You are reading content from Scuttlebutt
@Dominic

like or not to like

forking from @jolyon's suggestion on the dislike thead change "like" to "sweet" because that thread was big enough with just discussing negative expressions.

Some history on the like/dig/yup button: originally @paul just chose "dig". It's 60s slang, actually it was pulled from jazz before that I think. Later I made patchbay, and I didn't really like Dig that much, so I made it "Yup". What I liked about Yup is that it literally means "yes", but can be taken as like or approve too. But I feel, often "like" "favorite" buttons are really used as just acknowledgements, it could just be letting them know that you had read that.

We also experimented with selectable "reactions" at some point, but I think paul removed it because it cluttered the interface.

Underneath, it was still a vote message: {type: 'vote', value: 1}. -1 would be a downvote, which is how flag messages worked. Since I changed the button, I also added a expression: 'yup' field, because i figured there should be some record of whether they clicked a different button.

@ev
felt it was confusing to have two different button labels, and made a PR to change it to "dig" in patchbay, which I reluctantly merged. We also changed the expression value.

Anyway, then at patchwork@3, @matt changed it to "like" to be more mainstream compatible. But, he didn't change the expression, so it still looks like "dig" from the database. (we should fix this, but have been too busy, wouldn't be a very hard PR, hint hint nudge nudge)

User has chosen not to be hosted publicly
@juul

The problem with both "like" and "sweet" is that you might think something is important or worthwhile but you don't want to say "oh that's sweet!" or "i like that" if something bad happens and someone is simply reporting on it.

What about "boost" as in "boost this post's visibility" or to stay with the nautical theme "hoist" or "bowse" or "buoy" or "float".

@juul

Or if it's never going to be used for deciding what content to show first maybe "boost" isn't the right kind of thing either.

Dig reminds me of digg.com

What about "vibe" as in vibing with something, or "twinkle" as in the occupy "twinkle fingers" hand-signal. Or just go with a symbol like heart or thumbs up.

@nanomonkey

I bellyfeel and goodthink this prolefeed.

@Dominic

@juul currently there is no "amplification" effect currently. It's not used to rank or bubble up anywhere. All that happens is some ticks are shown at the top of the message and if you hover over the ticks, who hit dug is shown. Also, the author of the message gets a notification, and even then, "notifications" are not very in your face. In all, it's a pretty subdued system currently. But I think I like that.

@Dominic

@juul I'd click "vibe" though!

@Kodo

Why don't we just have emoji reactions like slack? I might even get around to a PR one day :P

@Aljoscha

Please keep in mind that not everyone is a native English speaker. If I saw a "vibe" button, I'd simply be confused, whereas pretty much everyone with some knowledge of the language knows verbs like "like".

User has chosen not to be hosted publicly
@dangerousbeans

1be45038c55ada4354156accb5a7672a.jpg

@grammolan

With the prospect that the term must be translated into other languages and cultures, and each can be given subtly different meanings depending on the choice of words, we may have to take a step back and think about why we use and what we want to use this function for.

I've been thinking about why and why I'm currently using the Dig button:

  1. Signal to someone that I found his contribution very valuable, with the motivation to reward him for it and to motivate further contributions with similar quality.

  2. Recommend the post to others as worth reading. Because I know that the "digs" appear in my account timeline and can be viewed by interested people. So I can help them come across interesting content.

Unfortunately, there is still no function to view the "most digged" posts in the circle of friends (or is it?). This would have to make the latter functionality even more valuable.

@robin

Cross-posting from the original thread on downvoting, as it's relevant here too.

%lTV0F2i...

In short, I think we should entirely ditch like/dislike (or whatever) as it is far too simplistic and crams much human emotion into too simple a response.

@dangerousbeans

How about free-form reaction tagging, but with a bandwagon collection of existing reactions to jump onto? That way each thread could have it's own unique things going on, or each cluster of people

@robin

i agree, free-form tagging is much better as it encourages more creativity/thought, but it still encourages single word responses (think how tagging is generally used) and thus tends to compress complex ideas or responses into overly simplistic replies.

@mikey

++ free-form reactions

either as emojis (as we see in Slack, Facebook, GitHub, etc) or text (as in the original Patchwork back in the day). probably emojis, then we'll probably want blob urls as emojis. :smile:

also, and this is more for the negative reactions, i wonder if we might want use codes to determine what a given reaction means to the interface, like error codes.

@dangerousbeans

maybe we copy what the brain does and have a memory and emotion?

eg: appreciated with fear

@mikey
  1. what i'm saying
  2. what i mean
@Dominic

@dinosaur but if I knew what I meant I would have just said it!

User has chosen not to be hosted publicly
@tiago

free-form seems cool and future-proof but i feel it kinda deviates from what i see as the main point for these things: not cluttering the interface. think about an announcement that warrants thousands of different opinions, and the #guysThatTagEverything.

i think the ux should distinguish two different cases: the "(common) quick response" section separate from "(emotional) reaction" section.

quick responses:
apart from the non-cluttering, i think this is about the 1% internet behavior thing --not needing to write a short post, and encouraging a response that might not be made otherwise.
I suggest using predefined literals like "thanks" and "sorry".
not as convinced about the "+1" and "-1", but i think they'd be useful without being reddit-toxic, as there's no ranking involved and it perfectly fits the use case.

reactions:
+1 (face-only) emojis. if it's about conveying emotion, there's a face for that. Added advantage that it looks very distinct from a quick-response.
I think emotional contagion effects should be optional, allowing reactions to public posts to be private.

@Zach!

I've been thinking about why and why I'm currently using the Dig button:

Signal to someone that I found his contribution very valuable, with the motivation to reward him for it and to motivate further contributions with similar quality.

Recommend the post to others as worth reading. Because I know that the "digs" appear in my account timeline and can be viewed by interested people. So I can help them come across interesting content.

@Lenny Abramov

May I propose something unorthodox?

Ditch the "like" button altogether.

I've been thinking about this for years, and I believe it's the right thing to do. I think pre-canned interactions should be reserved for emergency cases, like spam, or to very special ones, like polls.

Downsides of "like" or similar buttons:

  • They constrain communication.

  • Constrained messages can be easily aggregated (e.g. "like" count). Doing that implicitly attributes more value to the "most popular" posts. (That's good for special purpose platforms, like Stackoverflow where the solution that worked for most people will probably work for you, but not in general).

  • Showing the "like" count is bound to create a positive feedback loop.

Upsides of a "like"-free social network:

  • People will be more active. The 1% rule is not a law of nature, it's a consequence of how online interaction was designed.

  • People will be less likely to develop a "feedback anxiety". They won't post to be "liked", they'll post to give a real contribution, and be replied.

  • Different networks (because on SSB there are many social networks, right?) will develop their own social rules, e.g. always reply to "good" posts with an approving message and never to bad ones, or conversely always stigmatize the "bad" ones but never approve the good ones, or just reply when there's something substantial to contribute, etc.. Although I'd surely prefer a kiwi-cultured social network ( %huPiuUK... ) to Facebook, I think culture should emerge from society and not be spoon-fed by the app.

@Zach!

hahaha, i hit publish to doublecheck how the quote syntax works, realized I did it wrong, and accidentally published anyway. But I think checking the intent or value of a like/vote/vibe button, as @grammolan did, is useful.

This is my personal take!

For the first part (show acknowledgement/agreement), I do find the button useful for this but honestly, I appreciate seeing likes simply because I know that my post was read by other people. I mean this in a practical and not emotional sense. I am still learning how channels work, and visibility within this entire subjective social graph. I will post a message and wonder if it is essentially private, due to a channel I wrote it in or way that I wrote it. Seeing likes on a post assuages that doubt.

From this angle, likes is is a useful feature because of the relative infancy of scuttle, and my own personal lack of understanding of the architecture. I am not sure how useful they'd be over time as both the scuttleverse and my knowledge grows.

The second point (Recommend the post to others as worth reading) could be super interesting, especially if we had emoji reactions. Likes don't "boost" a post, so we aren't recommending a post to others by making it more prominent in their feed. Rather, the likes are added to your log. When I check out a person's profile I could see all the things they've written, the people they follow, and the posts they've liked. In this way, the likes could act as a curated list of articles that you have some emotional connection to.

Now imagine that these "curated lists" were more easily scannable and there was a matrix of emotions you could add to a post. Then my profile would show the stuff I've written, the friends I have, plus the things that I find sweet, the things that break my heart, stuff that shocks or surprises me, and all the things that made me go :aerial_tramway: (and these I don't necessarily know how to define except as :aerial_tramway: , but there's some sort of pattern when you see them all collected together). Also, this curation is largely subconscious. I read a post, i give a reaction, but over time the collection of these create some whole other meta-meaning.

Or maybe there's some whole other function we'd want for this type of feature that we have not yet defined, which is even cooler!

@tiago

@Lenny Abramov I liked your post, as in I appreciated the argument and it made me smile, but i don't totally agree.
However, i don't feel enough energy right now to explain why i don't exactly agree, particularly this fairly fundamental part:

it's a consequence of how online interaction was designed.

So for now, I'll leave it at that, but I suspect it looks bad to leave my reply like this. (Sorry!)

I'd prefer to be able to use a smile, and explicitly not the (hypothetical) +1.
I'd be happier with such less impactful hint, and go about my chores.

It's also both a little annoying and amusing that this very post fuels your argument.
Not my general perspective though.

@nanomonkey

I think a combination of the emoji response line that slack does and the edit history icon line that the smallest federated wiki would be slick. Then you could quickly see that a post has been edited and responded to by others with minimal intrusiveness. Similar responses could be combined into one icon with a mouse over for each icon that gives you more information on the edit date or how many people :poop:'ed a response.

Screen Shot 2017-05-13 at 8.32.33 AM.png

@joeyh

Share might be a more fundamental action than Like on ssb. Sharing a post could let your friends see content that doesn't normally propagate to them at all, so it has actual value as opposed to incrementing a counter.

@alex

I actually agree with the option of removing the like button altogether.

For the "like button feature" to be kept it would need to be seriously rethought.
I just posted %tBI9+wB... a few moments ago

@Zach!

These threads on the like button and the dislike/mute functionality make me happy, and for this reason:

I often think about the creation of the GUI at Xerox PARC, and what happens when you design for a specific audience. We had this awesome and confusing new tool of the computer, but you operated it through terse, arcane commands. But there was a demand from businesses for the computers' power. And large businesses were the ones who could afford computers. And so you translate this tremendous, new, expanding power into metaphors understandable to business and office work. Thus we get desktops, and files that belong in folders, and documents that look and act as paper that can later be thrown into the trash bin.

But what if computers weren't needed by businesses first, and instead there was a tremendous need from the forestry and ecological services? And so instead of a desk, the initial metaphor for personal use was a tree: you have various branches and leaves of the branch that you tend or graft onto other branches. And if a leaf wasn't needed anymore you could drop it on the forest floor to be turned into mulch.

If that was the story we started with, where would computers be today, how would they have expanded and mutated the metaphor? Because now we have an innate understanding of desktops and files, even though we live in an increasingly nomadic and paperless world. The metaphor has detached from its origin, and now only describes itself. But this means there is a common feeling that computers couldn't exist any other way, that a file and folder structure was the natural, inevitable approach.

I fell in love with the scuttleverse because of the people who are inhabiting it. Regardless of UX/UI, I continually come back because here I found people discussing practical ways of building their own airships, and what life is like doing guerilla gardening in Berlin or living in a self-reliant shack on top of a lava flow. There's a distinct social anarchist bent to the discussion, and folks are not only discussing alternate societies at length, but also have the skills to realize them.

We are the audience of the scuttleverse, since we build these interfaces and protocols around us. So I am intently interested in the metaphors we use. How do we translate the ideal, expansive power of SSB into words and metaphors we understand? And can do this using new metaphors, ones that do not originate from a social media need or a capitalist office space?

@cel

function to view the "most digged" posts in the circle of friends

i experimented with making some UI for this in #patchfoo: %xfm0ViS...

@devsort

+1 for slack like emojis instead of "like"

@Kodo

I think you mean :thumbsup:

@keks

Another use-case: I sometimes use them as bookmarks for myself. There is no UI for that yet (except manually querying on the command line), but at least the data is out.

@interfect

I use Patchwork and I want digs back. I really liked the rainbow fingers.

@dangerousbeans

without thinking I liked @keks's post to show agreement because I use the likes as bookmarks too

@robin

Expanding on my point about making "block" the last, nuclear option, after all else has been tried, i have a further suggestion. At the moment, blocking an identity means I can't see what they say. Perhaps we should make the block two-way, so they also can't see what I write. If a person is that toxic that we resort to blocking them, perhaps we should completely exclude them from our part of the network.

@Vendan

Personally, I've seen the transition of a community from '+1' posts to emoji feedback(github), and it was a very very good thing. I see no advantage of ditching the idea of a simple feedback response, as requiring responses to be full additional posts clutters the display (lots of +1, "Cool", "nifty" posts), and results in, if anything, less feedback ("meh, I've got nothing real to add, so I just won't post" vs. "I liked this")

@dangerousbeans

What about if we treat each post like a lifeform which you can feed so it becomes stronger

@interfect

Maybe we should have the different vote types announce associated emojis (or just be :emoji names:), and your software could pick up on the ones in use on the network and present the most common ones to you, or you could enter your own.

@Dominic

@tiago

I think emotional contagion effects should be optional, allowing reactions to public posts to be private.

we could do this! you could encrypt your acknowledgement so only the author of that message knew you felt that.

It's really interesting to see the diversity of opinions on this simple thing, ranging from removing it all together as @Lenny Abramov
and @serapath to using it sincerely, as I feel @Zach! expresses:

I appreciate seeing likes simply because I know that my post was read by other people.

Personally, I'm with Zach here. I often use the button just to let someone know I have read it. It's similar to nodding when someone talks. I also use them in private messages when I don't need to add any more content, but want them to know I did read it. It's funny, because now that pw is my primary means of communication, I feel my self wanting a "nod" button in other places, like twitter DMs or emails.

I can also symphasize with the toxic-like-syndrome. I disabled them on twitter (by switching to tweetdeck, where they can be disabled) but I don't feel this way about patchwork. It would be easy to disable likes, for filter likes so you only see from people you directly follow.

There is obviously a wide range of ways people want to express themselves.
It there a common ground?

You don't have to use the button, but a significant proportion of the total ssb messages created are created with this button. (trying to avoid using the word ;)

@tiago

so just wandering the problem-space.

there's a threshold from where you chose to click a button vs actually writing something.
that threshold might not only be about effort/time. sometimes i have stuff to say, but the relative screen space i'll use doesn't feel proper and i'll settle for a "reaction".
say instead of a simple remark, or adding another point to an argument.

this dichotomy doesn't feel ideal to me. at least in order to facilitate the kind of discussion/brainstorms this particular community leans to.
maybe an annotate inspired feature e.g. https://genius.com/4232847 could turn these ux decisions into more of a gradient. it could also scratch much of the lack-of-edit itch.

@cel

i've sometimes thought "ack" could be a good dig/vote expression

@Soggypretzels

I think what @tiago said about wanting to respond but not having anything you deem worthy of the screen real-estate is key for me. For that reason I don't find a lot of personal value in adding that ability to have highly custom reactions as I would mostly use the most standard reaction to minimize the number of different reactions a post has.

I feel if I wasn't original enough to reply with my own thoughts, why would I have something interesting to say with the emoji I pick.

I think we should also look at what features might be enabled (or prevented) down the road based on the reaction system. I think the reaction system could be a powerful tool in enabling content discovery and content burying. I wonder if allowing a lot of flexibility in reactions now might make it harder to leverage the system later. If voting is being giving more responsibility later, will people both vote and react?

@robin

@/%40AsA9DgDDdO24c4FFIbuE9LB3b1xFTzHVn6AKArWnS%2Bo%3D.ed25519

May I propose something unorthodox?

this is not unorthodox, it's wonderful. it encourages a qualitative, not quantitative approach. i agree with everything you said, thank you for putting in words what i couldn't (oh, the irony!)

@robin

@Zach_

But what if computers weren't needed by businesses first, and
instead there was a tremendous need from the forestry and
ecological services?

stunning, absolutely stunning. well said.

@Dominic

Thought Experiment: What if we implemented emoji reactions for the people who wanted one, but everyone else switched it off, so they just got a "reaction counter". What would happen?

@Dominic

@Soggypretzels it's interesting that several people have mentioned amplification or discovery. I feel this may be the root of some of the more toxic feedback in centralized systems, but I think if it edged towards curation - adding useful information, rather than virality - it could be great.

@robin

emoji still feels very limiting (they are after all fixed by unicode or whatever), compared to the endless wonder of a generative language (like english or most other spoken/written languages).

@Greg K Nicholson

I agree. I don't understand conceptually what the difference is between clicking “Like” and writing a short reply that adds nothing insightful.

Practically though, I think a short reply without insight is actually better than a “Like” button, because it's more human and less cold to just say “Thanks!”. You get to choose your nuance of approval. So let's optimise for thoughtful discourse and meaningful interaction.

If a UI really wants to count something, it could count the number of (direct and indirect) replies, which seems to me to be a reasonable heuristic for “this post was interesting and/or useful”.

@Lenny Abramov

Current uses of the "like" button that could be implemented otherwise

It's clear that the "like" button is being used for several different purposes. I think the current mechanism is not generic enough to fit them all. Below are some humble proposals on how to address some of those needs.

Recommend to others

(This was raised by @Zach!). We can devise a separate mechanism for that. Recommendations should be shown on your profile, not on the post you're recommending: if someone is already seeing the post, what's the point in recommending it? (The advantage of attracting the attention of someone skimming through posts must be weighed against the danger of triggering a positive feedback loop).

Remember, bookmark

As suggested by @grammolan, there's currently no "bookmark" feature. I agree we need one, but it should be a separate mechanism. Bookmarks are for yourself, so they should be private. For instance I may bookmark an offensive post not because I want others to see it but because I want to reply properly another day.

Don't waste screen real-estate

@tiago and others pointed out that a post as short as a single :laughing: would take up a lot of space for so little information. This could be addressed by a UI that compresses very short messages and stacks them side by side. Also relevant for this issue is @tiago's proposal of

allowing reactions to public posts to be private.

The author of a funny joke would see a grid of laughing emojis and LOLs, while everybody else would just see the joke and the public replies to it.

This could be extended to longer replies. Sometimes I want to give feedback that's relevant for the author, but not so much for other people. Or maybe I just want to make sure I understood the post properly, so that I can give a public reply that makes sense. Of course I can already send private messages, but it would be nicer to keep the context of the discussion.

[BTW, @tiago, thanks for the "partial" reply of yesterday, that's so much better than a "like"! And I think I understood your position.]

Nod, acknoledge, communicate you've read the post

This could be addressed in the same way as the previous point. A grid of "ack", "read", "seen", :ok_hand:... would be quick to scan. In any case, acknoledgements are only relevant for the author, so they should be private.

Proposed features that are too special-purpose

@Vendan:

I've seen the transition of a community from '+1' posts to emoji feedback(github), and it was a very very good thing

I totally agree, but Github is an example of a special-purpose social network: it's for making software, thus often you need to upvote/downvote proposals. However SSB is (primarily) general-purpose. We can't just assume everything can be voted. That might fit with some twisted notion of democracy, but it's hardly justifiable in a rational way.

What's going to happen when you allow to vote, say, a selfie? What used to be a dystopia has unfortunately become a tragic reality. People literally die for this sort of stuff. I know I sound melodramatic, but we should carefully consider these choices, and ask ourselves questions like: Are we willing to trade vulnerable kids' lives for an uncluttered UI?

@tiago:

this dichotomy [click a button vs actually writing something] doesn't feel ideal to me. at least in order to facilitate the kind of discussion/brainstorms this particular community leans to. maybe an annotate inspired feature e.g. https://genius.com/4232847 could turn these ux decisions into more of a gradient.

Annotations are neat, but again, this proposal is admittedly aimed at a special purpose ("the kind of discussion/brainstorms this particular community leans to").

This issue can't be addressed in isolation

I'm really enjoying this discussion, and I think it's very useful and even necessary. I obviously resonate more with some positions than others, but it's so interesting to see all those different opinions and proposals emerge! That said, this feature can't be isolated from all the others. As much as we try to limit the scope of the discussion (this thread originated from a thread on "dislike"), the question whether or how to implement constrained feedback mechanisms depends on the whole design of the UX. I'm now convinced that simply removing the "like" button wouldn't improve things, unless we replace it with a consistent set of features that serve the purposes discussed above, and probably more.

All this means that we can discuss features one at a time, but we can't really decide on them in isolation. In addition to this sort of discussions, I believe we also need to discuss at a higher level, one that integrates the separate evalutions into consistent designs. I'm talking about a consistent project, written by one person, that is founded upon some assumptions about reality, that follows some guiding principles, and defines the UX accordingly. I'm not proposing that SSB be designed this way, only that we invest some time in detailing several such "ideal" social networks. It would be a useful exercise, that would provide a different kind of insights compared to discussions on specific issues.

Since I raised the issue, I promise I'll write down my own ideal social network design (and probably post it on Medium or something like that, where I can edit it). I can't promise it'll be good, but at least it'll be consistent.

@Howard Klein

@jolyon This is really a tangent but I find it kind of mysterious to conflate dissent with things like people being assholes online. :p I think of muting someone like walking away from a guy at a bar who is trying to cause trouble. It's a reversible action that minimally reacts to unpleasant social situations without increasing tension in the way starting an argument would.

@Howard Klein

Great post @Lenny Abramov!

I just noticed that they way Patchwork displays replies (every reply is in order and has equal graphical prominence) makes it less appealing to post a quick little reply to someone. Maybe it would be better to have different treatment for different types of replies. Further toward the branches of the "comment tree" replies might be displayed inset and relatively smaller compared to their parent comments.

@Soggypretzels

@Dominic I think a key might be discovery of content that is particularly popular with people who you are 'closer' to, rather than just finding all the content that is popular.

I don't necessarily want to see a post that has thousands of likes, but if 3 of my 4 most-interacted-with friends all like a post, I would want to see that.

If a post is popular across the network but also popular with the people who's opinions I hold in high regard I would like that post brought to my attention. Not all that is popular is bad and it should be up to each user to surround themselves with people who post and interact with things they are interested in.

@Dominic

I also think it's far more important what the button does than what it says.
For example, on twitter, the "like" button was labled "favorite", which suggests a behaviour akin to a bookmark. But, because the effect was to notify the author of the message it serves as an acknowledgement. And because it also puts your avatar on the message, it serves as an endorsement.

You could remove the name, and just use a meaningless symbol, say, a banana and if it had the intended effect, people would figure that out and use it that way.

So if you wanted a bookmark it should remember that for yourself, but not show an effect on the message, or notify the author.

User has chosen not to be hosted publicly
@robin

I think a short reply without insight is actually better than a “Like” button,
because it's more human and less cold to just say “Thanks!”. You get to
choose your nuance of approval. So let's optimise for thoughtful discourse
and meaningful interaction.

exactly!

@Vendan

I totally agree, but Github is an example of a special-purpose social network: it’s for making software, thus often you need to upvote/downvote proposals. However SSB is (primarily) general-purpose. We can’t just assume everything can be voted. That might fit with some twisted notion of democracy, but it’s hardly justifiable in a rational way.
What’s going to happen when you allow to vote, say, a selfie? What used to be a dystopia has unfortunately become a tragic reality. People literally die for this sort of stuff. I know I sound melodramatic, but we should carefully consider these choices, and ask ourselves questions like: Are we willing to trade vulnerable kids’ lives for an uncluttered UI?

But blocking support for custom/emoti voting does nothing to make this harder. All it means is people post hate messages as replies rather then votes. Trying to technologically censor (and that's what you are doing, essentially) just by making an action slightly harder is pointless. Attack the root cause, not the symptoms! Implement a system for "flagging"(downvoting, essentially) hateful feeds so people can just have their client ignore them or even just not replicate them at all! Educate people that follow and include offensive feeds! If you don't want to be surrounded by toxic sludge, get out of the waste reclamation pool!

And I know this could be seen as victim bashing, but really... in real life, we tell domestic abuse victims to "Get away from the abuser, get out of the situation". Why treat interactions on the internet so much differently? This isn't facebook or twitter, where you get opened up to everyone by default. You choose to follow who you follow, you choose to download the feeds you do. Don't let the bullies stay and just try to control them and reform them. Punish them. Kick them out of your feeds, shun them from your communities, expose their actions to all that know them. (After all, you now have signed proof they posted the message)

@robin

@Lenny_Abramov
Excellent ideas, i agree with all, but want to add something to this:

Github is an example of a special-purpose social network

yes, it is special purpose. however, the software is still an artefact of socialising. in all social interactions, by definition the social must come first. the software is a product of the interaction of humans. we are humans before (and after!) we are coders. the social interaction always in always is superior to what it produces. in a more concrete sense, even though the folks on github code, they also communicate in human/humans ways

@robin

@Vendan

Trying to technologically censor (and that's what you are doing, essentially)
just by making an action slightly harder is pointless. Attack the root cause,
not the symptoms! Implement a system for "flagging"(downvoting, essentially)
hateful feeds so people can just have their client ignore them or even just not
replicate them at all! Educate people that follow and include offensive feeds!
If you don't want to be surrounded by toxic sludge, get out of the waste
reclamation pool!

blocking feeds is good. the problem is how and why and when we do it. not technologically, but when/why we decide to do it. the problem with dislike/flag/downvote, etc. is it encourages mob mentality. no thought, just click downvote yourself. if we encourage discussion over an allegedly bad situation (remembering that not all bad behaviour is as obvious as some), then we maybe come to a better solution that merely blocking someone who is a bit clumsy, rather than a misogynist, homophobic, racist, etc.

@robin

@Vendan

it encourages mob mentality. no thought, just click downvote yourself

damn, missed a bit. it does this, because there is no subtlety, no intent, no reasoning. it's "block" or "don't block", nothing to explain why, or to explore other options.

@robin

@Lenny_Abramov

founded upon some assumptions about reality

this is problematic, particularly when one person is trying to design the thing. we're in an age where reality is not fixed. premodernity it was, nature was everything. given the plasticity of culture now, firstly reality is a moving target, secondly it's shaped by the multiply recursive feedback loops between many, many individuals, and thirdly the very act of trying to design ssb or its interface effects the reality we are trying to measure. heisenberg discovered this a century ago (it's impossible to measure the location of a particle without changing its speed, and vice versa [1]), but at the same time and independently, Lacan (and later Derrida) did work in pscyhoanalysis and linguistics to show the same was true with culture [2]. as soon as we try to measure/nail down/label culture, so we inherently change it. the consequence of this is that we must involve as many as possible (and make it infinitely modifiable), in order to do anything meaningful and useful. this is the beauty of the GPL of course, it renders a finality impossible.

[1] https://en.wikipedia.org/wiki/Heisenberg's_uncertainty_principle
[2] http://repository.law.miami.edu/cgi/viewcontent.cgi?article=1339&context=umlr
http://www.signosemio.com/derrida/deconstruction-and-differance.asp
http://www.informationphilosopher.com/solutions/philosophers/derrida/

User has chosen not to be hosted publicly
@Vendan

Well, 2 things:

1) I'd be perfectly fine with "block with reason" or something. Heck, just flagging specific messages could act as a "if I've flagged more then I've liked, ding that feed for various things, 2x more then I've liked, nuke it", and so the message you are flagging could act as the reason as to why you are dinging that feed.

2) I'm also meaning this as a "I should be able to nuke a feed, period". I'm going to be building this into my client. If I don't like a feed, I should have the means and ability to remove it from my computer and my FOAF display and such. While this may seem more "safe space" stupidity, this is more along the lines of "My computer == my house, I get to control what comes and goes". If someone else sees my "I've blocked this feed" and decides to do likewise, that's on them.

User has chosen not to be hosted publicly
@robin

My computer == my house, I get to control what comes and goes

but it's not. you are a part of a network, a network which is built through the mutual interaction of multiple persons and is greater because of that. the idea in a networked world that everyone has their own property is false. the whole point of building a decentralised network is that it blows apart the idea of property, of hoarding, of selfishness, of exclusion. if we design correctly, we encourage sharing, reciprocity and mutual benefits. that doesn't work as soon as someone says "no, this is mine. get lost". there are echoes of game theory in here, under that nonsense it's always in everyone's best interest to "fuck their buddy" (yeah, bad choice of words), rather than cooperate. Adam Curtis made a wonderful documentary about this, includng interviewing john nash, who designed game theory in the 60s at the Rand corporation. even then, in the era of property, it was problematic (apparently, when the secretaries took part in nash's trials, they all opted to share/cooperate....). in an era of shared production (facebook, google, the majority of work places), the suggestion of cutting yourself off from everyone else is counter-intuitive.

my suggestion isn't that we shouldn't exclude bad behaviour, it's that we should do it with something reasoned, discussed, rather than through something as trivial and overly-simplistic as clicking a button which has the power to exclude. where's the subtlety, the nuance in it? where's the humanity?

@robin

@Vendan
more correctly, yes the (physical) computer is yours (although there is a good argument against this, but we'll put that to one side for the moment), but the network belongs (if we can use such a word when we are discussing the commons) to everyone who takes part in it. even the bits that reside on your computer are important to others - we store the data of others, we share it with peers, we collaborate endlessly. how can one person take a piece and say "this is mine", if that is the case?

@angelo

@robin I had been operating under the idea that SSB is a bit more like the real world -- when you don't like someone you generally try and remove yourself from their presence. You may have friends that keep dragging them in, but you, yourself, can generally exclude them. You can even avoid them entirely if you'd like.

In my understanding, SSB operates similarly. With the FOAF model a toxic/unwanted person would basically be "cast out" from part of the network graph because people in that part of the graph would unfollow them and they'd drop out of FOAF. This doesn't prevent them from continuing to be a part of other areas of the network graph.

If someone is so bad that I want to exclude them entirely from me, my computer, and my little node on the network, so be it. That doesn't exclude them from any other area of the network. But that's me saying I don't want them in my corner and I don't want to assist them in spreading further.

Any one person's node isn't able to completely shut someone down or cast them out entirely. That's the wonders of the decentralized network. There's no central authority. BUT, I am allowed to be free of people I don't want to associate with. And that's powerful.

So, at the end of the day, I like @Vendan's idea to be able to completely "nuke" a feed.

@robin

So, at the end of the day, I like @Vendan's idea to be able to completely "nuke" a feed.

again: I'M NOT ARGUING FOR NOT BEING ABLE TO BLOCK A PERSON. I've said this many times, I'm not sure why so many people keep bringing up that I am! read my posts properly, i have not suggested the thing you are bringing up!

my point (explaining it again!) is the process we use to block a person. the process @vendan and others propose it too simplistic/clumsy. i am arguing for something more sophisticated, because we as humans are sophisticated, intelligent beings who can make smart choices, better than the unsubtle "in or out".

@Dominic

@robin thanks for the clarification - would you make a proposal for how you think blocking should work? (prehaps in another thread?)

@robin

I had been operating under the idea that SSB is a bit more like the real
world -- when you don't like someone you generally try and remove
yourself from their presence

also, if that person is already part of your social networks (friends, colleagues. family, etc.), would you really just walk away? i would imagine you'd talk to your other friends first, find out if the behaviour is a one off. perhaps get another friend to talk and clear things up, perhaps it was a misunderstanding. imagining all bad interactions are "me in a bar, talking to some random, abusive drunk guy" is a stretch, although that case may happen.

@Vendan

It's a gossip network. It works by people repeating what they have been told. If you are trying to say that the good of the network requires that I not control my devices, then we have 2 very different value systems. And frankly, you are going to lose. You can't force me to repeat every feed, cause SSB is an open protocol. You can't hardcode it into the client, cause I wrote my own. You can't make it a requirement of the network, cause I'll just fork it. No matter what you do, you can't require me to interpret the network the same way you do, cause that's the entire point of the network, it's one of the fundamental rights. My feed is my property, my rendering is my property, my very source code to my client is my property. I've opened them up and made them available to everyone, but that does not change the fact that I still own them.

You've said over and over that you aren't against blocking people, that you are just against the process I propose, but you don't seem to have read my posts. I'm advocating for actually putting a reason on the blocking, for involving the community in doing so, for attempting, as far as possible, to educate and reform. But I'm also stating that yes, we need a nuke, cause some people are abusive, and we WILL see child porn and other such horrors at some point. I'm building a "nuke it from low earth orbit" feature cause some things truly do need to be wiped off of my computer, if for no other reason then that it's legally, ethically, and morally dangerous and horrifying to keep them.

@robin

@Dominic
done already (in the grandfather thread) reposted here, in a new thread.

%UmAxICI...

@robin

@Vendan

If you are trying to say that the good of the network requires that I not
control my devices, then we have 2 very different value systems. And
frankly, you are going to lose. You can't force me to repeat every feed,
cause SSB is an open protocol.

no, i said none of that.

that you suggest win/lose says a lot. this is not a competition!

@robin

I guess the bigger point is that this is a social problem. we should think of how to resolve the social problem, before we implement it in technology. if we try to fix the social problem with technology, we are skipping past an important step.

@Vendan

Life itself is a competition, just one that humans have gotten very very very good at winning. SSB is, at it's core, a competition, and makes that open and free for anyone to participate in. I'm keeping likes in my client, I'll continue to click the button, and put them in my feed and render them. You can strip them out of your display and ignore them completely, and now we're "competing" to see which choice gains consensus. Do more people continue to use votes? Do more people drop them? That's a contest, though a very low key and "calm" one.

And yes, this is all a very social problem, as I've said from the start.

@Dominic

@Vendan @robin we already have arbitration. At least, that is what I've been trying to do, at the cultural level, not at a technical level. I don't really know how to get people to see things from each others perspectives, I mean, you certainly can't make them. I guess they need to agree to arbitration? Should arbitration have a technical manifestitation? what would it look like? But we have private messages with multiple recipients, so if your friends are having a fight, you can say something to them.

But so far, the couple of trolls that have turned up, have generally left on their own, after a bit of, what you might call, friendly arbitration.

@Vendan is right, he's writing his own version of ssb from scratch. If anyone can reinterpret ssb, it's him. There is only fairly low level requirements to the protocol, and a lot of "behaviour" on top of that which can be changed without breaking the network. In bitcoin world they call it a soft-fork.

The respect in which it is not a contest though, is that it's not necessary for any particular interpretation to "win". Already, we have multiple interpretations (via multiple clients) and if this thread shows us anything, it's that there is a need for more interpretations, not less. Because they can't be "right" or "wrong", it's too subjective.

@Lenny Abramov

@Vendan

But blocking support for custom/emoti voting does nothing to make this harder. All it means is people post hate messages as replies rather then votes. Trying to technologically censor (and that's what you are doing, essentially) just by making an action slightly harder is pointless.

I'm not technologically censoring, I'm just saying that voting makes sense in specific situations only, so it shouldn't be attached to every post as a standard feature. It's sort of like the "join" button: it makes sense next to an invitation to an event, but not in general. No one would say blocking support for a "join" button in every single post is technological censorship.

If voting is so important it should have it's own feature: polls, i.e. a special type of posts that admit a limited selection of replies.

Few words about voting. Are "reactions" (emojis) a form of voting? I think so. There's voting whenever you group and count replies. Even if the options are all positive, e.g. :laughing: and :+1:, twenty :laughing: as comments to a perfectly serious profile picture are effectively negative votes. An even when the only option is :+1:, three of them feel as ostracism when all your friends get always 50+.

Regarding bullying: I know this won't stop bullying, but it would provide one less chance. Let's just keep this in mind. Let's weigh this extreme negative consequence against the positive effects (if there are any) of enabling voting in every post.

@Lenny Abramov

@robin

we're in an age where reality is not fixed. premodernity it was, nature was everything. given the plasticity of culture now, firstly reality is a moving target, secondly it's shaped by the multiply recursive feedback loops between many, many individuals, and thirdly the very act of trying to design ssb or its interface effects the reality we are trying to measure.

These are also assumptions about reality! There's nothing wrong with that. Since we can't get reality in itself, we have to make assumptions. Let's make them explicit - just like you've done - and let's build upon those.

@Vendan

An even when the only option is :+1:, three of them feel as ostracism when all your friends get always 50+.

And then it becomes 1 reply versus 20+ replies, and how is that any better... I still say you are trying to impose social change via technological control, and I really don't think that's going to be effective as you hope.

@ezdiy

@Dominic Arbitration works now, but on bigger scales costs a lot of energy (and pub owners are already kind of forced into that positition, for now).

People on this issue can be divided into two groups:

  • Those who prefer to not have any kind of hivemind, use /ignore, but don't distribute blocklists
  • People who eagerly distribute and use blocklists

Both approaches can work, to their own advantages and disadvantages. What is great about SSB is that end user can choose either. If anything, I'd provide end users a clear choice which of the two approaches to moderation they want to participate in, and not force em into one or another category by whatever the client implements (as that would be client forcing opinion on moderation).

@alex

actually i agree with @Vendan and am not so sure what's the value of "like"s.

Maybe they should not be public but just shown to the author? :P ...or maybe there is no need for like's at all

@alex

What about having a counter that says: seen by X people instead of likes if that's what it's for.

@joegle

What about embedding smart contract replies so that we make arguments more interesting? #bitcoin

@Lenny Abramov

Private reactions and replies

I’ve thought about short replies and “reactions” – where the need for them comes from and how it can be adapted to SBB.

It’s interesting to note that in instant messaging there’s only one type of message, and that’s enough to convey complex thoughts, nods, uh-huhs, emotions, etc. Obviously face-to-face interaction is much richer, but it’s still possible to render many things through text and emojis, and no big dilemma arises. Instant messaging is a simple medium with big limits, but there isn’t much we can do about it.

On the other hand, online public discussion is very different from instant messaging. I wouldn’t compare it to face-to-face interaction among two or few people, but rather to a large gathering where people take turns to express complete thoughts to a quiet audience. In such physical gatherings, one can distinguish several types of interaction:

a. A spoken statement to everybody.

b. Collective feedback: applause, laugh, boo…

c. Direct feedback from listener to speaker, through gaze, nodding, smiling, yawning…

d. One-to-one talk (probably whispered and brief).

e. Direct questions to the public, likely answered by raising hands.

In SBB a clearly corresponds to a public post. “Likes” (and hypothetical closed-choice “reactions”) are perhaps a good translation of b, because they are constrained and contagious. There is no specific feature that matches c, while d is served by private messages, although the context is not automatically tracked. There’s also no specific feature for e.

As far as I understand from several comments in this thread, many of us want something like c. That is, a short feedback for the poster that doesn’t need to be replied, is possibly nuanced, and isn't relevant for the public. That’s completely different from, say, an applause, which is chiefly a collective behavior (how embarrassing to clap your hands alone!). So I think it deserves its own feature.

Private feedbacks that require a reply should also have their specific space, as I argued in a previous post.

So I made this ugly mock-up to show how I would display private reactions and private replies (this is what robin would see):

private-replies.jpg

Reactions are presented as a grid of avatars. By hovering on the avatar the actual message is shown. Reactions are free-form: they can be emojis, a word, a short sentence like “Interesting! Reminds me of Shakespeare”. Just like a speaker scans faces in an audience to gather feedback, this system would allow you to see at a glance who's read your post, to know the reactions of people whose opinion you value most, and to simply disregard the superficial and the trolls. The rationale here is that faces are the first important piece of information, and the actual message is important only when the face is. Contrast this with the usual reaction-counter, where you first see quite meaningless numbers next to standardized messages, and only after hovering can you know who’s behind the numbers.

Private replies are easy. Since they’re private they’re like independent threads, and can be shown in separate tabs. The whole private replies section could be expanded/collapsed.

I’ve argued above that “likes” (and public, constrained “reactions”) have something in common with applauses and collective laughs. That makes sense, but there’s the crucial difference that “likes” get counted, and the count stays there forever. This makes them more akin to votes. Indeed, stripping those collective behaviors of their fleeting quality feels like a complete distortion of their nature. All in all, there seems to be no good translation of collective reactions in the online world.

Polls would be a totally legitimate feature. A post could define its own accepted public “reactions”, which would be counted and publicly displayed.

@Vendan

That is, a short feedback for the poster that doesn’t need to be replied, is possibly nuanced, and isn’t relevant for the public.

My issue is that this is still just a technological control to attempt to fix a social ill. What's to stop a bunch of trolls from all directing the same demeaning reaction to a post? Just cause the reactions are private doesn't stop any kind of out of band "here, attack this post". Heck, SSB includes private messaging as well, so they can even use SSB to privately arrange an "attack" on a post. It seems two faced to say both:

What’s going to happen when you allow to vote, say, a selfie? What used to be a dystopia has unfortunately become a tragic reality. People literally die for this sort of stuff. I know I sound melodramatic, but we should carefully consider these choices, and ask ourselves questions like: Are we willing to trade vulnerable kids’ lives for an uncluttered UI?

AND

simply disregard the superficial and the trolls.

Also:

I’ve argued above that “likes” (and public, constrained “reactions”) have something in common with applauses and collective laughs. That makes sense, but there’s the crucial difference that “likes” get counted, and the count stays there forever. This makes them more akin to votes. Indeed, stripping those collective behaviors of their fleeting quality feels like a complete distortion of their nature. All in all, there seems to be no good translation of collective reactions in the online world.

About the only real distinction is that applause doesn't get precisely counted and measured. Speeches are just as fleeting as applause (in that either is just stored as memories in those involved, unless it's recorded). In person conversations don't usually involve both sides wired up for everything including video recording, but SSB conversations, even private, involve recording everything permanently, and there's no way to prevent one side from decrypting and posting everything .

I think attempting to create a direct 1:1 between real life and digital is a bad idea.

@Lenny Abramov

My issue is that this is still just a technological control to attempt to fix a social ill.

I think you misunderstood my goals. It's probably my fault, because I made that pathetic remark about kids dying because of bullying... but I'm not trying to fix a specific social ill. I'm just aware that UX design affects society, and I try to justify the design choices looking at real-life and digital interactions that I think are healthy and effective. You can call that "technological control" if you like, but there's no way out of it, everybody does it when they design a social network.

simply disregard the superficial and the trolls.

My point is that if you don't see the reactions until you hover on the avatars, it's much easier to disregard the trolls, or just people you don't know / don't care about.

Speeches are just as fleeting as applause

Yes, but if you freeze speeches you don't alter their nature, while if you freeze applause you do. That's my feeling, at least. Not a very solid argument, I admit.

I think attempting to create a direct 1:1 between real life and digital is a bad idea.

That's a good point. But it doesn't erase the need to justify each design decision. Why do we have a like button? Why do we count likes? Why do we show them publicly? Why do we want reactions? Why that set of emojis, and not a different one? That's what I'm trying to do. Not fixing specific social ills.

Also, my description of that real-life situation didn't come out of the blue. I included the nodding, for instance, because Dominic said he uses the like button for that purpose. So I asked myself "When do people nod?" and "Is there a better way to translate nodding into digital interactions?".

@Vendan

Yes, but if you freeze speeches you don’t alter their nature, while if you freeze applause you do. That’s my feeling, at least. Not a very solid argument, I admit.

If you freeze applause, all you get is a volume of clapping. You often hear it after speeches, after all. The distinction is that social media tends to label applauders, not just record that there was applause, or a general estimate of how much there was. That's one reason for collating count and having to jump to see names.

That’s a good point. But it doesn’t erase the need to justify each design decision. Why do we have a like button? Why do we count likes? Why do we show them publicly? Why do we want reactions? Why that set of emojis, and not a different one? That’s what I’m trying to do. Not fixing specific social ills.

Bluntly, probably because most other social networks (esp. the major ones), have a like button. Familiarity eases transition. Though it should be noted they don't generally have a "dislike" button.

@Lenny Abramov

Here's my exercise in social network design: Focus, Diversity, Memory. How to Design a Better Social Network. The idea is actually a very old of mine, I've been thinking about it for some years. It's very simple, but the post is already long enough.

I'm also thinking of a more "social" social network, like an extension of the current SBB. I'll write another post about that.

@Lenny Abramov

Sorry, this should go in it's own thread.

Join Scuttlebutt now