I wish I could believe the implication of the article, that the Mastodon UX improves the quality of discourse, but I strongly suspect that there's one much more important Twitter feature that Mastodon lacks that is vastly more important than any mentioned: Twitter just has far more users.
Eliminating classes of problematic behavior and incentives by way of avoiding or altering technical features is always going to be a game of whack-a-mole. When you get above some threshold—maybe it's some sort of 'relationship distance', or maybe just a raw number of people—then people will always find a way to game the system. I don't think it's a bug or misfeature in the system, though the system can certainly contribute to where the threshold lies. It's a bug in collective human behavior.
So far, the most likely solution to the problem of a mass point-to-point communication system is to not have a mass point-to-point communication system. Shard it down until people are able to behave reasonably well, even if that means you can't communicate with just anybody. That completely breaks some use cases, but it turns out that anything that allows those use cases turns into a cesspool. Maybe somebody can invent something someday, but I'm doubtful that technical measures or UX can ever compete with our rage and other base urges.
Yeah, I would say that features of the software can definitely affect the community but there's a limit.
If you actually have Mastodon grow to include any significant portion of Twitter's userbase then you're going to get much of those same behaviors.
Like, on the "dunking" example: people will just use screenshots. Memorably, Twitter didn't have even retweets well into its life, but the users created a de facto system for doing it. This "pave the cowpaths" feature growth was once a defining aspect of Twitter.
Many of the people who like Twitter like to dunk, they like to read dunks, and they'll do it wherever they go.
> So far, the most likely solution to the problem of a mass point-to-point communication system is to not have a mass point-to-point communication system. Shard it down until people are able to behave reasonably well, even if that means you can't communicate with just anybody.
Yes, that is one of the main points of the Fediverse. There are people from all sorts of viewpoints there, all the way from Gab to LGBT instances, but my home instance is my "window into the Fediverse", both via instance filtering/silencing/blocking and via organic reach (content that people on your server never interact with won't reach your server), which creates this organic sharding.
In essence, I interact with "a corner of the Fediverse", with a fuzzy/organic boundary, and I don't expect to be able to communicate universally.
QUOTE FROM ARTICLE (sorry, not sure how to properly quote on HN):
Note how this is all emergent behaviour. It’s not based on blocks, though blocks are important to weed off bad parts of the subgraph. But independently from that, each instance’s view of the fediverse is unique, and without anybody doing anything, it naturally converges towards a community that makes sense to the instance’s members.
In practice this results in a feeling of social cohesion and intimacy that I haven’t felt online since early Livejournal days.
Isn’t that a bubble? How do you avoid an echo chamber?
You can follow people with contrasting values if you want, nothing’s stopping you and you don’t need admin permission to do that. But why would you want to write to people who don’t want to read what you write? It’s no more a bubble than your circle of friends.
(MY NOTE: Also, you can create multiple accounts if you want multiple "windows into the fediverse". Or create your own server. Or join one of the "free speech" ones.)
I get a lot more interaction from a lot more of a diverse crowd here than I did on twitter, because a single shared space like twitter ends up creating a superstar vs. mere mortals dichotomy, and as a mere mortal not posting discourse bait, I never had anybody read my stuff in the first place. Here there’s a small group of people who always interact with my stuff, and I know everybody by name, and I love them all.
But how do I promote my toots to a large audience in a system like this?
That's the neat part, you don’t.
How do I make everything easily discoverable by a large audience?
Did anyone else have to read this title several times to parse it? "Twitter features _that_ Mastodon is better for not having" is better, but still seems strange
edit: I think what trips me up is that "features" is used here as a noun but because of it's placement in the sentence I want to read it like a verb
I am not a native speaker and I have always found US news headlines to be weirdly constructed to the point I can't tell if it's a normal headline or if there's a mistake in it. What's worse is that in my native language I also notice a decline in the quality of writing, to the point where sometimes whole sentences are indecipherable.
In this case I only needed to read it once. I often feel like "that" and "and" words are dropped in headlines. I suppose it saves some precious bytes. /s
Even after reading it several times, I could not parse that title, precisely because I kept interpreting "features" as a verb. Viewing the actual article yields the real headline, which makes more sense:
4 Twitter features Mastodon is better for not having
That "4" at the beginning makes all the difference!
Twitter’s problem is that it encourages engagement with the absolute worst and most divisive users. Don’t agree with an antisemitic tweet? Your best option is to leave a reply and ratio the poster.
Twitter desperately needs a public downvote button with a mechanic similar to HN. Flame bait should be pushed to the bottom of everyone’s feed, not the top.
Have to say I wasn’t expecting “antisemitism is fine” as a counter argument to downvotes. Suffice to say this comment isn’t the slam dunk you think it is.
Yes it is, and I don't care about how popular my comments are. In many slavic countries antisemitism is the default and normal opinion. Semitic people are not the sacred cows they are in american culture.
Which Slavic countries, specifically? I come from one, and even though there are (borderline) antisemitic jokes (an X, Y and Jew go into a bar, or Holocaust-related jokes which i personally choose to interpret as just a specific brand of dark humour), antisemitism is not a "default and normal opinion". It's skinhead/football hooligan fringes.
So, which Slavic country has a majority of antisemites, today? The horrors of WW2 have taught all of us better, one would think (idiots such as yourself being the exception, not the norm; clarification: yes, you have to be an idiot to be antisemitic or racist or any other similar description).
mastodons biggest shortcoming remains the lack of e2e encrypted DM's. Giving volunteer internet janitors access to your private messages is a recipe for disaster
Because of the friction starting a public convo, going to DM, then negotiating which private messaging system to then switch to again. Most people won't bother.
Having person-to-person messages be unencrypted is irresponsible in 2022, the same way offering a web service for public consumption without TLS is. Encryption should be the default and the user should not have to even consider the threat of their messages being read by unauthorized parties.
The most irresponsible thing you can do is make a security promise you can't keep. Mastadon's usage is overwhelmingly browser-based; achieving reliably end-to-end security between users of browser-based apps remains an open problem. Taking a short step back from that: if you're going to try to give people secure messaging, you should have that goal from the start. Matrix is a good case study in what happens when you don't do this.
You can still high-horse Mastodon: just tell them they shouldn't have private messaging at all. That seems like a reasonable take.
Since matrix is an open standard and everything would it be possible to build a matrix client into mastodon? That would be really interesting, if it became a plug and play messaging client for open source projects that include some sort of DMs.
I just included it as a default case. If you want to assume malintent, that's on you. The more interesting cases are the last two because they reveal that messaging cannot be made private because you cannot prevent the counterparty from leaking information.
There are plenty of actual chat apps out there like Signal or Matrix that you should be using for this sort of thing. Neither Twitter or Mastodon are the right tool for this.
As much as twitter, Reddit and Facebook don’t like to admit it, it seems all would tell you that “spirited debate” drives significant engagement and traffic. God knows I’m guilty of it.
I’ve seen “thoughtful” conversation platforms succeed but they tend to be fairly niche and/or heavily moderated. I’m sure the suggestions here would make Twitter something different. But I’m not sure it would be “better”.
Not so sure about quote bit. I really like the quotes because I follow people for perspective on articles and content. Seeing the article summary and the comment as to why it’s good/bad/recommended reading is perfect. It’s basically hundreds of people being unpaid news aggregators and filters for me.
If their tweets were “this is a good article on why X is good [link]” that would be a much worse UX for content aggregation.
I have zero interest in discourse (and most people I followed don’t engage in any, nor do I).
Also, last time I tried mastodon (many years ago) there was a lot of focus on the distributed/federated thing. I found it hard to emulate the global timeline of posts from just people I follow. Maybe it was just me not understanding the UI, but I failed to create a Twitter feed like experience back then.
I hope Mastodon (ActivityPub protocol to be precise) could be statically hosted, like a podcast or rss, yes it's only one-way but hell it will accelerate adoption.
Unfortunately that does not work with Mastodon. When you view a user's profile instead of loading their current list of Activities Mastodon will only show you the ones which have already been propagated and cached on the current instance.
In the case of a static list of activities that won't work.
In my opinion it's a mistake to keep functionality like this, and there is a bug report on their tracker about it but I don't think it gathered enough interest from the devs to make it better.
The "no full-text search" benefit of Mastodon seems unsustainable, if Mastodon succeeds at Twitter's scale; not because the platform will add the feature, but because others will do it for them. Isn't that what happened to Twitter? (I might be misremembering.)
That's not how categories and social constructs work. These categories (age, sex, gender) are words that are subjectively defined by us based on utility and common consensus, but nevertheless refer to hard objective reality. Someone with age 35 cannot say they're age 5, because the definition of that particular construct precludes that. The construct points to the objective reality of time elapsed since birth, which can't be changed by your whims.
Agree for age and sex, but gender? Most of us came from a society were there was only two, but ask any hindu about Hijras? So for you Hindus were wrong, but how can they be? Or this is just o society thing (like the color blue being attributed to men since the industrial revolution, and to female aristocrats before).
Yes, even gender. Gender is just another subjectively chosen category, which points to objective reality. In this case, it points to internal brain states, instead of sex chromosomes. It's like the word "Belieber". That's a category which refers to self-described fans of Justin Bieber. There is nothing factually incorrect with such a categorization. In fact, categories themselves logically can't be factually incorrect. It's only the faulty application of an existing category that can be factually incorrect. Categories can have fuzzy borders, but that's often just the nature of words whose definition emerges from conversation and common consensus.
> So for you Hindus were wrong
Hindus were also right -- in their local context where the word "gender" is defined that particular way. There is no globally right or wrong way to define the word "gender", because "gender" is just a word whose definition is decided by local common consensus, just like any other word in a language.
Different societies have different words for common greetings with slightly different implications on meaning. None of them are right or wrong because these words are simply containers that can hold whatever we want them to hold. And it's up to common consensus to put stuff into those containers.
The underlying reality doesn't depend on the society you're in, just the categories and definitions of words that we create to make sense of reality and chunk that reality into digestible abstractions. So it's weak social constructionism, yeah.
Age is a useful construct for humans. It's an easy to communicate concept based on a shared wall clock we can reference. It abstracts away a lot of more complicated and nuanced factors.
People can accrue all sorts of physical changes at different rates (that still fall within a tight distribution). Some people mentally mature faster than others due to experience, biochemical differences, etc. Some people biologically age faster due to genes, stress, injury, illness, or disease. The volume of UV exposure one receives changes your appearance.
Age happens to closely correlate with so many societal concepts we care about: schooling, maturity and behavioral expectations, dating and marriage, career, having children, getting cancer, dying, etc. Age is universally useful and interesting to us.
It is, but this is such a reductionist, materialist view of the human condition that only bad techbros up to no good would want to talk about it.
Of course Age is about your temporal identity and the intersection of it with social structures of discrimination and oppression. This important fact is not yet well-known, but worry not, soon we'll have a field of study and several journals dedicated to it.
Nazis aren't a unique one time occurrence. But we can call them fascists or neo nazis too if you want.
Anyway, people who use a lot of racial slurs and have nazi symbols in their avatars seemed really happy about Musk "restoring free speech". It's a legitimate concern that Twitter will turn into a place that's much more friendly to these people - and thus also much more hostile to anyone to their left.
There are many better ways to disagree with people than doing the Nazi association. "He is one of / i associate him with the baddies" is cheap and dishonest.
Making a site "nazi friendly" doesn't mean you call people nazis. It just means you tolerate their existence to the detriment of other people wanting to stay. Free speech absolutism - even misguided - can and has caused this. Been to any chan's /pol/ board recently?
> Conclusion: nonprofit social media enables healthier design
Twitter is(was?) popular because it was not healthy at first place, it was just that it used to be a "safe space" for a certain category of people.
I quote the blogpost:
> Now that a narcissistic billionaire bought Twitter and will likely make it Nazi-friendly (or otherwise ruin it)
These (also narcissistic) people aren't interested in healthy design, they want an exclusive megaphone, while denying their ideological opponent the same thing. They aren't interested in Mastodon or they would have moved to Mastodon a long time ago, when the people they denied the megaphone moved to alternative social media.
Free and open source software does not discriminate based on political ideology. Unlike Twitter, which is a single instance controlled by a single corporation, Mastodon allows each instance operator to set their own rules. Each Mastodon user can choose the instance (or instances, since most Mastodon clients support multiple accounts) to register their account on.
So is the only meaningful difference the scale of an instance? If a single Mastodon instance becomes Twitter-size, then its owners will have the same power over the users and will face the same challenges as the current Twitter? They won't be able to ban users cross-instance, but if they ban you on that instance, wouldn't you have to create a new account somewhere else (i.e. start from zero).
Mastodon will not evolve into a single-instance monolith like Twitter because fediverse users are too diverse to agree on a single set of rules. Some instances are general purpose while others focus on specific topics. Some welcome certain types of content while others don't.
If you choose an instance with rules that you agree to follow, you're not likely to be blocked in the first place. But if your Mastodon account is blocked on an instance, it's easy to make a new one on a different instance. There are many other instances to choose from and you can regain your social graph by following everyone again.
I could absolutely see people saying that the power-generators are Nazi-friendly in this circumstance, and that the people selling/servicing them should be criticized accordingly.
Going by that logic, naturally grown food is Nazi-friendly because Nazis can just plant them in their backyard. Should people selling seeds/seedlings be criticized accordingly?
That whole argument seems like a badly disguised call for total surveillance.
> Should people selling seeds/seedlings be criticized accordingly?
Well, the line of cancelling someone until they literally cannot attain food has not been crossed... yet. But if say the Nazis were selling organic food to fund their activities, plenty of people would say it's wrong to sell seeds to the Nazis.
I'm not saying that I support this, just that this is what I see the "duty to cancel" arguments implying under these circumstances.
I'm sorry if I implied you do/don't support any of the arguments we're discussing. I seem to have a bit of confrontational style of arguing, I'll try to tone it down in the future.
On the original point - it seems to me that the whole cancel culture is just an attempt by a vocal minority to bypass democratic system. Centuries of lawmaking have made some things legal and others not, walking very thin lines between good and bad decisions in order to maintain the balance between totalitarism and lawlessness - and suddenly a mob of random internet dwellers think they can get their way by throwing tantrums and bend the rules as they wish.
I'm annoyed at the very efficiency with which mob manages to cancel whoever does anything remotely "unacceptable" by their unwritten, "come on dude we all know that", rules... It seems like the futute is likely to be lead by loud demagogues skilled at tapping into the mob rage of the society and cleverly weaponizing it against their political enemies. Like always.
It's fair to say you can pick a Mastodon server based on rules that make sense to you, which is not the case for Twitter. If your server admin starts posting nutty propaganda (as Musk did today; https://www.nbcnews.com/news/us-news/elon-musk-new-owner-twi...), you can migrate somewhere else in a matter of minutes.
(not familiar with Mastodon tech) Would I be able to move my social graph as well in your example, or just content? E.g. if I first registered an account at mastodon.se and many other mastodon.se users followed me, would they still see my updates if I move to a different instance (or if I am banned from mastodon.se)?
Is it fair to say that now if I follow you on any instance and then you are banned from that instance, I will continue following you and seeing your posts because your identity is owned by your self-hosted instance?
Mastodon is just an interface for the larger Fediverse network. It does not control the network any more than Microsoft Outlook controls the email network.
Similarly, email is a "nazi friendly" technology. Are you going to not use email because of that? It's not reasonable to base every choice in your life around whether a "nazi" could potentially make the same choice.
Yes and no. Mastodon has been popular with the alt right because they have been banned from other platforms but these "nazi" servers are probably not part of network you join. I doubt a car enthusiasts server would be following any political server.
Since every server can pick which servers they want to federate with certain groups will not connect with each other. This however doesn't prevent you from having two accounts.
The solution is mentioned above. Unironic nazis/white nationalists/white supremacists will only last long on free-speech-oriented instances which are likely to be defederated from any instances which explicitly disallow hate speech. Pick one of those latter instances as your home, and you will most likely never see a "nazi" on your timeline unless you go out of your way to go hunting for one.
Yeah, it's ironic that Twitter is allegedly more "Nazi-friendly" than Mastadon - presumably this "Nazi-friendly" shift will be due to a pullback on the aggressive, centralized moderation of mainstream conservative viewpoints. But Mastadon doesn't have this to begin with! I guess the argument is that within Mastadon, it's easier to create your own filter bubbles that would allow you to enjoy an experience untainted by "Nazis"?
Your local instance moderators choose are responsible for who their server interacts with. Ultimately this creates many segmented branching networks of federation, and with an appropriate instance, nazis won’t exist in your network.
> I guess the argument is that within Mastadon, it's easier to create your own filter bubbles that would allow you to enjoy an experience untainted by "Nazis"?
This is absolutely the argument.
It seems that for some people, the ability to control your own environment is not enough - they want the authority to control everyone else's environment, to make it impossible for people they disagree with to exist.
Of course, it always starts with Nazis, because Nazis are universally hated and not arguing against Nazis will get you labeled as one very quick. But in order to be able to make it impossible for one group of people to exist, you must have the ability to make it impossible for any group of people to exist. And that's power I would not trust anyone with.
That's true on paper, but is glossing over a huge issue: the Fediverse is strongly politicized. Mastodon instances are under pressure to not federate with undesirables, under threat of being defederated by others. As a result, you have one "primary" network of instances whose admins and members either subscribe to specific kind of political extremism, or are at least willing to acquiesce to it for the sake of the users to have anyone to talk to, and a forest of small clusters. The "primary" network and secondary clusters don't talk to each other, and associating yourself with the latter is a quick way to be booted off the former.
Technically, this is federation working as intended, but it's not as diverse and inclusive as it advertises - on the contrary, it's more like a HOA that has a right to ban condo owners from talking to each other.
It's a nice place to be when you're not into politicking, but then so is Twitter, so it's not that big of a win. A more compelling argument for Mastodon is that it's community-funded, so it's not out to sell your eyeballs.
Eliminating classes of problematic behavior and incentives by way of avoiding or altering technical features is always going to be a game of whack-a-mole. When you get above some threshold—maybe it's some sort of 'relationship distance', or maybe just a raw number of people—then people will always find a way to game the system. I don't think it's a bug or misfeature in the system, though the system can certainly contribute to where the threshold lies. It's a bug in collective human behavior.
So far, the most likely solution to the problem of a mass point-to-point communication system is to not have a mass point-to-point communication system. Shard it down until people are able to behave reasonably well, even if that means you can't communicate with just anybody. That completely breaks some use cases, but it turns out that anything that allows those use cases turns into a cesspool. Maybe somebody can invent something someday, but I'm doubtful that technical measures or UX can ever compete with our rage and other base urges.