Improving "answerer" experience

Question

This is gonna be "meta" and not have a specific question in mind, but an invitation for discussion. I think that's in-scope for the meta areas...

I've been a member of this place for "2 years, 7 months" according to my profile. In all that time I've not asked a single question. Mostly because I've 30 years of Unix experience and the couple of times I started a question I worked out the answer ("talk to the bear" methodology).

So I approach this place as someone who wants to answer questions; to help others, and (hopefully) teach along the way.

Today I look at "newest questions" and if something seems like it's worth writing an answer then I'll go for it.

I'm wondering if there is a way of making that experience better.

One idea is to allow sufficiently reputable people to add answers to closed questions (perhaps with review requirements). A number of questions get closed "duplicate", but if you look at the duplicate question then it doesn't fully answer the asked question; there maybe a 80% overlap, so the original questioner may not be satisfied.

Another idea is to be able ignore users or tags. There's one user who is a prolific asker of questions which Pauli would have said "That is not only not right, it is not even wrong!". I've tried answering these questions, in the past, and then got frustrated with the comments, which frequently try to extend the scope of the question and demonstrate the original question wasn't what was really intended. Whenever I see a question by this user I want to close the browser tab, which means that other questions don't get a look in.

I like to answer questions (or, sometimes, just to nudge people in comments to allow people to answer it themselves). In the real world I run a "three strikes" rule; if you ask me a question I'll try and answer it. If you come back with essentially the same question then I'll assume I didn't do a good job first time and go into more depth. A third time and I'll explain it in so much detail... you better not ask a fourth time.

How do other people approach persistent askers of bad questions, to make their "answer question" process smoother?

Answer 1

1. Are we twins?  I've been here 4¼ years, and posted 432 answers and 0 questions.

2. You talk to a bear?  I talk to a cat (or a committee of cats, if I can get them to stay organized).  And apparently the Stack Exchange standard is to talk to a duck.    :-)

Seriously,

3. Interesting idea about allowing sufficiently reputable people to add answers to closed questions.  You know, of course, that the standard procedure is a combination of

   • Edit the question to clarify that it is different from the "duplicate", and to improve its quality in general.
   • Vote to reopen.
   • Post a comment saying
     • this question is different from the "duplicate" because ...
     • please reopen this question; I have discovered a truly remarkable answer for this question which comments are too small to contain.        :-)
   • Post on Meta, saying the above.
   • Go into Chat and plead your case.
     
     

   I'm not sure we need to circumvent the process, but let me think about it for six to eight weeks.

4. Regarding questions as moving targets; see Should extensive question editing be ruled out?

   D'oh!  I just noticed that you've already commented there.

Answer 2

I am also someone who

wants to answer questions; to help others, and (hopefully) teach along the way.

I'm a little shocked that I've posted over 1200 answers over the 3 1/2 years I've been here so far. I like to answer questions that have some Venn diagram overlap of:

• in, or near, my area of knowledge
• demonstrates a misunderstanding on the part of the asker

Sometimes a question is heavy on the first criterion, and I answer it because it's fun to try something new (sorting in sed or in bash); very often I learn something new along the way.

Questions meeting the second criterion are interesting to me because I like teaching people about UNIX, and it's gratifying to see people learning.

As an answerer, if I'm curious about a question, I'll open it in a new tab and take a look. If my Venn diagram doesn't light up (because I can't understand the question or don't have the knowledge or I can't tell where the asker's knowledge starts or stops or etc), then I close the tab and move on. (This is separate from the "reviewer" side of me that would VTC an unclear question, close as duplicate or too broad, etc; and also voting up/down).

Watching and ignore tags helps to a small degree (obfuscated by the large number of text-processing questions and mis-tagged questions).

This is another reason I wrote https://unix.meta.stackexchange.com/a/5016/117549 -- if an asker manages to get the right words into the title and excerpt, it stands a better chance of me opening it up in a new tab to potentially answer it. If the asker's effort starts and ends with a title like "Need a Linux command for below output", it's a red flag.

I concur with G-Man's thoughts on wrongly-closed-as-duplicate questions.

Regarding moving targets, I try to force myself to leave clarifying comments on questions that could go in various different directions. I try to answer only after it's clear what the problem is and when a solution is possible. At that point, it should be clear that the question is answered, and I'll redirect further (extensive) questions to the "Questions are free!" new-question link. Of course, if the follow-up questions are about understanding the current answer, that's in-scope. Edits to the question should not invalidate existing answers, which is why it's important to clarify the question (or VTC if there's no follow-up).

Answer 3

In response to a question on MSO about providing the ability to ignore specific users, the powers-that-be said, a few months ago, that they're considering offering some "option to mute interactions in a more granular way."

I primarily use unix.stackexchange.com as an answerer, and my interface is the Questions page, most recent first. What I'd find useful is a way to mute a question that I know I won't be answering, so that I don't keep seeing it again and again when I revisit the Questions page.

Answer 4

Way back when I mentioned in chat that I'd like to ignore some users, somebody pointed me to Ignore Users Script ... but that script hasn't been updated since 2009, and didn't work for me. So I threw most of it out and rewrote it. Here:

// ==UserScript==
// @name         SE Ignorer
// @namespace    https://tampermonkey.net/
// @version      0.1
// @description  try to take over the world!
// @author       Muru
// @include        *://*.stackexchange.com/*
// @include        *://stackoverflow.com/*
// @include        *://askubuntu.com/*
// @include        *://superuser.com/*
// @include        *://serverfault.com/*
// @grant        none
// ==/UserScript==


(function (func) {
    // helper for injecting the script into the page
    var script = document.createElement("script");
    // insert a script element...

    script.type = 'text/javascript';
    script.appendChild(
        // with a function which is called on document ready
        document.createTextNode("$(" + func + ")")
    );
    // run the script after other scripts
    document.body.appendChild(script);
})(function(){
    'use strict';
    var userids = [];
    var qlists = ['#questions', '#question-mini-list', '.search-results', '.js-search-results'];
    var targetNode = null;
    for (i = 0; i < qlists.length; i++) {
        targetNode = $(qlists[i])[0];
        if (targetNode != null) {
            break;
        }
    }
    if (targetNode == null) {
        console.log("No question lists to filter!");
        return;
    }

    // Options for the observer (which mutations to observe)
    var config = { attributes: true, childList: true, subtree: true };

    // Callback function to execute when mutations are observed
    var callback = function(mutationsList) {
        for(var mutation of mutationsList) {
            if (mutation.type == 'childList') {
                letsJQuery();
            }
        }
    };

    // Create an observer instance linked to the callback function
    var observer = new MutationObserver(callback);

    // Start observing the target node for configured mutations
    observer.observe(targetNode, config);
    function letsJQuery() {
        $.each(userids, function(){
            $('.user-details a[href*="/'+this+'/"]').closest('.question-summary').remove();
            $('.started a[href*="/'+this+'/"]').closest('.question-summary').remove();
            $('a.comment-user[href*="/'+this+'/"]').closest('.comment').remove();
        });
    };
    letsJQuery();
});

Notes:

• You have to manually fill in the userids array using the user IDs, not usernames (so that's 70524 for me and not muru). Check the profile link: https://unix.stackexchange.com/users/70524/foobar - it's the number before the username.
• It removes entries from the questions page and search results, which were last modified by the listed users.
• I don't do much JavaScript, so I don't know how bad it is, so I haven't updated the original.

Answer 5

I'm a nobody here but I came from the dark side or even darker if you go back far enough and what I did on Ask Ubuntu if the user:

• posted a comment with an additional question is:

  That is an [entirely new question](https://unix.stackexchange.com/questions/ask) to which I don't necessarily have the answer, but feel free to ask a new one: We don't charge extra for an additional question!  **;-)**  Feel free to include a link to [this one]() if it provides context.
  

  which renders as:

  That is an entirely new question to which I don't necessarily have the answer, but feel free to ask a new one: We don't charge extra for an additional question! ;-) Feel free to include a link to this one if it provides context..

• asked me to elaborate: I did, but sadly only did once and then reverted to the above canned comment, but I like your system better, so I'm going to change my behaviour here on Unix & Linux.

;-)