Never in my life did I think people held such strong opinions on “reaction” buttons. I really don’t see how they could be such a negative influence on the platform? I’m not going to say they necessarily add anything either, just not sure why people are so worked up over it.
10 Likes
Reactions instead of POLLS?
It’s not about reactions, it’s about unnecessary change. This has happened before.
7 Likes
I just want to point out that your rhetorical suggestions are misleading as to the variety that could be offered without tapping into the negative reactions.
Here is a fairly limited selection of potential non-negative reactions.
I personally think that if you’re to justify the added layer of complexity of the new system then you ought to add a lot more than six new reactions covering a limited range of expression.
11 Likes
If Dark Souls and Elden Ring have taught me anything, it’s that the point down gesture can definitely be negative lol
12 Likes
Did you have napalm for breakfast?
7 Likes
The clear compromise is a palette of reaction kanji.
(I’m also in favor of the vanilla like system. )
5 Likes
I like this reply. I think I went too far in steelmanning seanblue’s argument and ending my discussion, but the inverse would’ve a been a deconstruction too condescending for me to be comfortable to post.
For me, the simple fact of the matter is that there are too many flaws in the Wanikani team’s initial implementation to come to any conclusions. There are too few ratings, their meanings overlap too much and they only cover a very small slice of what could be considered a complete rating system. The ratings themselves are also way too similar looking to an absurd degree, which I think is its biggest sin.
Here’s some rules that successful implementations I see follow. Mostly off the top of my head, so not comprehensive:
1: Distinct shapes: This means all emoji faces are off the table as well as hand emojis. You get one exception in that you can only use one. Use two and you have two ratings that look far too similar.
2: Distinct colors: This double means all emoji faces are off the table. No, seriously. Emoji faces and emoji hands make for terrible ratings. This also means no grayscaling of ratings under any circumstances.
3: Just the right amount: Usually not less than 10 but not more than 20. Having a single “like” is always too little. It’s just the simplest implementation. All it says is “I would like to bring attention to this post.” which is why social media companies love it, because advertisers love it, because it provides the absolute simplest form to disseminate users’ personal data. Inversely, having every emoji/icon available is too much. People won’t use them properly if there isn’t a distinct consensus, which leads to decision fatigue, which leads to disuse.
4: Names: Naming your ratings properly is important for users coming to a consensus on usage and encouraging them to use your ratings properly. Giving names that simply describe what the rating looks like leads to ambiguity and at worse can lead to a user believing a rating is useless because no connection can be made on its use.
5: Ratings, not reactions: Users should think about what quality a post has and not just what they emotionally feel when looking at a post. Thinking about it only in a reaction kind of way limits you simply to emotions, when so many posts are neutral in varied ways and not meant to elicit any strong emotional response. Thinking about it in this way helps lead to a full palette of ratings, if the color requirements didn’t make it a full palette already.
Combining these together and you get something like this:
Agree
Friendly
Cute
Optimistic
Sad
Artistic
Informative
Good Idea
Winner
This is all from a real forum. I’ve removed the negative ones and kept the ones that I think could be used in Wanikani in some way. Given the Wanikani team’s silly and artistic sensibilities, they could probably easily expand this list with useful and more specific ratings.
There are other things to be said here about presentation and UI interaction, but that’s very much on a per-forum basis as layouts and interaction aren’t nearly as universal. This is all going to be down to Discourse’s limits.
In retrospect, I think the only reason why I haven’t seen a bad implementation is because people figured this out years before I even joined the forumspace and haven’t been trying to reinvent the wheel since. Meanwhile, the Wanikani team was just… testing a feature without intentions to fully implement it. I can’t fault them for just trying things out, but I’m going to be harsh on the Wanikani team and say that there wasn’t an excuse to not make their ratings feature complete before letting the public test it. Now users are making extrapolations that only apply here because of their uniquely bad implementation. There’s an art to using multiple ratings that simply doesn’t apply to a single-rating system. Whether or not such sentiments can be reversed, only will and time will tell.
P.S.
Does discourse really not have a user friendly image embed button? I had to type out the raw BBCode for the images and that seems a little user unfriendly. Just an odd oversight.
11 Likes
Works the same in a Desktop browser, you just get a file picker instead of defaulting to your photos.
Haven’t had to do that since PHPBB. Not even sure I remember how. lol
6 Likes
I think you make some sensible points, but as for not “using them properly” or not “having a distinct consensus” those aren’t necessarily problems to me and I think maybe you’re not giving people enough credit. My guess is that over time long-running threads would establish their own consensus for the use of different symbols specific to the thread’s topics and ways of interacting. Here having more icons allows for more creativity and specificity in establishing visual norms and I think this could have a positive effect on the most active recurring threads without detracting from the rest of the forum. Some broader consensus might also be established organically, but I think having a “default” suite of named symbols (+ a choice of many more) could probably help provide guidance and coherence across the forum.
Edit: I also think that taking a top-down approach to “defining” the meaning of the sybols would limit the creativity with which they could be used. The could stand for optimism as you say, or for pride, or perhaps for something else, and by leaving the meaning open you’d let the users use the symbols in the way that is most meaningful to them. With the trade off that such meanings would require time to be established and an active set of users to maintain. (But the specific set of defined symbols you shared seems really well designed to be fair)
5 Likes
I would have reacted with to your post, if it was possible. Have a confetti ball instead.
7 Likes
I can’t really agree to this, unless I understand which side of the software are you on, and alternatives.
What I see is Discourse may suffer from aging… but somehow it survives (in popularity and features) so far.
5 Likes
You mean the design of the backend? If @Humin had the frontend design in mind, then I agree. I am not too fond of it either.
Of the newer (then, so like 5-8 years ago) Linux distributions which split off from Debian proper at some point used Discourse for their new forum and it was a complete nightmare to navigate through longer running threads, because specific information was more critical than what person A said and when. The page tab system makes a lot more sense for a forum and is easier to navigate both on mobile and on desktop.
Regarding post scoring - Reddit uses thumbs up and thumbs down. It just works.
6 Likes
Reddit sells Gifting stickers, though.
Also, I can’t just host Reddit anywhere.
4 Likes
That system heavily biases toward majority opinion and encourages short oneliners. I feel like it never really worked that well, and people still don’t know how to use downvotes properly. Especially on a informative discussion forum it doesn’t sound like a good fit.
I wouldn’t really categorize it as forum software, anyway.
13 Likes
I mean, this is even less sensible to me honestly. Being averse to change simply because it’s change, without any consideration for what that change actually entails. Not to imply that change for the sake of change is inherently good, but I don’t think it’s inherently bad either.
4 Likes
Being adverse to change as a gut reaction makes sense to me. If you’re prefectly happy with the way things are then a sudden change can either mean that your experience stays good, or that it gets worse.
For example: if change = 50% probability of “good” and 50% probaility of “bad”, whereas not-change = 100% probability of “good” then people who are already satisfied will have a natural bias agaist change. There is also the mental switching cost of getting used to a new system, which means that even if the new stuff is as good as the old stuff, the net result of the change is slitghtly worse than if you didn’t attempt it. So for change to be worthwhile there needs to be a decent probability of it turning out, not “good,” but “really good.”
I don’t think we’ve seen this, but people are reacting to a potential disruption without any suggestion that it would even offset the switching costs. And I don’t think voicing one’s opinion here is an indication of “strong opinions on reaction buttons” as you term it, sometimes talking about changes to the WK forum is just a welcome distraction from what you’re actually supposed to be focusing on (like my exams ).
9 Likes
This is a very clear statement that it’s about change in general.
This still implies that change must be inherently good or bad, and my point is that this change falls pretty squarely in true neutral for me. Who cares what the reactions are? The actual point of the forum is replies and discourse, like we’re having now.
3 Likes
I think with any change, the question is what does it add? And what is trying to be accomplished with the change? (Also @Beyond_Sleepy had a good point about the cost of switching.)
What does being able to clap at a post add that giving a like doesn’t? What does the laughing emoji add that a like doesn’t already convey? (I think most of us understand that liking a post that is essentially a joke, is appreciating/laughing at/with the joke.)
Does this just add complexity and UI clutter without actually adding a tangible benefit? That is my question. However, I can only answer it for myself.
Giving a like is easy. That is just noting whether a post added something to the discussion/thread. Whether the add was info, a joke/laugh, additional experience with X, etc.
But when I suddenly have a plethora of reactions then two different modes becomes available (for me and everyone else): those who engage with the extra reactions and perhaps puts additional meaning into them, and those that don’t (and then either just uses likes or leaves no reactions at all anymore).
More complexity comes with more choices which can be good, but it is also adds additional burden. Learning how a community uses the reactions, and also, what if someone uses the laugh emoji on a really sad post? Even without any “negative” reactions, with even these options you can create negative reactions by mismatching them with the content of a post.
Personally, I think only have likes are superior, because there are no unnecessary questions added (why did you give a rainbow when I talked about a sad manga? Why use a hug emoji when I’ve never interacted with you before (can feel intimate)), less/minimal potentiality for abuse, less ui, and less complexity for new users. (Most people already start “interacting” with a forum by lurking to figure out the culture, also adding reaction culture to it means an additional step to figure out.)
In conclusion, I’m against it because I don’t see what it adds that are worth any of the cost. More reactions just doesn’t do it for me. If I want to react more deeply, I use the reply function. I’m not opposed to changing my mind if there are benefits that I’m not seeing.
Even if it stays, I’m likely to just keep only using the like/heart one anyway, because the fewer micro-decisions I have to do in a day, the happier I am.
20 Likes
Change as a potential disruption. Change can be neutral if it doesn’t disrupt things, I’m just saying that it’s not strange for people to be wary of potential disruptions. In general.
What I meant to imply was that “the state of the forum” is inherently either good or bad (or I guess neutral) for the user, and a “change to the forum” has an inherent chance of affecting whether it’s good or bad. The change could turn out to be neutral, but we can’t always know that in advance which is why change generates discussion:)
4 Likes
My bad on assuming you were referring to the positivity when you called the reactions similar. I’ve participated in discussions on the Discourse Meta forums about this feature, and the lack of negative options just happened to be in the forefront of my mind because of that. Interestingly, none of the criteria you mentioned for what makes good options has come up over there, so I don’t think they’re evaluating this as deeply as you are or (and perhaps not as deeply as they should be).
Maybe I’m just resistant to change, but I honestly don’t see what any of those options add to the WaniKani forums. Perhaps they’d be fun in Campfire, but I mostly participate in the Japanese Language section section. Yes, I’m biased because of that, but since WaniKani is a Japanese learning platform I think that’s a reasonable bias to have. I don’t think (debatable) improvements to Campfire is worth making the core parts of the forum more confusing
I find this so interesting. Maybe I’m just using the wrong sites, but I’ve actually never seen something like this, except on Facebook. I’m excluding chat systems like Slack and Discord where you can react to messages with any emoji. I’ve literally never seen a forum with anything other than likes, upvotes, and downvotes.
Well, not a forum, but I guess GitHub has this for issues. It works reasonably well there, but that’s because downvotes and expressing dislike for something is important on an issue tracking site. So I don’t know. I’m still struggling to see how it would work well on WaniKani regardless of the options provided.
14 Likes