worked, and he calls me names, and makes noise just to annoy me a;so he is so annoying that i can't even be in the same room as himi don't have a link" and some random formatting attempts. Yes, it's in ELU, not IPS. – shoover May 4 '18 at 18:34
  • Assuming we accept the axiom that there is a problem, is it primarily with new askers or with new answerers? My home base is SO, but I wander throughout the SE universe, and IMHO there are far more bad first questions than there are bad first answers. Most of the bad first answers are genuinely trying to be helpful, and just need to be trained how to do it better. Most of the bad first questions are badly asked, or provide no context/code/details, or are duplicates of questions already asked and answered multiple times, or are just plain incomprehensible. Oh, I forgot "gimme teh codez". :( – shoover May 4 '18 at 18:57
  • Standards for how low-quality answers/questions are inconsistently applied. – jxh Aug 30 '18 at 23:32
10

(I'm not sure that this'll count as an answer. Or even be useful. It's just my ramblings based on the question.)


Yep, I totally agree that experienced users commenting can come across as condescending. I know I'm guilty of it - hell, two out of your three example quotes are mine! (That said, I thought I'd done a better job than normal on those of avoiding condescension... apparently not!)

So, why does this happen then? Mainly it's because I come across a lot of low quality answers (this may be due to frequenting the LQP review queue...), and what I want to do is downvote them into oblivion as being useless. But that often seems mean to an innocent newbie who's only trying to help. So I try and bottle my frustration and write helpful comments instead... but they are often still coloured by my frustration. Often I'll write a comment and then delete it (and skip the review) when I realise I can't Be Nice while saying what I want to say.

So, canned comments can help with this, as at least they'll have been written by someone in a calm state of mind. I think canned comments on the "review answers from new users" would be useful. And a bank of canned comments (much bigger than that you linked) would be useful. NVZ had some good ones (though maybe he's more active on ELL than ELU? I'm not sure.)

Mari-Lou A in comments has suggested idownvotedbecau.se as a good site for canned comments. On SO people will just post a link to the relevant page from that site. One from there that's very relevant to us is idownvotedbecau.se/noresearch. [Note that I've seen these a few times and thought they were fake/joke links, and the "no research" in the comment was the whole message. If you follow the link there's actually a nicely written out explanation of why "no research" is a problem.]


But I think the real problem is that the barrier to posting as a new user is too low. Users can come in from google, read a question, and decide to post an answer without knowing anything about the site. They don't even need to read any other answers on the page.

Trying to get people to edit their answers to fit our Standards, after they think they'e already done, is not going to make them happy. We need to enforce Standards before they post. We need to tell them, up front, something along the lines of:

Please don't post an answer unless it meets these criteria:

  • It adds something new that others haven't already covered

  • It's backed up by evidence. Dictionary definitions, properly referenced (see [here]{link} for how to reference), are often the best way to do this. Alternatively examples of actual usage may be good, but again they need to be referenced. NOTE THAT YOUR NATIVE EXPERIENCE IS ALMOST CERTAINLY NOT SUFFICIENT EVIDENCE.

The existing info when you try and enter an answer is minimal. If a newbie tries to enter a one word answer, they get told that it's too short. But they don't get told why it's too short. (Or do they? I may be wrong. I think it just says "we're looking for answers with some sort of explanation"? Maybe we need to be less polite here and more forceful: "*If your answer doesn't meet the minimum character limit, it's because your answer is no good. A one word answer may be wrong or it may be right. We want answers that prove they're right, with explanation and research to back it up.*"


The other problem is that many users simply don't care. For your linked example there was a model answer already, but the newbie didn't follow its example. I then pointed the newbie to the model answer, but he chose to ignore the suggestion to incorporate it.

Hence why I think we should make posting harder. Will it put people off? Hell, yes. But those who are prepared to understand the Standards are those who we want to join, because they'll actually respect the Standards. The lazy ones who want to post their one word idea and leave aren't who we want anyway.


I have no idea how implementable my ideas are, or whether they'd require SO staff to write something specific for ELU. This whole post can probably be disregarded as rambling. I feel better for writing it, and considered not posting it. But I've decided to go ahead and post it anyway.

  • New users are already given advice upfront before posting an answer. The problem is it's super generic advice. Hopefully it can be changed... – Laurel May 1 '18 at 22:03
  • 1
    "The other problem is that many users simply don't care." It's often established users with thousands of rep who don't care about site standards. – curiousdannii May 2 '18 at 1:59
  • @curiousdannii - In comments, I agree. In low quality answers it's normally new / low-rep users. – AndyT May 2 '18 at 8:05
  • 1
    I agree, some structural solutions would have a much bigger impact than any piecemeal changes individual contributors can make. From the blog post referenced in the question, it sounds like there may be a new "landing page" for brand-new askers coming soon; if that works out well, perhaps the development team will turn its attention to something similar for brand-new answerers, as well. ("We’re planning to test a new “beginner” ask page that breaks the question box into multiple fields – one for each of the key things answerers need to help") – 1006a May 2 '18 at 19:03
  • 3
    I've discovered this on SO idownvotedbecau.se whaddya think? Too direct? Click on the "research" link, I think it's well written and could easily have been scripted for EL&U. My main reserve is its presentation style, it looks too severe. – Mari-Lou A May 3 '18 at 8:34
  • @Mari-LouA - I likey! But I've seen those before and assumed they were jokes, not an actual link. Only by actually clicking your link did I discover that it's an actual useful explanation. Still, I might try it. – AndyT May 3 '18 at 9:09
  • 1
    @Mari-LouA That link deserves more prominence. :) – Lawrence May 3 '18 at 9:16
  • Stack Overflow did an experiment with a structured template for their questions. Perhaps we could run our own experiment here as well. – Lawrence May 3 '18 at 9:19
  • I don't have the time nor the inclination to turn this into an answer but I'm glad we're on the same wavelengths here. The no research page is rather good. I've got it. It lacks a touch of humour, we at SE are taking ourselves too earnestly nowadays. – Mari-Lou A May 3 '18 at 9:23
  • @Mari-LouA - Ok, I've shoehorned into my own answer then. – AndyT May 3 '18 at 9:29
  • @Lawrence - Hmm. They're calling it a small success, but it's very small. 3% fewer "bad" questions (i.e. those downvoted into oblivion), which were evenly converted into "neutral" questions (no downvotes, but no answers either) and "good" questions (upvotes and/or answers). Not as encouraging as I'd have hoped. – AndyT May 3 '18 at 9:31
11

While there is the occasional rude individual, perceived but unintended slight, or even an otherwise polite user acting out of frustration, I suspect that the lion's share of the tension comes from a mismatch between who the site says it's for and who actually turns up to ask questions.

  • Who the tour says EL&U is for: linguists, etymologists, and serious English language enthusiasts.

  • Who turns up: mostly, people who aren't linguists, etymologists or serious English language enthusiasts. By some accounts, the number of EL&U users in the first 2 categories can be counted on one hand with fingers left over.

Perhaps it's time to rethink the wording in the tour.

  • Only three users then? :P I can name four. But point taken. I upvoted ages ago so this is me just saying "right on brother!" – Mari-Lou A Apr 29 '18 at 18:33
  • Haha, you might be right :) . But are all 4 still active on EL&U? – Lawrence Apr 30 '18 at 0:51
  • By "reword the wording in the tour" do you mean "officially change the site's mission statement to include casual users"? Does changing the mission statement mean we will accept single word answers? If not then this doesn't address the actual problem. – AndyT May 1 '18 at 15:37
  • @AndyT Yes, I mean a recasting of the official target audience. About single word answers: as far as I know, Stack Exchange has always required answers to be supported. That’s the issue, I suppose: if we want a community of non-expert English users, we already have ELL. If we want a community of expert English users, there will be some that feel disenfranchised. Perhaps what we really want is an expert community that non-experts can pose questions to. This non-homogeneous composition would be an interesting experiment. – Lawrence May 1 '18 at 18:05
  • In that case, I think we just merge the two site. The professionals can ask on [linguistic.se] – Ooker May 3 '18 at 12:34
5

It is undeniable that the minimum site expectations of an acceptable answer must be clearly explained to the new member:

(1) it should answer the question as asked by OP

(2) it should be adequately supported by appropriate references.

So we only need create a standard message that says just that and is prefaced by a welcome to ELU. This standard message can be easily copy-pasted as a comment by any user. This type of thing is being done to guide both answers and comments from new users on Interpersonal Skills SE.

Sample message (comment):

Welcome to English Language & Usage and thank you for posting this answer. Please note that answers on this site should actually answer the question (rather than be a commentary or anecdote) and answers need to be supported by adequate references. Could you update your answer to follow these guidelines? Please see the Help Centre for more details.

Note that I am not saying anything about "high standards", "low quality" or "linguists" so there is only a focus on the minimum expectations of an answer from this website. This sort of comment can only rarely be interpreted as unwelcoming or hostile to a new user. We are simply pointing them in the right direction and I don't see any "negative" wording here. Can you think of any other "minimum requirements?" -- if so, then please point out in comments.

  • It's not the LQAs that's the "problem", they rarely cause the most seasoned of users to foam at the mouth. It's the VLQQuestions that prickles most users or haven't you noticed? Is it just me? – Mari-Lou A May 1 '18 at 7:38
  • You are right but OP of this meta post is specifically asking about LQA from new users @Mari-lou A, as OP clarified in a comment [extract: "(...) You mostly focus on poor questions whereas I was focusing on poor (but well-intentioned) answers. I think as it is, questions can be dealt with easily by closing / putting on hold (...) When answering a question that's not necessarily the case, an answerer may feel they've already done their part. – JJJ 2 days ago"] – English Student May 1 '18 at 7:46
  • True, true I had forgotten that. I think though it becomes a greater problem when the newcomer has accumulated rep but keeps getting DVs from users who argue that questions are off topic, or answers should be supported, or users need to check whether there are duplicates before answering, or citations must always be attributed correctly, or that OED is not the same as EOD, or that a question is GR (gen. reference), or that answers must show original content, or... etc. etc. – Mari-Lou A May 1 '18 at 9:59
  • It can get a little overwhelming to be faced with so many "guidelines". I'm positive that we've scared off some (potentially) very good native speakers contributors in the past. – Mari-Lou A May 1 '18 at 9:59
  • Yes indeed, and I saw some of them go: but each sympathetic senior can make such a difference! I learned all about it as a new user on multiple SE sites over the last 1 year, and found that the guidance of the few highly welcoming and encouraging members like you can help a committed new user to get through the "newbie" phase and learn to contribute towards what the site really wants to achieve @Mari-lou A. – English Student May 1 '18 at 12:03
  • 1
    Making that sincere effort is all we can do @Mari-lou A. If the new user is unwilling to learn then he is not a good fit for SE and won't last long here. The next new user might get an anonymous DV/CV because the most friendly member can become tired after all that wasted effort! I do agree that we have a lot of new users who don't want to learn how the site works. Maybe only 1 in 30 new users becomes a regular contributor here. – English Student May 2 '18 at 11:56
  • I meant to delete the previous two comments because I just wanted to show that even when experienced users make the effort to reach out to newcomers and offer good friendly advice it can be ignored, and I mean really ignored. Not once has that user replied. Maybe they were offended? Boh. – Mari-Lou A May 3 '18 at 8:32
  • I been thinking about it and "offended" seems both unlikely and unjustified; maybe many new users don't want to communicate with regulars about English L&U or the most useful way to approach this website: maybe they just want to get their homework done and that's it. [So much hassle to get an answer for my homework?] That puts a whole new complexion on being "welcoming" to new users and raises another meta Q: "how to deal with new users who don't want to know how this site works?" If I do post it, please write an answer @Mari-lou A and we will take this discussion forward on that page! – English Student May 3 '18 at 13:03
  • @EnglishStudent - I want to ask you a favor, if you don't mind -- I remember, back when you were a fairly recent arrival at ELU, I wrote a comment below one of your posts, with a tip for participating successfully on ELU, and I remember that you took it pretty well -- which may say more about your inherent graciousness than anything else -- but I wanted to look up what I wrote to see if it would be a helpful contribution to this question. But I can't find it. Maybe you remember better than I do approximately how far along you were in your ELU trajectory, or what the topic was? – aparente001 May 6 '18 at 20:05
  • Let me try to remember @aparente001. I answered a huge number of questions in those early days and some comments might have been deleted for one reason or another (or, more likely, moved en masse to chat.) I will try to find your comment and get back to you with the link. Of course you are one of the most welcoming senior members on ELU! – English Student May 6 '18 at 20:11
  • @EnglishStudent - (cough) thank you. I don't consider myself a senior member. But some of the lessons I've learned along the way about what's expected at SE and especially about ELU were more painful than they needed to be. Hence my interest in this question. – aparente001 May 6 '18 at 20:39
  • "some of the lessons I've learned along the way about what's expected at SE and especially about ELU were more painful than they needed to be." __ I remember your comment now, just struggling to locate it @aparente001. Now I think it could even be on meta! Please check back here in 1-2 days to see if I could find it (in case you missed the ping.) – English Student May 6 '18 at 21:34
  • @EnglishStudent - Oh -- I forgot to look on Meta! I will. Anyway, no hurry, and if it's gone, it's gone. Thanks. – aparente001 May 6 '18 at 22:00
  • You are most welcome @aparente001. – English Student May 6 '18 at 23:45
5

SO could help with setting an expectation of required editing of OP:

?"We understand that if you are confused or stuck it might be hard to word your problem on your first attempt (but please try! (link to how to ask a question) you will get better results)"

?"Most questions will need clarification from you - please be prepared to quickly edit your question especially in the first hour after asking based on comments you receive asking for more information. You may need to make multiple edits to your question."

?"It is natural for volunteers expect the return courtesy of contributing time to improving a question and the overall quality of the site benefits from clear questions with clear answers that will be useful to others in the future."

?"Insufficient answers can also be a problem. Please do not be offended if another answer very similar to your own but with more detail is added if your answer is less complete or insufficient to the guidelines of the given SO forum"

On that last point, I think a culture shift here to encourage a more complete version of the same general answer could be encouraged rather than discouraged because it came in later.

Ideally, a comment might give a person a window of time to improve an insufficient answer but rigid rules about that would not make sense - that is more a matter of board custom. Also, two of the same 'answer' could both be useful with different information for each and rather than voting one down and one up, perhaps people could be more inclined to vote both up or remain neutral on one and vote for the one they feel is more complete.

4

This is the first introduction to your pages. Pretty nice, right? Nowhere does it say "Please begin your tour by looking at our standards for answers and questions. Then dive in. You will be encouraged to modify either your questions, or answers, or comments along the way. We hope you find it inviting enough to return and stay awhile." (Can you tell I wrote ad copy at one time?)

enter image description here

  • Where did you see this? I thought I was pretty familiar with the system, but this is completely new to me :/ . It looks like something automatically generated that would be the same across the network (with changes to the name and links, obviously). – Laurel May 3 '18 at 4:25
  • I think it is part of the SE "welcome email" for new users @Laurel. – English Student May 3 '18 at 13:32
  • @EnglishStudent you’re right. I never read mine. But it’s the exact same email except it’s for the computer science site (the first site I joined). You only get one such email, so I guess it has to be generic . – Laurel May 3 '18 at 19:32
  • 1
    I agree that the SE introduction, with its welcoming tone and omission of any indication that EL&U has fairly strict standards regarding what constitutes an acceptable answer, sets newcomers up for a nasty surprise when they provide an answer off the top of their head and draw downvotes and criticism from site users who are enforcing the site's actual standards. The newcomer wants to help and generously shares his or her knowledge—and then gets called out for answering improperly. That's a serious disconnect that SE itself encourages through its breezy, no-strings-attached welcome message. – Sven Yargs May 3 '18 at 22:16
2

I consider this the most important part of the blog post:

Too often, someone comes here to ask a question, only to be told that they did it wrong. They get snarky or condescending comments for not explaining what they’ve tried (that didn’t work). They get an answer… but the answerer gets scolded for “encouraging ‘low-quality’ questions.” They get downvoted, but don’t know why, or called lazy for not speaking English fluently. Or sometimes, everything actually goes well, and they get an answer! So they thank the poster… only to be told that on Stack Overflow, “please” and “thank you” are considered noise. All these experiences add up to making Stack Overflow a very unwelcoming place for far too many.

It's striking that SO experiences the same problems as EL&U, as expressed in a variety of posts on EL&U meta.

The possibility is that there's something wrong with the basic model.

The model is entirely based on discipline--what you must do, what you must not do--not only as a new user, but as an experienced and frequent user, one who, as Dan Bron points out, provides and curates the entire content. Scolding, admonishing, instructing. Telling users, new or not-so-new, what we do not do (homework, proofreading, SWRs without a sentence, reference questions, things easy to look up, etc. etc. etc.) and what we must not do (especially answering questions in comments and failing to provide references).

This is all done in an effort to improve the quality of the questions and the answers, but the process itself--the curating--is time-consuming, mind-numbing, and productive of an annoyance that starts to show up in comments. (Did you look up those words? And just what was it that you didn't understand? Where is your research?) And these comments are the main reason for the perception that the site is unwelcoming.

So, I would recommend an experiment: Give up all the negative feedback and the instructions. Ask a questioner to clarify only if you really want additional information so you can answer. Ignore any questions you consider too poor or confused to pursue. (While avoiding negative feedback, I'd accept using the voting tools. Let people go to the help pages to figure out why a question or answer was downvoted.)

Clean up the leftovers with a couple of programs--delete any questions that have no comments after 48 hours (or whatever), delete any questions with no answers after 3 days or so. Let the voting do the curating.

  • 1
    Please read carefully. I never suggested removing answers without upvotes. I've posted to meta before and gotten the same sort of response--criticism based on intentional misreading. – Xanne Apr 28 '18 at 17:57
  • 1
    If you take the approach that you get more with honey than with vinegar, there is no need to instruct users on how to answer questions. If they want to know why they got downvoted, there's material on help (it is there, isn't it?). I am not enthusiastic about the close-vote routine--the reasons are unwelcoming, part of the discipline. – Xanne Apr 28 '18 at 18:30
  • I intended my extended comment to apply to both questions and answers and to new and more experienced users. – Xanne Apr 28 '18 at 22:53

You must log in to answer this question.

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