Mastodon isn’t just a replacement for Twitter
noemamag.com
noemamag.com
Just Mastodon Isn’t a Replacement for Twitter
Mastodon Just Isn’t a Replacement for Twitter
Mastodon Isn’t Just a Replacement for Twitter
Mastodon Isn’t a Just Replacement for Twitter
Mastodon Isn’t a Replacement Just for Twitter
Mastodon Isn’t a Replacement for Just Twitter
Mastodon Isn’t a Replacement for Twitter Just
"how's the article coming along?"
"hold on, I'm twitterjusting it"
It's meant to convey a sense of defeat. So in this case the sentence would read as implying that the author wanted Mastodon to replace Twitter but that it's not able and that fact is causing her emotional pain.
I giggled at them trying to explain Mastodon, though:
> There is no such thing as a social network called Mastodon! Instead there are thousands of independent social networks called Mastodon Instances.
...three paragraphs later...
> When we say “Mastodon” we mean the whole network of Mastodon Instances
I know what they were going for, but it seems to undercut itself. :)
Is there a summary that can be read <5 minutes?
It doesn't need to aim for hockeystick growth or guard its data to create value for investors. It doesn't need to manipulate users for high engagement metrics to impress advertisers.
Personally, I started to look at nostr [0]. It gets rid of this idea of decentalization-by-federation and makes nodes mere relays of information. The architecture is infinitely simpler and it a lot easier to scale. I'm wondering how easy it would be to fork an Mastodon/ActivityPub client and make it work with a nostr relay, and I will probably take the christmas break to add a nostr relay to communick.
- Both are useful. They serve different use-cases. There are elements of each I like and ... am frustrated by.
- Mastodon is mostly short-form content analagous to Twitter, and strongly favours media (images, video). Other thread participants are not notified of new replies unless specifically tagged within it.
- Diaspora* supports longer-form, formatted (Markdown) posts, with a single time-ordered discussion below which can be moderated by the post's author. Participants are notified of additional comments within a thread, and can return to it, even after years.
- Those two distinct formats already strongly shape the type and nature of discussions found on each.
- Mastodon has seen vastly more development, including addressing several issues I've long had both specific to Mastodon (CW behaviour is ... improving, though I've still got concerns) and both platforms (Mastodon now supports edits to existing toots, Diaspora* ... does not). By comparison, Diaspora*'s development is all but moribund.
- Mastodon's developers have been far more responsive, and empathetic, than Diaspora*'s. How much of this is just green-behind-the-ears freshness I'm not sure.
- Mastodon presents non-members with useful information. Visit an instance and you can actually see activity and discussion on that instance. Hit up a Diaspora* instance and ... you're given a big Fuck You with a registration sign-up.
- Neither platform is especially search-friendly, either on-platform (both support only Hashtag search) nor through third-party general Web search. Diaspora* seems all but entirely opaque on that point, Mastodon ... at least surfaces some content.
- Diaspora*'s original, oldest, and largest instance went titsup.com earlier this year, after years and years of all-but-absent management and oversight. Even the contact emails and forms were offline for years before the instance died. Another large instance, created by and for refugees from Google+, ceased active operations after its administrator died unexpectedly (also after a long period of radio silence). Mind I've had a couple of Mastodon instances fold under me and/or show sketchy management as well. Bus factor and continuity planning are major issues on both platforms, and seem almost entirely unconsidered.
- Diaspora*'s total MAU is a small fraction of Mastodon's. There are a number of people I appreciate greatly on Diaspora*, but it's ludicrous to suggest that there's a substantial community of any sort.
- Diaspora*'s abuse tools are if anything even more rudimentary than Mastodon's. One of the principle mechanisms are private lists of abusive / spamming profiles maintained and privately posted by individuals within their own groups. Admins can do some instance blocking, but ordinary members cannot. An "ignored" profile still sees content posted by the ignoring party, and the ignored's comments on third-party posts are still visible to the ignoring party. (There may be reasons and arguments in favour of this, but the functionality is unexpected and confusing to many.)
The only reason I've heard someone suggest to move from Twitter is because the new owner opened his mouth about political views.
If they have to split their efforts among 10,000 independent groups, maybe the EV-chasers will go away and people will talk about things other than the latest scandal or movie.
- No data mining
- No opaque algorithm that can not be tuned by the user. Okay, Mastodon has only one "algorithm" (aka reverse chronological order) but there is nothing stopping us from developing clients (user-agents) that filter the timeline according to what we think it's important. [0]
- Who cares about "all the users"? The important thing is that the you can connect, follow and interact with a large-enough group of interesting people. Even if the whole network has "only" 8M people, my timeline on Mastodon has certainly become more interesting than the twitter one.
[0] https://mastodon.communick.com/web/@raphael/1093815543157191...
Like, especially, a bad actor could run several large instances with no public relationship and so all kinds of correlation.
- I can not put tracking cookies on other people's clients. Twitter puts tracking cookies, can collect location data, can get your contact list...
- If any instance starts manipulating messages from its users to add tracking redirections, people would immediately lose trust over it and drop it. With twitter, there is absolutely nothing you can do and people treat t.co as an inevitable necessity.
- Even if someone went on to analyze the data and tried to use in the messages to give people personalized ads in some custom client, what would be the incentive for people to stick to it? If the publisher wants to have a business model where they want to share ad revenue with the users, they might as well go with something like Brave.
For people that use Twitter but care about their public square on the internet being slightly more sustainable and human-scale, Mastodon's model makes a lot of sense. Mastodon provides the tool for building your networks, making it possible for small organizations to run instances basically as utilities. I'm particularly excited for the co-ops I've seen in this space.
Can you be a bit more specific? This literally applies to the top 6 media corporations, Amazon, several banks, and pretty much every other large corporation.
The use by Musk of the term "public square" when he's running a private corporation on his own whims is what is causing him problems.
So far my experience is reminding me of 1990’s style connecting and chatting with interesting random people from around the world who are friendly and quirky, and I find being there a “nice” experience.
So. I've been on Twitter 15 years, since early 2007. I feel a lot of affection to it in some ways, but particularly over the decade, it has become in many ways a pretty unpleasant experience. And yet I was still there. Matt Levine on Twitter:
> My favorite might be Ben Thompson’s rather bearish post, which argues essentially that Twitter is a very unpleasant product that most people do not want to use, but is extremely appealing and addictive to a small minority of us with significant personality flaws.
He's probably not wrong.
Elon has either promised to, or already has, made the site worse in a number of ways:
- The paid checkmark thing. People who pay for this will, Musk has promised, have their tweets essentially boosted by the algorithm. Frankly, I don't particularly want to hear from the sort of person who's buying the checkmarks, and paying for attention is a corrosive design dynamic. Even dating apps, which invented it, at this point mostly realise it can only be used sparingly without destroying the experience (for instance, Grindr has a thing where you can pay a couple of euro to appear higher in the grid of people than you should for an hour or something, but not permanently, because then the grid would be permanently full of the sort of people who pay for attention, and no-one wants that.)
- He's bringing back all manner of complete monsters. There are the big high-profile ones, of course, but also thousands of local monsters. Irish twitter got a lot better a few years back when a couple of prominent local Neo-Nazis were sent on their way, say. Twitter moderation was already too lax; this is very much going the wrong direction.
- A lot of the people I liked following have either already left, or are much less active.
- I find Elon personally extremely irritating, and there's just no escaping him on Twitter right now.
After 15 years, I'm basically done. I'm keeping the account for now (you never know; there's a bit of a history of social media sites being re-sold after a bad acquisition), but Mastodon is now filling the gap for me. It's pretty good; there are some annoyances, but in general it scratches the same itch as Twitter without a lot of the irritants.
I do think one mistake Elon makes is that he seems to think everyone loves Twitter. My suspicion is that most Twitter users, or at least most heavy users, find Twitter barely tolerable, but addictive. It doesn't take much to make people in that situation say "screw this".
I don't expect Twitter to go bankrupt, but it will need desperate measures to stay afloat. When it'll come to "either we do this or we go bust", they won't stop at anything. It's likely going to be a massive sell-off of user privacy on a scale that will make people run to Facebook for safety.
Has it still? A lot of interesting people I used to follow on Twitter either don't post much there anymore or straight out deleted their accounts when moving to Mastodon. I haven't deleted mine, but I certainly feel much less inclined to cross-post things to Twitter anymore as well.
Your handle consists of your alias and your home instance. ie, "@sohcahtoa82@nerds.social"
If I want to move to another instance, say, jocks.social, then my handle would become @sohcahtoa82@jocks.social, and that's assuming @sohcahtoa82 wasn't already taken on jocks.social.
Instance mods can block entire instances, making it possible for an instance (or set of instances) to become isolated from the rest of the Fediverse. At that point, those instances might as well be a completely different site, since they'll no longer communicate.
And I somewhat expect that to happen. Many instance admins will block instances that become known to be cess pools of hate speech.
It's puzzling to me that "Discord" is even in the same sentence as Twitter or Mastodon. It serves an entirely different use case, being designed for real-time communication.
Why Mastodon isn't as large as twitter is a question by itself. Twitter's continued existence doesn't matter.
I honestly don't get the hype.
As users begin migrating to the noncommercial fediverse, they need to reconsider their expectations for social media — and bring them in line with what we expect from other arenas of social life. We need to learn how to become more like engaged democratic citizens in the life of our networks.'
All right, running with this: corporate sponsored "public spaces" aren't really public (as they are capitalist entities). Let's add "Conway's Law" [1.] to this, and extrapolate that perhaps democracies currently are repeating the design of the communication structures that comprise it (of which there are myriad, but assume the largest/loudest have more sway). Let's pick one nation, the US, and question: is the US democracy constrained by the current corporate-sponsored public-spaces/communication structures (which have been old media, i.e. print, but seems now new media, i.e. all online). A 'fediversification' of the public mind seems really important here, the question is how far can/does it go (techno-first generations seems fine, but let's say baby-boomers perhaps not so much?).
I'm curious what others think about this, and what resources can be shared to provide more insight into this question.
[1.] Conway's Law: “Organizations, who design systems, are constrained to produce designs which are copies of the communication structures of these organizations.”
I keep wondering, who pays to run all those servers, handle users on the servers, and for the people to spend time keeping it all up to date?
Will it end up with some companies running servers for people. Still, where does the money come from to cover the expenses?
Today, volunteers and goodwill. Tomorrow, likely the same actors that run free VPNs today, ie the tier-2 ad-tech/data-broker shadow industry.
That's the sort of money people happily pour into hobbies without a second thought. Collectively my Discord server of like 30 people pay more than that on their premium offering. All it takes is one person in your community to foot a relatively trivial bill. Or, orchestrate a Patreon or similar.
To an addict like I am, Twitter timeline is something measured in tweets per second. It is a constant flow, a time synchronized operation, not something that I expect to be event based.
Mastodon devs and ActivityPub spec devs must have a complete opposite of that, that users only post content when there is something to say, and the post will be pushed when a post was made.
The latter model is certainly doable with the ActivityPub protocol, $20/month budget, and by stretch can be handled by Mastodon as it is built. But I think there is this misunderstanding, and it just seems that none of it is built to cater for those with the former model.
That information is public for at least one moderately popular instance, fosstodon.org. According to https://hub.fosstodon.org/about/#current-funding the current cost is $1895.50 per month, plus $200.00 for the CDN. (And yes, last time I looked, their Patreon is going to be more than enough to cover that.)
You'll join a community subject to agreements that are periodically renegotiated, but those negotiations and the resulting enforcement will be done by idle members rather than those too busy to notice minor infractions. Likely a better fit with human psychology than living in a monolithic and decaying walled-garden, but also likely to result in spaces where people largely talk about what they do outside of them. Some people will still commute for business and will bring their cultural conflicts with them.
There's no reason for that in the decentralized space. As a side effect of my oft-repeated belief on HN that I'm not convinced that a centralized single massive community of everyone is even really possible, expressed well prior to Elon buying Twitter, I have no presence on any of the major social media sites. I have many communities. HN is merely one of them. Each of them has a different social contract. Many of them are mutually contradictory in that either the implicit social mores or explicit social rules would exclude each other if taken in their totality... this is less impressive or weird than it sounds, the moreso as you take more and more restrictive views of what constitutes a "match".
This is, in my opinion, normal. It reflects the reality I live in. What is mandatory and verboten at the grocery store, the strip joint, your kid's school concert, a college classroom, etc. all differ intrinsically, and there's no getting around that. The attempt to jam the entire Internet into one walled garden was always doomed to failure, by its very nature. I think that while there are always technical pulls to centralization, and marketing/capitalistic pulls to centralization, in the end, the social pressures that make it impossible to control everyone in one space will always win out. I expect we're in a multi-decade decentralization here... not necessarily to the extreme that everyone runs their own instance, sure. But the idea that we can all agree on a very small set of social websites, and we will all agree that Silicon Valley Liberalism is the One True Social Contract is an absurd illusion brought on by too much VC money, just one more victim of the end of 0% interest rates.
So as the Twitter culture shock spreads out, I would say, the next culture shock thing to break is the idea that you have "one" social media presence or that such a thing is necessarily possible. Prepare for the Internet to spin out and once again match the reality of the real world, with different spaces for different purposes. You may not be able to have "one" identity any more than you can live in one social space in the real world. Indeed, the entire idea is kind of silly in the end.
Key word being "hopefully". People either don't understand or care about the power differentials that these mega corps perpetuate and how it affects us, especially long term. People dislike Musk as a person, they'll be happier than ever jumping in bed with any other megalomaniac billionaire as long as they have the self-restraint to hide their cynical worldview.
This doesn't mean that federation, decentralization and privacy is dead in the water, but it does mean that the stakes are high, and you need to offer a better product than what ad-tech bloat gives you today, so it's certainly not impossible. You need critical mass, ease of use, technical simplicity and resilience against abuse, and then strike at a good opportunity for mass migration.
I don't think Mastodon today fulfills any of the criteria, let alone the core privacy features that are difficult to retrofit. I also don't think it ever will, because the OSS style governance model always leads to hard-to-use products and fractured communities whenever large and difficult decisions need to be made. (That doesn't mean that the fediverse can't be an excellent refuge for geeks and certain niches, which is great both for them and an incredibly useful lesson for future projects!)
Realistically, I think the Signal model is our best bet, given the world today. An ideologically convinced medium sized business/foundation built on OSS tech, with decently aligned incentives, but that respects UX enough to build a product that everyone can use.
There's a waiting list while the devs are frantically in the middle of actually writing it.
https://www.thewrap.com/what-is-post-twitter-alternative-new...
[1] https://archive.ph/xYeYY - TPM article without paywall
[2] https://techcrunch.com/2022/11/28/post-news-twitter-alternat...
[3] https://www.npr.org/2022/11/21/1138064606/elon-musk-twitter-...
Meanwhile in China...
I haven't followed anyone new, so this traffic is coming from people who stopped posting for a bit and then returned. Some of them I know are crossposting from mastodon, but not all of them.
Personally have been splitting time between Twitter & Mastodon, although using Twitter more. Mastodon has a higher percentage of my high quality follows but a lot of them either dont post much there or are crossposting to Twitter anyways. And I prefer Twitter the application a heckuva lot more than Mastodon.