Issue already exists: https://github.com/diaspora/diaspora/issues/2999
Is there any way to finally close a discussion on loomio?
I’d like to suggest to add the existing issue to the 0.0.3.0 milestone. Can any project admin on GitHub proceed or is this a stupid idea?
Why has it to be in 0.0.3.0?
No specific reason. I just thought it has been decided on so let’s proceed with action. Afaik, there are currently only two milestones on GH. 0.0.3.0 and 0.1.0.0. Feel free to create 0.0.4.0 if you think it’s better in that milestone.
I see nobody actively working on it so I’d prefer no milestone at all.
Yeah, we put a milestone when someone works on a pull request and we can estimate the time to do it. About like in comments, nobody has time to work on it for the moment…
It’s easy to decide that it would be good for something to be done. Actually finding a way to do it (and people with time and energy to do it) is not so easy. Likeable comments had to be removed because they were putting an enormous strain on the servers of big pods. Now this may be a fairly straightforward problem to solve, or it may not be. Until the reason for this problem has been identified, and one or more people have said they’ll work on it, it wouldn’t be appropriate to set a milestone, certainly not an imminent one.
As always, I’m sorry I can’t help with coding things.
Sorry for bumping this as per this new thread. Of course there always are “higher priorities”, but someone might be kind enough to update the Loomio thread with what’s happening on github. Someone was working on bringing the feature back, but the current status of the progress is unknown. Can someone ping the guy working on it on github? Thanks.
What is the status on this feature? It’s a feature request for 6 years now and I think it still it is a wanted feature. Are there any technical issues blocking this or is it “just” time?
The federation part is done and working, the backend is mostly done and it’s missing a frontend. So it’s “just time”.
At the end of the diaspora comments there are opinions posted that are against the “likes-on-comments” feature. See this comment and comments below: https://pod.interlin.nl/posts/71a750604c560137e949268acd52edbf#40632ce04cde0137957d005056264835
I just would like (no pun intended) to give my opinion on the decision:
@robb_nl, that’s an interesting contrast with your comment two previously in this discussion.
• having less than 20 voters
Everyone in diaspora* was given opportunity to comment and vote on this issue. There was a clear majority by those who did bother to vote. (I’ve got a feeling that the only ‘no’ vote was mine.) If every decision had to pass a quorum (and at what level would you set that quorum?) this project would be stuck in stasis even more than it has been.
• an option to edit posts and comments would be FAR more valuable.
You are doubtless right, but that is a complete irrelevance because implementing this feature in no way impinges on the implementation of the other.
• Please don’t call it a “like” button.
There’s a discussion to be had there about what it should be called, but I think that would be best kept separate from implementation of the feature, which was just about whether it was worth solving the database issues in order to extend an existing feature once again to cover comments. The naming issue applies to the same feature on posts as well as comments, and that isn’t covered by this discussion.
@goob, I can understand that my comments in this discussion might sound as a paradox. Let me try to explain:
IF a +1 feature is implemented as an option to just express that you (firmly) agree with a comment, then it can be of value and an asset for diaspora. This situation happened to me several times.
However, my fear is that the +1/like feature will end up as a way to accumulate as many +1’s/likes as possible and the sole reason for posting will be accumulating +1’s/likes. IMO this behavior is bad for discussion. In that latter case, I think it would be better not to have the feature.
At the none-distributed social networks such a feature is also (heavily) abused by the service itself for profiling users. This probably is the biggest horror of the feature… (although I don’t see this to be a problem for d* because of the distributed structure)
I agree with all of those points about the downsides of having such a feature in a social network. However, your view and mine have always been in a minority in diaspora*, so we’re not going to get traction for removal of this feature.
There has always been a ‘love’/‘like’ feature in diaspora*. Originally it was available on comments as well as posts. However, as the network and individual pods grew in size, this led to the databases of larger pods blowing up. ‘Likes’ on comments therefore had to be removed in order for the network to continue functioning, until such time as this technical database issue could be solved.
The vote above was merely a confirmatory vote that it was OK to restore likes to comments once this technical issue had been solved. It has now been solved, and it makes no sense to have likes on posts but not comments.
Your points are not about whether ‘likes’ should be re-enabled on comments as they are already enabled on posts; your point are about the merits or otherwise of having any sort of one-click response in diaspora*. That belongs in a separate discussion from this one, and is therefore off topic here.
There might well be some old discussions in Discourse that you can reanimate; I certainly remember discussing that issue a number of times, and some of those discussions were probably here.
12 yes-votes is the majority of people who cared enough about this feature to participate in this poll, which had a total of 15 votes. Would be be nice to have more engagement? Sometimes, absolutely. This Discourse has something around 1200 users who actually participated at least once, which is a very small number of diaspora users. But that’s just the way it is. We can’t force people to participate here, and some people simply don’t care.
Yes, and eating steak would be tastier than drinking water. Having another feature that feels more useful than likes on comments does not change the fact that likes for comments have already support in the federation layer, an established UX concept, as well as general consensus as no one complained about this feature when it was inside diaspora a few years ago.
That’s not even a discussion to have at this point. The primary interaction on posts is called “like” as well, and naming comment-likes a different verb would confuse everyone. We have had many discussions in the past about finding a better verb, but they all died, as it turned out that picking something more suitable that can also be translated in a lot of languages is actually an arduous task. If you have a revolutionizing idea, please find those discussions and resurrect one of them.
We already have likes for posts, and I seriously don’t get the impression that having these changes anything in terms of engagement. diaspora* does not incentivize receiving likes at all - it won’t make the post appear in more people’s streams, it won’t make a post stay there longer. It’s just that - people showing their appreciation.
In contrast, if we’d remove it entirely, I feel like either overall engagement drops significantly, or, even worse, people start posting “nice!” comments, which would make the comment section unreadable for everyone else.
I’ve seen this discussion evolve around the same “interacting with a single click reduces the number of actual comments on a post”, but I have never seen any evidence for that. If anything, those posts which receive a lot of likes also receive a lot of comments, sometimes even from the same people. Also, I’ve seen a lot of posts which don’t receive lots of likes, but a metric ton of comments. This position also seems to forget that sometimes, “I like this.” is simply the only reaction that people might have to a post. That could either be a beautiful photo, a nice story, or something else a user might appreciate without having further opinions or feedback on.
The number of likes is actually a handy tool all by itself. For the HQ account, for example, merely looking at the numbers can teach us that the “where is my feature?” post got more engagement than the “welcome” post. This could teach us to post more development insights and more repeated welcome content. The same is true for people posting comics or photos to diaspora, which can use the number of likes to gauge what kind of content their audience enjoys. That, paired with the fact that there is zero evidence for having a like button actually impacting the non-likes-engagement at all, leads me to the conclusion there is no reason to block someone from re-adding likes on comments - especially since we had them in the past, and only removed them because of a severe flaw in database design.
If we’re accepting that likes are a thing on diaspora* right now, and re-adding them to comments just provides more consistency, we could actually move past this topic and talk about likes in general, which we also had a brief chat on at the hackathon.
No matter how much we like or hate click-interactions, there is no arguing against the fact that they provide a way for users to express their opinions. Our job should not be to take expression away from users because “this is evil, you should write actual comments because everything else is meaningless”. We already limit user’s abilities greatly with the reasoning of privacy concerns, and while that may be fine, I see no base for us not to fulfill that requirement.
What we also can agree on is the fact that having a boolean like feature is not the best situation we could be in. Having only “like” is actually creating a lot of really tricky situations, like the “can I ‘like’ a post talking about sad news?”, for example. The obvious solution to that is to go the way that so many platforms these days do: by leveraging a more broad range of reactions. GitHub, for example, offers several emoji-based reactions:
Slack is taking this idea several steps further, and is simply offering the entire emoji range (plus custom emojis):
And I actually really like this. It makes a huge difference on a post if you react with 
, 
, 
, or 
. Having these kinds of interactions would actually get rid of the edge cases outlined earlier. Obviously, that’s not perfect either, as having a bunch of very colorful emojis in an otherwise pretty grayscale user interface will ruin everyone’s eyes and create lots of visual noise. Maybe there are solutions for that - like hiding away the emojis behind a “42 reactions” link and only showing the actual emojis when hovering over that, and/or by making emojis black-and-white per default. Maybe it’s a horrible idea. That’s a discussion to have for future iterations, but by no means is this blocking likes on comments in any way.
A acknowledge that the second half of this comment is kinda off-topic and more suitable in a general discussion about likes. If someone resurrects one of those, I’m happy to move the text around.
No, you didn’t even participate. 
A very rough mock-up of some possible approaches I uploaded when we were discussing this a few years ago:
Is it time to split the last few comments off into a new thread, or perhaps to one of the existing discussions about implementing emoji?
In a healthy situation, to have a valid decision making poll, there should be a minimum participation. A poll, that has been held 7!!! years ago with 15 votes can hardly be seen as a representation of the (current) diaspora community.
In a healthy situation, decisions are not based on polls that can be subject to multiple misleading factors (i.e., people not caring, or people suddenly caring because a particular peer group decided to raid a discussion, resulting in a large number of votes for one option that might not even represent the majority, but just the loudest user base). In a healthy situation, decisions are based on factual arguments, logical reasoning, and common sense.
Unsurprisingly, this was one of the reasons why we moved away from Loomio many moons ago and with that, stopped doing votes per se. We switched to a more discussion-oriented model where people can express their opinion if they can argue adequately (“I don’t like this because I don’t like this!” is not a proper argument). Funnily enough, this forces people to actually express their opinion, rather than just clicking a “like” button. But contrary to social network posts, “like” clicks on Loomio actual had an impact, and more often than not, decisions got made by people who managed to collect the largest amount of populists, rather than the people who had the best arguments.
In this discussion, the lack of factual arguments is almost laughable, and a bit disappointing.
On the one hand, we have the people who seem to be able to make logical arguments like
while the other side brings up pseudo-arguments like
and instead of finding proper arguments to support their opinion, that side just keeps on spamming the same old arguments all over the place, even in areas where it actually causes people trouble to do their work (like a pull request, for example), and creating new posts on diaspora attacking members of this projects individually for stating their opinion, and badmouthing a community members contributions to the project because you don’t like it. What an achievement.
If you have a proper argument against likes on comments that have not been disproven already (and remember that an argument contains claims, reasons, and evidence; even anecdotal evidence may be fine in some discussions, unless it’s already been proven not to be applicable), then please state your arguments so we can have a proper debate on those.
If you do not have appropriate arguments supporting your point, then please stop whatever you are attempting right now. Repeating the same points over and over again will get you on people’s blocklists, but that’s about it. You won’t change the outcome of this discussion and the fact that the pull request in question will be merged once the work is done if you’re stuck in an endless loop. This is valid for everyone, not just you, @robb_nl, but for all the other people reading this thread and complaining about how horrible this project and my personality in particular are. This project is lead by whoever has the most reasonable and most convincing arguments, not whoever can shout their opinion in the loudest voice.
If the technical resources had been available, comment likes would have been reintroduced 7 years ago.
And there has been no representation of current community opinion since then, until you and a couple of other people started voicing your opinions since the feature has been built. Which is a bit late.
I don’t like ‘likes’ or any one-click interactions, as I said before, and I wish diaspora* didn’t have any such thing. However, the weight of opinion has always been heavily in favour of them. And I simply cannot see how reenabling ‘likes’ on comments, which should always have been there and would have been had the database issues not occurred, will have any negative impact on interactions that ‘likes’ on posts, which have always been available, do not have.
Do you have any evidence that enabling likes on comments had a corrosive effect on conversations that likes on posts don’t have?
Not on diaspora. But on other social media platforms it has, massively.
Again, IF a like function on comments is solely to express: “I couldn’t have said it better and I agree” then it can be an asset for diaspora. My fear (and no I can not prove I am right with hard evidence) is that diaspora heads to the same like-hungry BS that is happening on Facebook, Instagram and the likes.
(I’ll use Facebook as an example; other commercially operated social networks are no doubt similar to a greater or lesser degree.)
Facebook’s entire business model is based on using the ‘like’ function to build of profiles of people who use the site that can be marketed to advertisers etc. It thus aims to make people click ‘like’ as many times as possible. To do this, it filters content to show you primarily the things it thinks you will click ‘like’ on. And in longer discussions, it shows the comments that have the most likes in preferences to the other comments in the discussion.
Its UX is therefore geared around using the ‘like’ function to encourage more use of the ‘like’ function. In order for people to get their content seen, they therefore have to tailor their content to what Facebook’s algorithms think will get lots of ‘likes’.
And so on.
None of the above is done by diaspora* software, nor will it be. Posts are not filtered by number of likes (in spite of demands for this by a number of users over the years), nor is content thought likely to get more ‘likes’ given preference. Indeed, there is no ranking at all of content.
There is therefore no incentive at all for people to tailor their content in any way in order for it to get greater exposure. The only way to do that is to use appropriate hashtags in public posts so that the post will be shown in more people’s streams. (Indeed, tag-spamming is a greater danger in diaspora* than ‘likes’ will ever be.)
While there might be an inbuilt bias towards writing content of a type that would get more exposure on Facebook in people who have recently arrived from that site, their content won’t get more exposure here and, if they decide to stay, it’s likely their content will change over time as a result.
Likewise, if someone has become used to ‘gaming’ the Facebook algorithms (actually, becoming a slave to them, because that’s exactly what Facebook wants) in order to get more attention for their comments in discussions, they will find that their comments actually get no more attention than anyone else’s. They are likely to either change their behaviour over time or to give up on diaspora* because their interactions aren’t getting the same kind of attention as they did on Facebook.
I don’t imagine that people who are used to using diaspora* rather than Facebook will suddenly starting commenting differently because there’s now the chance of someone pressing a ‘like’ button on their comment, because it will have zero impact on the interactions other than the knowledge that person X decided to press the ‘like’ button on that comment. There will be no other benefit to that action, so no incentive to write comments that are likely to get that reaction.
While ‘like’ is somewhat more nebulous that a straight ‘I agree’ (I’ve had people ‘like’ quite down-beat posts I’ve written, in, I can only assume, an attempt to sympathise when they didn’t know what to say in response) its presence in diaspora* has no effect other than to let the person you’re responding to know that you wanted to interact. There might be people who are ‘like-hungry’, and they can certainly collect ‘likes’ if they want, but there will be no impact other than their personal knowledge that they have X number of likes. They won’t be more prominent in the network. That’s done by having more contacts, and you’ll get contacts more by saying things that interest people.
As I said, I honestly can’t see why ‘likes’ being available on comments should have any impact different from ‘likes’ being available on posts. As you said, you have no evidence of a negative impact in diaspora*. Fine. Let’s see what actually happens when ‘likes’ are available once again on comments. I certainly don’t remember a negative impact when they were available previously. If it proves to have a strong negative impact on interactions, we can then revisit this and see whether this restored feature should be removed. Until then, we’re just musing on suppositions.
(This has become far longer than intended, so I’m going to post it ‘raw’, not proof-read.)
