1. 6

This was positively motivated by this comment.

A lot of times I see comments which are somehow on topic, but give the impression that the comment’s author didn’t actually read the post, or at least didn’t stick long enough for the gist. Seeing that more people feel similarly, maybe we can do something about it as a community?

  1.  

  2. 6

    Why do people post?

    1. To ask questions to learn. Good. That’s a big part of this
    2. To tell people things about the article/topic. Good. “
    3. To be in the mix of people talking, even though they haven’t anything to say. Normal human behavior. Ignore or help
    4. To get a reaction from other people. Needs social interaction. You don’t want to engage? Ignore.

    None of this is different from a gathering of people. except that anonymity can encourage people to be more obnoxious that they would be face to face. (And in some cases, it goes the other way too, let us not forget)

    In a gathering of people there would be social cues that you are not keeping with group think. Here there are not. But I find it easier to ignore people and not hurt them.

    I like this. If there is a useless comment, I can gently tell the commenter they should read the article, or not be so rude etc. etc. If I’m in a bad mood, I just ignore it. In some ways this is much better than face to face because I can manage my emotions much better.

    So, in summary, no we should not be going out of our way to ding people for fairly harmless, human behavior on this site.

    Oh. Also, get rid of votes.

    1. 7

      Oh. Also, get rid of votes.

      Although I can see the usefulness in votes, I’m in favor of removing visible numbers. Incredible distraction.

      1. 4

        I added the vote numbers on main page links to my adblocker for a while, because I know I have a personal bias to ignore articles with fewer votes. After I removed that block rule, I felt less like I was going to click a link because of a low(er) score. It might work for you, or anyone else who reads this.

        1. 3

          Your advice might seem pretty obvious but I didn’t think of that, thanks.

          It seems to be .comments .voters .score, or just .voters .score if you want to include the story itself.

        2. 2

          The numbers only appear after a set limit (I believe votes in the range [-4,4] aren’t shown until after some time) and after some time.

          1. 2

            What is the rationale for that? It seems like a tacit acknowledgement that that votes are distracting or otherwise problematic. (I agree with kghose, we don’t need popularity contests here.)

            1. 1

              I really don’t know. I looked through the About page but didn’t find a reason.

              I think starting something like lobste.rs without something like “karma” just doesn’t work anymore. It’s like herpes now, you cannot get rid of it.

              1. 1

                A few days ago I actually spent like 10 or 15 minutes looking through the site source code trying to find where the little ~ comes from, just out of curiosity. Didn’t find it, gave up and went back to work.

                What is the value of preventing voting on new stories or comments? Does Reddit or HN have a similar thing? I don’t really use those sites, although I’m aware that the design of Lobsters is very much a reaction to them.

                As to whether all discussion forums must mimic Reddit’s gamification strategy, I respectfully disagree. I don’t think your herpes analogy makes any sense.

                1. 3

                  That actually hides the vote count for a certain time and/or threshold. The idea was some people will go back and forth in arguments trying to win upvotes from their side. The numbers become part of the mental feedback loop. Hiding them theoretically could reduce that a bit. I dont know if it works.

                  Drawback is I cant see upvotes to know if a specific person found an answer immediately helpful. They often do an upvote instead of comment.

                  1. 2

                    Drawback is I cant see upvotes to know if a specific person found an answer immediately helpful. They often do an upvote instead of comment.

                    And I wonder if that is brevity or a fear of lowering their points/comment score. I myself, prefer a personalized note. Nothing like a personalized typed note, appreciating another human being, sent over the internet under what may or may not have some relationship to my real name and/or personality.

                    Hmm: however that does raise the idea of having a “Thank” button (like some sites) and have that be de-anonymized - you can see a list of … thankers? … - for those who do not want to clutter up a conversation with notes.

                    1. 2

                      Oh I agree with that. The thanks and follow-ups I get make the site that much more fun. :) I just noticed a lot of people are low-participation where they rarely comment. Super, casual users of the site. The most signal I might get from them in a particular thread is an upvote or no vote. Some big fans of mine on both HN and Lobsters don’t even use the site: they just email me on occasion.

                      So, the unusual patterns here make me look for and like several forms of appreciation for effort I put in.

                      “however that does raise the idea of having a “Thank” button (like some sites) and have that be de-anonymized “

                      That’s interesting. Facebook does it with likes. I did find that feature useful as feedback on what people were enjoying or not. Also teaches you about their personal beliefs. Helps avoid ticking people off if you see they respond to this but not that. If keeping it simpler, one idea I had for Lobsters was a symbol or color change if the person you replied to upvotes the reply. Not one for downvotes, though, since that probably just be fuel for arguments. However, having one for upvotes allows a quick, lazy way to send a thanks for people that wouldn’t participate further. The idea came from the fact that upvoting a reply is already a thanks in a way.

                      1. 2

                        a “Thank” button

                        I like that idea. I know some find it gauche, but I personally quite appreciate the GitHub/Slack-style “reaction” feature that lets one respond with the entire range of emoji. Much more nuanced feedback than votes, doesn’t litter the forum with “+1” comments, doesn’t accumulate into a user-worthiness score or reorder the comment timeline. Having it be de-anonymized (on hover) both adds to the personal touch and encourages politeness.

                    2. 2

                      As to whether all discussion forums must mimic Reddit’s gamification strategy, I respectfully disagree. I don’t think your herpes analogy makes any sense.

                      I will try to expand.

                      I agree with you that the gamification of sites through karma count[1] is harmful to online discourse.

                      The perceived pressure to keep score does not encourage frank debate, in my opinion. It encourages groupthink.

                      My point is that nowadays, for many people, being able to hit a like/unlike button or similar is simply how interacting on the internet works. So if sites or services don’t offer it, they’re either pressured to do so, or ignored.

                      I guess using “meme” in its original sense would have been more correct, except that term is hijacked. And the herpes analogy was used as it’s a disease that seldom is fatal, can cause ugly side effects, and is deeply embedded in the host.

                      Note that lobste.rs, like HN, does make alterations to the basic upvote/downvote scheme. New users cannot downvote below a certain karma level.

                      [1] I believe it started with Slashdot, continued with REddit and HN, and has been carried over to this site.

                      1. 1

                        Thanks for the expansion. I think I mostly understood your meaning the first time around, but I still contend it’s a lousy analogy. People (and especially the sort who are inclined to use Lobsters) are much less stuck with a decade or so of mild psychological conditioning than they would be with an actual virus in their spinal fluid. Particularly when they can recognize that the conditioning is harmful. Expectations and norms change all the time.

            2. 1

              So, in summary, no we should not be going out of our way to ding people for fairly harmless, human behavior on this site.

              I generally agree with this, but it’s a very fine line–once you have a sufficient number of humans, even the fairly harmless, human behavior of defecating results in sewage everywhere. So, something must be done.

              Also, get rid of votes.

              Nice Carthago delenda est at the end. ;)

              1. 3

                Nice Carthago delenda est at the end

                @friendlysock dude, that made me LOL. In the office.

            3. 5

              This is a “problem” (if it’s really a problem) that is best solved with community building and the enforcement of social norms, not through a technical mechanism.

              1. 5

                You mean like having a meta thread?

                1. 2

                  “Enforcement of social norms” usually means “decentralized enforcement of social norms” – like social exclusion. Meta threads are part of a centralized system & are often intended to produce technical or legal solutions.

                  If somebody consistently acts like they haven’t read the linked article, they violate my personal expectations for good community ettiquitte & I’m less willing to engage with them or assume that their arguments are in good faith. If reading the article before commenting is a norm, then most people in the community will (without central collusion, but merely through generalizing from past interactions) act the same way. As a result, people start to feel less welcome after leaving dumb comments & more welcome after leaving smart ones – and folks who aren’t comfortable with the weight of the expectation that their comments need to be high-quality are going to eventually leave. That’s the normal operation of decentralized social norm enforcement.

                  Centralized social norm enforcement (without technical or legal enforcement) is something like editing the posting guidelines / CoC. Now, if I recall, the posting guidelines for this site already recommend high-quality comments & suggest that off-topic threads & reddit-style in-joke circle-jerks are unwelcome – which, to me, implies an expectation that people will put in the minimum effort to become familiar enough with the subject of a thread to leave good comments. In other words, we already have a guideline against this behavior (although perhaps if this is a persistent problem, we should make it more specific).

                  I’m personally OK with criticizing low-quality comments as an enforcement mechanism – after all, such a criticism isn’t fundamentally misapplied (it’s not an attack on a person but a corrective for a specific behavior generally acknowledged to be undesirable) – but I would understand if people would rather avoid that in favor of down-voting & other less-direct mechanisms in order to avoid the kind of toxic environment that you tend to get when people treat criticisms as insults & correctives to bad behavior are allowed to escallate into fights. Nobody has the presence of mind to be able to avoid falling into that trap all the time, unfortunately, and avoiding direct criticism of commenter behavior or leaving it to the mods is one way to limit the damage from that.

                  1. 3

                    I wasn’t aware we (finally) had official posting guidelines… or even formal position statements from our moderators.

                    I’ve seen far too many communities devolve from lack of moderation, explicit guidelines, and population growth control. I mean, the Eternal September is linked in the About page.

                    And I don’t see how a thread discussing social norms is part of the centralised system. There have been plenty of meta threads wherein people propose technical and legal solutions; but, fewer where anything formal came about. More often, we see a discussion that helps us each refine our opinions on what good community etiquette means.

                    1. 3

                      I wasn’t aware we (finally) had official posting guidelines… or even formal position statements from our moderators.

                      There was enough resistance to the CoC proposal that we don’t have such guidelines. Here is @pushcx’s preference is for how Lobsters do discussions. In practice, I think the closest thing to a guideline we have is to avoid fights/trolling where possible, look for positive/charitable interpretation of peoples’ comments by default, stick to factual claims/arguments over rhetoric, and try to bring in deep content in stories or comments. Especially the latter even if we’re screwing up on anything else. The admin’s and moderators also usually let the community decide things with votes and flags. The votes push down and collapse, but don’t fully censor, the dissent in majority of cases. The flags are handled in a remedial way that tries to avoid deletions or bans wherever possible.

                      You’ve been here a while and probably know most of that, if not all. I’m just dropping the summary here as clarification for new people or more casual readers of the site who may get confused reading your first sentence. Of course, this is just my opinion from observing since signup how admins and moderators act on metas, technical discussions, and intense debates.

                      EDIT to add admin’s prior encouragement about good, discussion etiquette.

                      1. 1

                        I wasn’t aware we (finally) had official posting guidelines…

                        I had to click through some when I signed up for the site, but I don’t know how to get to them now. Maybe they’re linked from the ‘about’ page or something?

                        I don’t see how a thread discussing social norms is part of the centralised system.

                        The facility for even applying a meta tag was at one point created by a moderator for the purpose of identifying this type of discussion.

                        Social norms often go through their entire life cycle without anybody directly talking about them or consciously noticing that they exist, because they tend to perform their entire task while remaining vague, nebulous feelings about certain people or actions being gauche (without anybody actually identifying the boundaries or locus of that feeling). So, just discussing them directly is more of a ‘centralized’ behavior than how they are typically enforced – let alone talking about them in a place specifically designed for talking about them.

                        1. 1

                          Social norms often go through their entire life cycle without anybody directly talking about them or consciously noticing that they exist, because they tend to perform their entire task while remaining vague, nebulous feelings about certain people or actions being gauche (without anybody actually identifying the boundaries or locus of that feeling).

                          While that can be true, I’ve discovered— to my shock and horror— that some people, even in the small groups (3+), both explicitly and frequently discuss social norms and organise to change them. I think these kinds of conversations happening in private is a recipe for misunderstanding, and as such always push for them to happen in public.

                          1. 2

                            That’s a fair point. If we’re going to discuss norms with an aim toward norm engineering, we should do it in public.

                            I’m just thinking that low-stakes hard-to-prove faults ought to be enforced with soft mechanisms – in other words, if failure to abide by a norm can only reliably be determined subjectively (like “this person didn’t carefully read the article”) and the failure isn’t so catastrophic that a small number mean doom for the whole community (the way that violence might), then enforcement should be done in a way that requires the consensus of multiple people who aren’t communicating about the subject (so that one person, no matter their social capital & ability to influence others, can’t unilaterally enforce based on a potentially-flawed understanding of the situation).

                            In other words, while we talk about the problem & might clarify the guidelines in order to make it more likely that folks consider it when deciding whether or not to upvote a comment, we shouldn’t allow discussion of it to become the basis of legal or technical enforcement. When we need legal or technical enforcement for subjective things, we often start relying on juries or forcing judges/mods to uphold substantially higher standards of behavior, and that kind of thing takes the kind of effort few online communities can muster – while laying the responsibility that way without supplying the resources leads to a situation that’s just as broken as if no norm enforcement was happening at all.

                            (There’s also the issue that defining norms sometimes changes them to make them easier to define, resulting in unexpected changes in enforcement once some definition takes hold, but I don’t think that problem particularly applies in this situation.)

                            There’s a lot of stuff that gets posted here that’s very long (like, hundreds of pages sometimes) or very technical, or both – in other words, this isn’t a strict expectation that everybody will read the whole linked article before commenting, or else the meatiest stuff will have a weeks- or months-long delay before anybody says anything about it at all! And, even articles that are short can have their point missed if they use specialized terminology unfamiliar to the commenter, are poorly written, or if the commenter in question simply didn’t have their morning coffee yet or didn’t sleep well. On top of that, high-quality discussion can result from comments that are tangential to the article itself.

                            What we’re really trying to avoid is the normalization of low-quality repetitive discussions of hot-button issues prompted by tangential relation to an article. This is truly distinct from both reading the article & being on-topic, and I wouldn’t trust any single user to be able to identify it reliably.

                            (I have a dog in this fight. As somebody who fairly frequently posts xanadu-related content, I know from experience that every comment thread is full of people repeating half-remembered misconceptions about the project & other people drawing on misunderstandings of the article coming from an unfamiliarity with the terminology & rhetorical style – so, I’d absolutely like the situation to improve. I’m pretty sure no mod is going to do a great job separating the folks whose confusion is understandable from the folks whose confusion is incompatible with a good faith effort to have read the post in such a circumstance – and a random sample of users on this site couldn’t either, simply because misinformation is widespread & primary documents are arcane and fragmentary. There are probably other subjects that produce similar reactions & that I’m not familiar enough with to identify.)

                2. 4

                  I think this something we can solve though (very gentle) shaming. Whenever someone leaves a comment that shows they didn’t read the article, I try to leave a comment saying “that’s explicitly covered in the article.”

                  1. 1

                    That’s a really good way, I’ve seen some of your comments already, but I must admit I was too thick to also recognize the gentle hint.

                  2. 4

                    Architecture in cities influences how people make decisions[1]. Maybe we can add something to the website that influences people to actually read and digest the posts before commenting.

                    A not so intrusive way, which is still based on trust, can be to allow commenters to set certain flags like: FULLY_READ, SKIMMED, or BARELY.

                    A more intrusive way would be to even mark a comment, if the author didn’t open the link. This is obviously easily bypassed, and might lead to a lot of false negatives.

                    1. 2

                      The action of me opening a link via RSS in Thunderbird is entirely separated from visiting this site and deciding to leave a comment, unless it’s an ask thread. I can’t imagine any unobtrusive, working addition to that.

                      I’m going to borrow a reply from xvirus:

                      Complex nonsolutions to simple nonproblems.

                      1. 4

                        That was me just brainstorming to start a discussion, instead of just presenting an issue, and spending no time at all considering solutions. Neither my imagination nor yours is the limit.

                        As for the quote, if you think this a non-problem, then you’d do some of your readers a service by being less indecisive, less ambiguous, and saying that clearly, instead of forcing me to wonder or ask if this is a non-problem for you?

                        1. 2

                          Complex nonsolutions to simple nonproblems.

                          Very Unix haters. And yet, people continue to use it, despite the numerous alternatives.

                          1. 1

                            +1 to the point about reading articles somewhere else.

                            I don’t agree with “complex nonsolutions” part. The problem of commenters not reading the article is real. I find myself guilty from time to time and I’d like something to help me keep myself in check.

                        2. 4

                          Obvious technical solution with undoubtedly horrible side-effects:

                          • Add a “Please RTFA” down-vote option.
                          1. 2

                            A long-time Lobster made a comment that got more upvotes from the community than the article itself and any other comments (including yours). Your plan is to tell the same community to make comments like that disappear. “Not a great plan.”

                            Reminds me of the 2-3 people that downvote something as off-topic and tell people in comments it’s off-topic for this site after 20-30 people have upvoted it showing they want it here. Call it adapting our social norms or tyranny of the majority, it’s clear the overall community favors many different kinds of posts and comments even if small subsets of its don’t want all of them here. This sub-group wants this, that sub-group that, and so on. Trying to fight popular trends just adds more noise. I discourage those metas and comments. Also, that’s one, less, deep dive or something on our front page.

                            EDIT to add: I think downvote as troll covers this already. I still think it’s a waste of time for anything popular. Opponents can show their disapproval in a way that’s visible to rest of community, though.

                            1. 4

                              Reminds me of the 2-3 people that downvote something as off-topic and tell people in comments it’s off-topic for this site after 20-30 people have upvoted it showing they want it here.

                              My voting has certainly been swayed by a certain friendly Lobster explaining his downvotes.

                              1. 1

                                Thanks for pointing it out. :) I’m not saying nothing can happen. I’m saying it usually doesn’t happen. So, such actions will normally aim to influence a small number of people in context-specific ways. People who are also open to that influence. Whereas, I think this is aimed at a site-wide standard and/or enforcement about something which opposition was the default.

                                Hope that clarifies what I meant better.

                              2. 2

                                A long-time Lobster made a comment that got more upvotes from the community than the article itself and any other comments (including yours). Your plan is to tell the same community to make comments like that disappear. “Not a great plan.”

                                A factual correction: the comment that got most votes from the community was the one pointing out the behavior, and the only comment I had on the post was under it saying that I made this thread to discuss that exact behavior. So the majority in this case was at least agreeing that it exits (obviously that doesn’t mean they wanna discourage it).

                                My plan is also not to change it, but to maybe change it, if the community so desires. At no point did I try to enforce anything.

                                1. 1

                                  Fair enough.

                              3. 1

                                I complain about it because I find it exasperating, but unlike some sites it rarely swamps the entire conversation here.

                                1. 1

                                  There is something to be said for the comments at the bottom of an article over the comments on a site like Lobsters: you can be fairly sure that pretty much everyone there has actually read what they’re commenting on.

                                  1. 1

                                    Your use of the word “there” is unclear in this case. Is it referring to Lobste.rs or the online site?

                                    Do you mean that commenters under an online article are more likely to have read the content? Because that is not my experience at all.