The Mastodon centralists are coming out of the woodwork today with, “But it makes onboarding easier!”
So what?
The goal here isn’t to grow for the purpose of growth. Growth, in and of itself, is meaningless.
The goal is to decentralize social media.
like this
Mark reshared this.
Chris Trottier
in reply to Chris Trottier • • •There’s such a thing as growth by subtraction.
What I mean by this is that if you want Mastodon to grow, you must be cognizant that Mastodon can’t be all things to all people.
To me, the promise of Mastodon is decentralization. Therefore Mastodon shouldn’t be the service for centralists.
That is, people who are looking for an exact replica of Twitter—corporate structure and all—should probably not use Mastodon.
Chris Trottier
in reply to Chris Trottier • • •History shows that shovelling people into one server and then saying, “Let’s explain to them about decentralization later” doesn’t work.
Most people will never care about decentralization. They just want to chat with their friends.
This will become more and more true as Mastodon gains popularity.
So it’s better to force people into making a choice at the outset. Yes, it’s annoying and causes friction. Many people find making any decision whatsoever to be “difficult”.
And that’s fine.
Chris Trottier
in reply to Chris Trottier • • •People are now conflating centralization with “accessibility”.
No, these things are not one and the same.
Alt text on images is accessibility.
Defaulting everyone onto one server is not accessibility.
Chris Trottier
in reply to Chris Trottier • • •Centralization does not make Mastodon “accessible” to so-called “regular people”.
Centralization turns Mastodon into Tribel, Hive, and Post.
Centralization takes everything unique about Mastodon and destroys it.
Chris Trottier
in reply to Chris Trottier • • •Let me tell you, if centralization of Mastodon is your goal, then basically every microblogging platform that exists right now is blowing it out of the water!
You should probably sign up for those services instead.
m@thias.hellqui.st :verified-skull: likes this.
Chris Trottier
in reply to Chris Trottier • • •I am so scared that mastodon.social could be acquired that I have set up plenty of my own Fediverse servers—on multiple domains—just in case.
This isn’t paranoia. It’s the sane thing to do.
And if you can, you should probably do it too.
FediThing has moved!
in reply to Chris Trottier • • •Also, don't use the word "mastodon" in the domains, because it is owned by the same people as mastodon.social:
social.growyourown.services/@h…
If there was a sale of mastodon.social, the trademark might be sold with it, and they could start changing the terms of use for anyone using a domain with the word "mastodon" in it.
Grow Your Own Services 🌱
2023-04-25 08:54:27
Chris Trottier
in reply to FediThing has moved! • • •Chris Trottier
in reply to Chris Trottier • • •People seem to think this is about "free and open source" idealism.
But I don't care about "free and open source" if the net effect is centralization.
Google uses lots of "free and open source". It is a monopoly.
FediThing has moved!
in reply to Chris Trottier • • •We need open decentralised ownership as much as open decentralised software.
Ownership isn't decentralised if everyone's on one server.
eshep likes this.
Chris Trottier
in reply to FediThing has moved! • • •Chris Trottier
in reply to Chris Trottier • • •Ironically, the people who are in favour of Mastodon’s centralization are, themselves, “free and open source” idealists.
They believe the AGPL is a magic bullet to ward off everything that can go wrong with Mastodon.
If everyone arrives at mastodon.social, it’s okay—you have the source code!
Reddit was once free and open source too. It is now a centralized walled garden.
Jonathan Arnold
in reply to Chris Trottier • • •Chris Trottier
in reply to Chris Trottier • • •Decentralization has nothing to do with idealism.
It has everything to do with being clear-eyed and practical.
On the contrary, believing that Twitter is the “public square”—despite being owned by surveillance capitalists—that is idealism!
Kiudecan
in reply to Chris Trottier • • •poetaster
in reply to Chris Trottier • • •Paul Schoe
in reply to Chris Trottier • • •If a person is so committed to distributed social media that she/he sets up their own Feduverse server, what is the use, as you advise, of setting up multiple servers?
Isn't one server even better for one administrator to support a decentralized system?
Because with only one server, you can concentrate all your efforts, monitoring, marketing on one server, giving better results (more users) than when that attention is divided along several servers.
Pier'ce
in reply to Chris Trottier • • •nhan
in reply to Chris Trottier • • •Why haven't you moved away from mastodon.social? I mean, using your own servers and start posting there and stop posting here?
Currently to promote decentralization we can have a policy where you have to move off a server when you reach a certain number of followers?
The Cadence Collective
in reply to Chris Trottier • • •St. Paul :llama_2:
in reply to Chris Trottier • • •the decentralized nature of Mastodon is what really keeps me coming back. I started on a larger more general purpose server, but I eventually migrated over to shakedown, and that’s made all the difference.
I love that I’ve found a community of like minded music lovers, and I can still bring all of the other interesting voices I follow into my timeline together!
TankFox
in reply to Chris Trottier • • •I love that it takes effort to get here because it filters out 90% of the entire world.
People climb mountains because the conversation is better at the top. By and large the only people up there are skilled, respectful, smart, and actually want to be there a lot.
This is exactly the same reason I decided to go to college
eshep likes this.
Stanley Black-Decker
in reply to Chris Trottier • • •Andrew Woods
in reply to Chris Trottier • • •jan
Unknown parent • • •Why is it, that open source tech projects so often get lost in complexity and miss the usability focus?
Calm down and allow various perspectives.
Chris Trottier
in reply to jan • • •@jnbrgr Because this isn't about "open source" -- it's about decentralization.
Usability is fine when it serves the objective of decentralization.
Kristian
in reply to Chris Trottier • • •@Chris Trottier It isn't about decentralization per se. It's about a very limited understanding of decentralization which actually boils down to a late-1990s approach of "decentralization just by having smaller centralized systems connected to each other" - as in individual servers having accounts on it, accounts that are heavily tied to these servers and pretty hard to move elsewhere. We see these problems if we look at, in example, how incredibly complex it would be to support migrating posts from one Mastodon account to another. Personally, I am sure decentralization and usability aren't per se mutually exclusive. But the _current_ approach of decentralization and usability are (_because_ of that strong ties betweeen account and instance, and because of the importance choosing an instance has in this process), and if we want decentralization to succeed, we'd better strive for solutions that have sane decentralization irrevocably while _still_ being usable even to the most non-tech user you could possibly imagine. Right now we're doing pretty much the opposite.
@jan :mastodon:
Chris Trottier
in reply to Kristian • • •@z428 @jnbrgr Yes, it is. It's entirely about decentralization.
Usability is fine so long as it creates the conditions for decentralization.
If it doesn't serve the purposes of decentralization, "usability" doesn't matter. In fact, such "usability" efforts are unusable.
Kristian
in reply to Chris Trottier • • •@Chris Trottier Couldn't disagree more. At the very end, social networks are tools that should serve some sort of purpose, and they should first and foremost focus on meeting the functional requirements of this purpose. No one would use a system just because it's "decentralized" even while it doesn't get the job done, and actually we have a lengthy track record of centralized platforms outpacing decentralized ones simply because they worked for people (Slack vs IRC, WhatsApp vs XMPP, ...).
Decentralization, privacy, usability, ... are "just" nonfunctional requirements. They're all important, most likely they're equally important, but first and foremost it should be a system that works for users, otherwise it won't be used by a reasonable amount of people. Maybe decentralization is just poorly designed if it breaks usability at the moment.
@jan :mastodon:
Adam Dalliance
in reply to Kristian • • •@z428 @jnbrgr
Functional requirement number one: Nobody owns my data but me, nobody can change the system from under me, nobody can put unblockable adverts into my feed, nobody at all owns the network.
There are other requirements after that, but that is the prime directive in this federation.
Kristian
in reply to Adam Dalliance • • •@Adam Dalliance Why, in this case, not simply use some computer with your data stored in a markdown file offline? If it's just about owning your data in a way so no one can interfere with them, and if this is the only requirement, then why bother putting them online at all? 😉 That's what I mean: Despite all these (totally relevant and undisputable) requirements, isn't it first and foremost about making data available, visible to, interactable with for others?
@Chris Trottier @jan :mastodon:
Adam Dalliance
in reply to Kristian • • •@z428 @jnbrgr
I didn't say it was the only requirement, in fact I explicitly said there are other requirements too.
Just that the ownership of my data and the non-ownership of the network are the prime ones.
I'll get a Bluesky account one day probably, but not until I can run my own server.
Adam Dalliance
in reply to Adam Dalliance • • •I think most of my friends who are leaving Big Social are indeed doing so by just keeping their diary offline instead. Most of them aren't moving elsewhere, they're just quitting.
Maybe a few private Whatsapp groups that I'm not invited to.
Kristian
in reply to Adam Dalliance • • •@Adam Dalliance In my environment, I don't really see much people moving as of now. There's a crowd of enthusiasts leaving behind Twitter, Facebook, ... but these mostly are the folks that have been in many places already before, people that aren't hesitant to doing so. Most of especially my Twitter bubble is extremely robust, doesn't care much about Musk and is staying there without any second thought. That's where my motivations to get some of them "over here" just result in a "why", and I don't really have good answers satifying to them, at least in terms of advantages outweighing disadvantages / effort required for that. If "owning your data" just would actually be "owning your data" for everyone without being required to run a server, things would be quite a bit easier to communicate - at the moment, for most of these "owning your data" in practise means "trust some random person to host your data well" (either by running a Mastodon / Fediverse instance for you or by providi
... show more@Adam Dalliance In my environment, I don't really see much people moving as of now. There's a crowd of enthusiasts leaving behind Twitter, Facebook, ... but these mostly are the folks that have been in many places already before, people that aren't hesitant to doing so. Most of especially my Twitter bubble is extremely robust, doesn't care much about Musk and is staying there without any second thought. That's where my motivations to get some of them "over here" just result in a "why", and I don't really have good answers satifying to them, at least in terms of advantages outweighing disadvantages / effort required for that. If "owning your data" just would actually be "owning your data" for everyone without being required to run a server, things would be quite a bit easier to communicate - at the moment, for most of these "owning your data" in practise means "trust some random person to host your data well" (either by running a Mastodon / Fediverse instance for you or by providing infrastructure for you to self-host your Mastodon/Fediverse instance). This really feels flawed in terms of "decentralization".
@Chris Trottier @jan :mastodon:
Chris Trottier
in reply to Kristian • • •Chris Trottier
Unknown parent • • •@jnbrgr @z428 @pre On the contrary, pushing growth for growth’s sake is the tail wagging the dog.
What does it matter if everyone uses Mastodon if Elon Musk, or someone like him, ends up owning it?
I’d rather use a social network that no one uses and owned by no one than a social network used by everyone and owned by one rich asshole.
In fact, that’s why I used Mastodon for many years!
Chris Trottier
Unknown parent • • •@z428 @pre @jnbrgr Nonsense. They can move to email. They can pick up a phone. They can send a text message. They can decide to simply not help surveillance capitalism.
If they want social media that can’t be bought by Elon Musk, well, time to learn about the benefits of decentralization.
canleaf08 ⌘ ✅
in reply to Chris Trottier • • •Kristian
in reply to Chris Trottier • • •@Chris Trottier Then why discuss this at all? The decentralized Fediverse will remain just what it used to be before Musks Twitter. Maybe Musk at some point will buy a bigger instance which will render this _instance_ more "centralized" again. Why bother? It won't change anything for you and the way you want to use Mastodon, would it? You still can run your own instance, you can block the large ones, you can move your account elsewhere and take your contacts with you. Why not simply ignore it? Why argue against how other use this system, then...? (No offense intended - I really don't understand it at this point...)
@Adam Dalliance @jan :mastodon:
Chris Trottier
in reply to Kristian • • •jan
Unknown parent • • •The maximum I can accept is an approach like the default maximum users setting on Pixelfed. And even this is VERY questionable as any instance can simply close for new registrations when THIS instance decides to do this.
Welcome culture & maximum freedom to co-exist with different approaches is essential!
Chris Trottier
in reply to jan • • •@jnbrgr Inevitably, I would love for everyone to run their own server, or use social media on a peer-to-peer basis.
To me, the Fediverse is a stop gap.
Jiří Fiala Total Landscaping
in reply to Chris Trottier • • •Henry
in reply to Chris Trottier • • •this is the thing, yes centralisation does make it easier for newcomers but it also fundamentally changes Mastodon that would be a departure from the original intention.
Centralisation in and of itself isn’t necessarily bad, but Mastodon is not meant to be a straight replacement for Twitter, instead it’s its’ own thing. The lack of centralisation may not suit everyone but that’s ok, there are other options out there.
I’m getting used to Mastodon now and finding it valuable to me
heapwolf
in reply to Chris Trottier • • •GitHub - socketsupply/socket: The simplest way to build native desktop & mobile apps for any OS using your favorite frontend libraries. Apps can even communicate directly with Modern P2P, no cloud required!
GitHubChris Trottier
in reply to heapwolf • • •DamnKimberlee
in reply to Chris Trottier • • •eshep
in reply to Chris Trottier • •Honest, curious question; not being a pedantic dickhead.
In this thread whenever "mastadon" is mentioned by someone, is it in direct reference to Mastadon, or referring the fediverse in general?
I think the synonyminization of mastodon/fediverse is part (most?) of the reason for this sorta thing.
m@thias.hellqui.st :verified-skull: likes this.
Matthias Bohlen
in reply to Chris Trottier • • •Yep, totally true. But I think the entire fediverse should have search. I cannot find my friends on Mastodon, except when they are on mastodon.social.
That's something that is truly lacking: discovery. People can opt-out if they don't want to be discovered.
Kristian likes this.
rudiev 🏳️⚧️🏳️🌈🇯🇴
in reply to Chris Trottier • • •eshep
Unknown parent • •m@thias.hellqui.st :verified-skull: likes this.
Kristian
in reply to Chris Trottier • • •LegalQuilts
in reply to Kristian • • •@z428 the first thing that needs to be done is talk to NON-tech users.
We had this discussion a bit ago. I remember someone went into the whole detailed idea of how people choose an email server. It was a very techy answer. How non-tech people choose a server is "Hey my friend uses gmail, guess I will go with that." Same thing with the Fediverse. How do non-tech people really thing, not how tech people think non-tech should think.
Chris Trottier
in reply to Kristian • • •@z428 The goal is to build social media that cannot be bought or sold by one entity or person such as Elon Musk.
And how to achieve that? Through decentralization.
How do we not achieve that? By centralizing Mastodon onto one server so that it can be acquired by an Elon Musk.
Kristian
in reply to Chris Trottier • • •Chris Trottier
in reply to Kristian • • •Your push for centralization ignores the elephant in the room: what does it matter if everyone leaves Twitter when Elon Musk could theoretically buy mastodon.social and now own a significant part of the Fediverse?
Is “easy onboarding” worth that very real possibility?
christopher :rebel: 38C3
in reply to Chris Trottier • • •@z428 because the protocol allows people to move to a new server.
you're saying all this pro federation subjective feelings stuff while using the largest server of them all.
Chris Trottier
in reply to christopher :rebel: 38C3 • • •@yawnbox @z428 Yes, I use mastodon.social. I also use peerverse.space, atomicpoet.org, and calckey.social.
I’m in multiple places being active. And when I’m on mastodon.social, my purpose is to talk about the importance of the Fediverse.
nhan
in reply to Chris Trottier • • •@yawnbox @z428 Why don't you leave mastodon.social? Have 1 post like this as an announcement, then move to a smaller server, redirect your followers there, and use that account to answer questions/replies of others. And just stop posting/commenting here.
You can make a new account in a new, very small server, and posting about the importance of the Fediverse. Aren't we proud that we can talk to anyone regardless of the servers? You can be a prime example yourself.
Kristian likes this.
Chris Trottier
in reply to nhan • • •I have a purpose on being on mastodon.social, just like I have a purpose on being on calckey.social.
See, the great thing about the Fediverse is that I can be in many places at once.
nhan
in reply to Chris Trottier • • •Chris Trottier
in reply to nhan • • •@nhan @yawnbox @z428 Leaving on the Fediverse is not easy. In fact, for accounts that have my amount of followers, it’s expensive. There’s the further matter of mastodon.social being the most federated server on the Fediverse. And there’s also the matter of if I were to leave mastodon.social, it would be to relocate on my own server.
But I also prefer atomicpoet.org to have another purpose.
nhan
in reply to nhan • • •Chris Trottier
in reply to nhan • • •@nhan @yawnbox @z428 I also run calckey.social, calckey.art, peerverse.space, vancity.social, atomicpoet.org—and others.
I’m doing my part for decentralization.
Kristian
Unknown parent • • •GhostOnTheHalfShell
in reply to Kristian • • •Mastodon’s on-boarding should take these unique features into account. Because pushing everything into one server also seems to be making a problem of its own.
Instances are staffed and funded by individual and groups and that essence needs to be made apparent. Servers at capacity should be flagged or cataloged as such; maybe what is needed is a way to help people spawn new ones as a consequence (hint. hint.)
so a mechanism to draw people in an help generate mastosprogs..
Kristian
in reply to GhostOnTheHalfShell • • •eshep
Unknown parent • •@GhostOnTheHalfShell, couldn't the same problem be solved simpler and on a larger scale by introducing people to the fediverse vs introducing them to mastadon?
Maybe instead of advertising "joinmastodon.org" and arguing about why/how it needs to change, maybe point people to places like the following:
fediverse.party
fediverse.observer
fedidb.org/software
Each of these sites provides them with different styles of useful information about why Mastodon would be a better fit for them than Misskey or Pleroma, or why they may enjoy Diaspora over an ActivityPub option. Maybe Lemmy is the one that stands out for them.
Does presenting folks with alternate mastodon instances not create a similar problem to presenting them with a single one, given there are many other options to choose from outside of mastodon?
Bianca
in reply to Chris Trottier • • •Joe
in reply to Chris Trottier • • •aardvark
in reply to Chris Trottier • • •caelanbp
in reply to Chris Trottier • • •Chris Trottier
Unknown parent • • •@arinbasu1 @yawnbox @z428 @nhan It’s not that I don’t want people to join mastodon.social, it’s that I don’t want this to be the default server.
If people are choosing to come here after being shown numerous options, then good.
eshep likes this.
Arindam Basu
in reply to Chris Trottier • • •This is merely a thought “experiment”, how might it be if we had something like “calckey.social” back in this time last year, when many people were leaving twitter in search of better alternatives, and we had people coming over to Mastodon but then ended up either not using or leaving because some of them found Mastodon to be confusing to use, and anytime people brought up the issue of Fediverse, these users would say Fediverse is too confusing (conflating Fediverse with Mastodon).
In hindsight, seems to me that if they discovered the power of #Fediverse with something like calckey.social, they would have stayed on. Sadly, although there was pleroma/misskey/calckey/foundkey, none of these gained the kind of user base that’d have been helpful to “spread” the fediverse around, 🙂 Mastodon is great but there is definitely room for others.
Calckey Social
Calckey Sociallike this
Kristian, eshep and Mark like this.
Mark reshared this.
Kristian
in reply to Arindam Basu • • •@Chris Trottier @christopher :verified_gay: @nhan
eshep
in reply to Arindam Basu • •Kristian likes this.
Kristian
in reply to eshep • • •@eshep The problem gets worse however in what isn't visible at first sight: Talking e-mail as the "classic" decentralized system, yes, there used to be a lot of different providers and clients and user interfaces but in the end you could be reasonably sure that, at the very end, sending, receiving and displaying mails would work well for all of these and between all of these. In 2020s Fediverse, things are slightly more ... complicated, knowing that how federation, commenting, notifications, ... works will more or less massively depend on which platform you are on and which platforms your contacts use. Like, I repeatedly am experiencing interesting issues (as in lost comments, lost messages, lost notifications when communicating between pixelfed and friendica). This is something you only notice once you've been into things for a while, and at this point reverting a change and moving elsewhere will become more difficult. (We're not talking about different instances blocking each oth
... show more@eshep The problem gets worse however in what isn't visible at first sight: Talking e-mail as the "classic" decentralized system, yes, there used to be a lot of different providers and clients and user interfaces but in the end you could be reasonably sure that, at the very end, sending, receiving and displaying mails would work well for all of these and between all of these. In 2020s Fediverse, things are slightly more ... complicated, knowing that how federation, commenting, notifications, ... works will more or less massively depend on which platform you are on and which platforms your contacts use. Like, I repeatedly am experiencing interesting issues (as in lost comments, lost messages, lost notifications when communicating between pixelfed and friendica). This is something you only notice once you've been into things for a while, and at this point reverting a change and moving elsewhere will become more difficult. (We're not talking about different instances blocking each other more or less intransparently now.) Even at this very "high-level" point of view, there's loads of room for usability improvements. 🙂️
@Chris Trottier @christopher :verified_gay: @Arindam Basu @nhan
eshep likes this.
joel b
in reply to Kristian • • •good thing email always works %100 of the time and doesn’t suffer from spam filters. I never have to deal with tickets where a user claims he/she never got their confirmation email despite my server logs showing it delivered. Corporate run networks are entirely transparent about moderation and the appeals process is widely known to be robust and fair. /s
give me a break
Kristian
in reply to joel b • • •@joel b I've been doing that stuff (handling corporate mail for a mid-sized organization) since the late 1990s and yes, there's a load of things that just could break here as well. That's why I didn't write "it works 100%" but rather "you can be reasonably sure it works" (and I _mean_ exactly that). 😉️ With e-mail I've seen interesting effects too, but all these years I rarely stumbled across situations where things got completely lost and no one had a clue where or how.
@Chris Trottier @christopher :verified_gay: @eshep @Arindam Basu @nhan
eshep likes this.
joel b
in reply to Kristian • • •I wish I shared your experience. Where do you find these users who are not clueless?
I can say in my ~6 months of running my own pleroma instance as a hobby that I have not noticed any trouble in my message delivery.
Kristian
in reply to joel b • • •@joel b There never was a shortage in clueless users over here too. Maybe our load wasn't that heavy though, just had around 7..9k of messages per day, most of them (document / drawing management for civil engineering) with larger documents, xls sheets, ... attached. Even this way, real serious issues (as in mails lost without being able to figure out where they ended up) have been rare, even knowing a lot of the external participants were somewhat low-tech back then, utilized mail systems such as provided by their dialup ISPs, or GMX / Yahoo in worst cases. Content blocking always was an issue, but it rarely managed to have messages lost. And except for a very small couple of contacts sending the dreaded winmail.dat stuff, we never saw messages we weren't unable to read or handle correctly.
... show moreOn the Fediverse (okay, taken I don't run my own instance) I've seen all too often situations where a comment on a post, or a comment on a comment on a post, would not show up on t
@joel b There never was a shortage in clueless users over here too. Maybe our load wasn't that heavy though, just had around 7..9k of messages per day, most of them (document / drawing management for civil engineering) with larger documents, xls sheets, ... attached. Even this way, real serious issues (as in mails lost without being able to figure out where they ended up) have been rare, even knowing a lot of the external participants were somewhat low-tech back then, utilized mail systems such as provided by their dialup ISPs, or GMX / Yahoo in worst cases. Content blocking always was an issue, but it rarely managed to have messages lost. And except for a very small couple of contacts sending the dreaded winmail.dat stuff, we never saw messages we weren't unable to read or handle correctly.
On the Fediverse (okay, taken I don't run my own instance) I've seen all too often situations where a comment on a post, or a comment on a comment on a post, would not show up on the target system, in some cases however be visible to others on other system (unless posted privately) and often spent hours in communicating with various actors (including instance admins and partially developers) to learn what went wrong. Same, until just a while ago, Friendica wasn't able to render Mastodon polls so these posts in Friendica would just show up as text posts and unless the textual part of it suggested there might be more you wouldn't even notice you're missing something. Or, in example, handling pictures: I can use pixelfed and Friendica to post messages with up to ten (pixelfed) or unlimited (? Friendica) amount of images attached but Mastodon generally will only display four of them, again without leaving a hint that there's more. Or pixelfed stories (which are absolutely unusable outside the pixelfed world). E-mail had a lot of mess in terms of formats and structures too but I never experienced this kind of chaos with e-mail throughout all these years, the worst to happen in this really was TNEF / winmail.dat or a message claiming to please install an HTML compliant mail client to read the content. I've never seen, in example, a message just partially rendered on some system. This is what is hard to impossible to explain to new users without driving them away (and it sheds a difficult light on all these images claiming the Fediverse is great because everything connects to everything...).
@Chris Trottier @christopher :verified_gay: @eshep @Arindam Basu @nhan
joel b
in reply to Kristian • • •fair points, all of them. I suggest much of this has to do with Mastodon being a flagship product but not always meeting the bar when it comes to interoperability. I will not speculate on reasons for that. It should be noted that I run pleroma, not mastodon.
For what it is worth I don’t do a lot of media heavy posting aside from Simpsons screenshots as snarky replies. They always seem to get thru. I just don’t have many followers so the videos I do post are not widely viewed. I can hotlink them to my mother in an SMS message and she can always see them so that’s my biggest use case covered.
Dropping 12 images into a post that may exceed 50mb is going to be a lot to process for any system. I am not familiar with Pixelfed as a user but I would assume they must offer a crazy amount of server side processing to serve low resolution previews and follow up with high resolution on request. I find without an algorithm to keep everyone’s funnel topped off the content on Activity Pub is very bursty.
The federated nature does exponentially expand ban
... show morefair points, all of them. I suggest much of this has to do with Mastodon being a flagship product but not always meeting the bar when it comes to interoperability. I will not speculate on reasons for that. It should be noted that I run pleroma, not mastodon.
For what it is worth I don’t do a lot of media heavy posting aside from Simpsons screenshots as snarky replies. They always seem to get thru. I just don’t have many followers so the videos I do post are not widely viewed. I can hotlink them to my mother in an SMS message and she can always see them so that’s my biggest use case covered.
Dropping 12 images into a post that may exceed 50mb is going to be a lot to process for any system. I am not familiar with Pixelfed as a user but I would assume they must offer a crazy amount of server side processing to serve low resolution previews and follow up with high resolution on request. I find without an algorithm to keep everyone’s funnel topped off the content on Activity Pub is very bursty.
The federated nature does exponentially expand bandwidth with scale because each activity must be published among each instance. Yes this means things can be undefined while the system reaches equilibrium and centralization can solve that. I just don’t think that benefit is worth the drawbacks of a centralized ecosystem over all. Lack of speed can always be overcome, in the worst case, by patience.
Kristian
in reply to joel b • • •@joel b Ah no, I'm not really suggesting a centralized system here, and I'm on the Fediverse after all because most likely it's the best we have at the moment. But then again, there has been plenty of well-worded criticism on the shortcomings of ActivityPub (like this one medium.com/@denschub/activityp… ). What bothers me a bit is: There's a lot of open aspects - in terms of standards quality / strictness (extensibility is evil and might be opposite to standardization?), handling technical issues (storage, bandwidth, ...) and so forth. And, which is worse: I think in order to address such issues (jwz.org/blog/2022/11/mastodon-…), the current decentralization is not really able to handle that well because decentrali
... show more@joel b Ah no, I'm not really suggesting a centralized system here, and I'm on the Fediverse after all because most likely it's the best we have at the moment. But then again, there has been plenty of well-worded criticism on the shortcomings of ActivityPub (like this one medium.com/@denschub/activityp… ). What bothers me a bit is: There's a lot of open aspects - in terms of standards quality / strictness (extensibility is evil and might be opposite to standardization?), handling technical issues (storage, bandwidth, ...) and so forth. And, which is worse: I think in order to address such issues (jwz.org/blog/2022/11/mastodon-…), the current decentralization is not really able to handle that well because decentralization means responsibility for handling these aspects is spread out all across a plethora of smaller and smallest actors who have neither incentives nor (at least in some cases) motivations to share such issues in a joint effort. This is a bit concerning (even though at this point we're way deeper than "just" user interface glitches or inconveniences in system communication. But I still feel the Fediverse has quite some way to go here, and maybe ... also in a way of getting organized better, of trying to figure out how to handle bandwidth, storage, ... in a sane way globally, not just limited to particular instances.
@Chris Trottier @christopher :verified_gay: @eshep @Arindam Basu @nhan
joel b
in reply to Kristian • • •Kristian likes this.
Chris Trottier
in reply to Kristian • • •@z428 @yawnbox @eshep @arinbasu1 @nhan @skotchygut I don’t know how anyone can agree with Dennis Schubert when Diaspora is effectively dead. Diaspora has had 13 years to validate their approach, and it clearly hasn’t worked.
If Diaspora integrated with ActivityPub, it would be alive and well instead of another project with wasted potential.
Kristian
in reply to joel b • • •@joel b Well, my personal opinion here is maybe too optimistic, yet: When I joined the FLOSS bandwagon in the 1990s, with (desktop) Linux gaining speed and projects such as GNOME and KDE emerging, there was an interesting spirit in that whole community, something like "by collaboration rather than competition we'd even be able to beat competitors like Microsoft". And ... in a way I think it will require that mindset and it will fail without it. 😔 And I wonder where's the actual problem, to be honest. Why not come up with a _strict_ standard, a standard that is not extensible, a standard that tries to get it right rather than being as loosely-defined as somehow possible for the sake of getting something done that is ready for end-users, ready to beat the ____ out of Twitter and Facebook literally by tomorrow because it's a joint effort by the brightest and most inspired minds out there? Why not doing this instead of coming up with yet another fork of a fork of a pr
... show more@joel b Well, my personal opinion here is maybe too optimistic, yet: When I joined the FLOSS bandwagon in the 1990s, with (desktop) Linux gaining speed and projects such as GNOME and KDE emerging, there was an interesting spirit in that whole community, something like "by collaboration rather than competition we'd even be able to beat competitors like Microsoft". And ... in a way I think it will require that mindset and it will fail without it. 😔 And I wonder where's the actual problem, to be honest. Why not come up with a _strict_ standard, a standard that is not extensible, a standard that tries to get it right rather than being as loosely-defined as somehow possible for the sake of getting something done that is ready for end-users, ready to beat the ____ out of Twitter and Facebook literally by tomorrow because it's a joint effort by the brightest and most inspired minds out there? Why not doing this instead of coming up with yet another fork of a fork of a predominant platform just because we disagree with the original maintainer about post length or that wording on the "Toot" button? Or the other way 'round: What could be a ... motivation or driver to actually _get_ the community together to sort these things out? For where we are now (and loosely getting back to the root of this post), I guess it's just a matter of time for some shady actor to throw up a solution that works for users, provides a stable environment, has a predictable behaviour and will leave the Fediverse again where it used to be before Mastodon and Musks Twitter: A cool playground for tech-savvy and privacy-aware people yet not something of interest to many others. That would be sad. But right now I don't at all see us being "organized" enough to avoid that.😔
@Chris Trottier @christopher :verified_gay: @eshep @Arindam Basu @nhan
Kristian
in reply to Chris Trottier • • •@Chris Trottier ... yes, because it's always the great solutions that dominate while the worse ones die away. That's why we're all using Linux desktops on our corporate computers and smartphones. Oh. Wait. ... . 😶
(Less sarcasm, more cynical though: There's a two-fold bad feeling about ActivityPub in me. The one part of it is it mostly being a _web_ standard. That could be a reason why it seems to blatantly ignore some aspects of reliabilty in delivery, things that have been implemented and defined way better in other protocols like SMTP ages ago. And the other part is that "extensibility". I've been into various kinds of software development since the mid-1990s, and if there's one thing in Dennis' outline I would sign off without second thought, then it's the approach that a server _must_ accept all data sent by some remote system without any validation or without any idea to check which kind of format has been sent over. This is as weird as it probably
... show more@Chris Trottier ... yes, because it's always the great solutions that dominate while the worse ones die away. That's why we're all using Linux desktops on our corporate computers and smartphones. Oh. Wait. ... . 😶
(Less sarcasm, more cynical though: There's a two-fold bad feeling about ActivityPub in me. The one part of it is it mostly being a _web_ standard. That could be a reason why it seems to blatantly ignore some aspects of reliabilty in delivery, things that have been implemented and defined way better in other protocols like SMTP ages ago. And the other part is that "extensibility". I've been into various kinds of software development since the mid-1990s, and if there's one thing in Dennis' outline I would sign off without second thought, then it's the approach that a server _must_ accept all data sent by some remote system without any validation or without any idea to check which kind of format has been sent over. This is as weird as it probably gets and I can only guess which kind of mindset is driving such an approach in a specification - asides maybe the desire of wanting to do a standard but not completely in a way that feels restrictive. In the post, he has this example of ...
"(1) Sorry, Alice, Bob is using software that is not compatible with us, so you can’t communicate with Bob here.
(2)Yes, you can communicate with Bob, but since he is using ExampleNet, please be aware that Bob will not receive your photo albums and will be unable to interact with those. Carol will see your photos, though, but unfortunately, she will not be able to see your geo-location updates. Moreover, because of technical limitations, Dan can comment on your posts, but we cannot make sure that Carol and Bob see those, because we cannot redistribute Dan’s comments."
... and I'd be glad to see you using Diasporas demise for explaining how this is wrong, assuming this is what I read in my timeline literally every day (in example talking about Mastodons dreaded 500 chars post limit, talking about picture attachments, talking about "Misskey Flavoured Markdown" that is unusable literally anywhere else except for Misskey and its forks).
@christopher :verified_gay: @eshep @Arindam Basu @nhan @joel b
Chris Trottier
in reply to Kristian • • •The approach your talking about has already been tried. That’s the approach Diaspora took. Diaspora failed.
And this is because the approach that you’re talking about doesn’t allow for new, novel, and innovative approaches to social media.
People don’t want mere clones. They want something fresh and new.
joel b
in reply to Kristian • • •I started my FOSS adventures in 1997 as soon as I could get my own computer that wasn’t shared by the rest of the family. I also remember those heady days though I was still quite young.
To attempt to answer your question I think such a standard will, in time, emerge from what we build today. It’s very difficult to nail the right design tradeoffs that will work for everyone. Far more so when we don’t get the benefit of a benevolent dictator who can ram thru a solution, for better or for worse.
I am okay with this tortoise and hare race because I know the closed systems will always disappoint and collapse around the egos of those who attempt to dominate them. It takes a village to raise a protocol.
Kristian likes this.
Kristian
in reply to Chris Trottier • • •(And no - it's not the approach Diaspora took. I'd dare to say this is an approach that hasn't been tried so far, mostly because - at least by now - it seems if there's one thing people are mostly unable to, then it's uniting behind _one_ common idea and trying to make that as good as possible by finding a consensus on how it should work. Ever checked how many of these open social networks started all along with or shortly after Diaspora? And in the end, most of them were, most of them still are clones by today. pixelfed is mostly a distributed Instagram clone. Mastodon is a federated... show more
(And no - it's not the approach Diaspora took. I'd dare to say this is an approach that hasn't been tried so far, mostly because - at least by now - it seems if there's one thing people are mostly unable to, then it's uniting behind _one_ common idea and trying to make that as good as possible by finding a consensus on how it should work. Ever checked how many of these open social networks started all along with or shortly after Diaspora? And in the end, most of them were, most of them still are clones by today. pixelfed is mostly a distributed Instagram clone. Mastodon is a federated Twitter clone. And so forth. Care to name _any_ service in the current Fediverse that is completely unique from what it does - its federated nature aside for a moment? Most of them are even more explicitely labeled as such - a "decentralized version of <you-name-it>".)
Chris Trottier
in reply to Kristian • • •@z428 The Diaspora approach is wrong because, at the end of the day, they talk to no other social network except for Friendica—and even that was met with outright hostility by Diaspora developers. Mike Macgirvin, who began Friendica, can tell you all about the numerous conflicts with Diaspora regarding interoperability with a protocol they made.
But mostly, it’s wrong because Diaspora has painted themselves into a corner. The service is stuck in an old Facebook paradigm.
Chris Trottier
in reply to Kristian • • •@z428 I don’t know where you’re hanging out, but I’m seeing lots of innovation on the Fediverse. Misskey has a radically different feature set from everything else. Bonfire is likewise excellent. As slow as Mastodon is in development sometimes, the fact I can edit my posts, then track those edits is fantastic.
None of this would be possible with Dennis Schubert’s approach to development.
Diaspora once had 650,000 accounts. Hardly any are active anymore. Why is that?
Kristian
in reply to joel b • • •@joel b I really would love to share your optimism here, but .... I have yet to see it. Two examples:
(1) Social networks. I've been on Twitter at least since 2010, Facebook since 2009, Flickr since 2005. In all of these (well except for Facebook, recently) within somewhat "grown" communities, some of them have been around for as long as that. Hard to say how many "alternative" social networks have been around, came and went by in the meantime. Identica. GNUSocial. Early Friendica, for a while. movim. Diaspora. The list is long. In most cases, we didn't talk about compatible evolution. In most cases it was all about like "oh you're still using this old <___>? here's where the music is." Ah yes, and of course all the new systems never cared about being compatible with the old ones, because why bother (see Diaspora vs ActivityPub for that). So in the end, while in > 15 years the proprietary platforms managed to provide a working environment to a vast load of
... show more@joel b I really would love to share your optimism here, but .... I have yet to see it. Two examples:
(1) Social networks. I've been on Twitter at least since 2010, Facebook since 2009, Flickr since 2005. In all of these (well except for Facebook, recently) within somewhat "grown" communities, some of them have been around for as long as that. Hard to say how many "alternative" social networks have been around, came and went by in the meantime. Identica. GNUSocial. Early Friendica, for a while. movim. Diaspora. The list is long. In most cases, we didn't talk about compatible evolution. In most cases it was all about like "oh you're still using this old <___>? here's where the music is." Ah yes, and of course all the new systems never cared about being compatible with the old ones, because why bother (see Diaspora vs ActivityPub for that). So in the end, while in > 15 years the proprietary platforms managed to provide a working environment to a vast load of people, however nasty the systems are, the alternative systems provided a fragmented, short-lived zoo of technical endeavours. I'd really love to believe ActivityPub is here to fix that, but I have serious doubts.
(2) Linux desktop. What was the first "Year Of The Linux desktop"? 2000? 2001? Every year since then? We had plenty of opportunities to provide alternatives to Microsoft: Windows Vista. Activation stuff in Windows 7 at least. Ads in Windows 11. Still, we're missing crucial parts required to, in example, make Linux systems a viable competitor for most business applications. Still, we have a plethora of half-baked, partially incompatible desktop environments, distributions, variations of distributions, ... . Meanwhile, Apple has a closely tied integration of smartphone and computer, and Microsoft has a desktop that can be fully provisioned from within Azure, while on Linux I still then and now struggle to mirror my smartphone notifications to my XFCE desktop or to transfer files between both devices without the cloud or an USB-C wire.
So, in both cases: Yes, I think building and crafting standards takes time and patience. But it also takes something else - commitment. And in too many situations, right now, maybe I'm too pessimistic to see that. It all too much just feels like the "alternative technologies" are just a playing field for people wanting to try out things and do tech stuff for the sake of it, rather than aiming at more ambitious goals.😔
@Chris Trottier @christopher :verified_gay: @eshep @Arindam Basu @nhan
Kristian
in reply to Chris Trottier • • •Kristian
in reply to Chris Trottier • • •@Chris Trottier Ok, if Misskey and Bonfire (both are good platforms) are "innovations", we don't need to any further discuss that I guess. Editable posts have been around in other platforms (open ones too) for quite a while as well... .
As for Diaspora. The core issue I see here is that essentially at some point things didn't move forth anymore because developers didn't care about working on / contributing to Diaspora anymore. It's always much more enticing to start something anew than working on an "old" platform (as in "a platform built and driven by someone else"). And this is one of my bigger gripes with all this situation.
Chris Trottier
in reply to Kristian • • •@z428 Of course things can be done better but have you ever heard the phrase “Perfect is the enemy of good?”
Well, that applies to decentralized social protocols too.
Diaspora is so focused on everything being “just so” that they won’t talk to anyone. And that’s a problem.
A protocol isn’t merely about defining what stuff does. More importantly, it’s a gentleman’s agreement that delivery systems will be on good behaviour when a message gets sent.
Not ideal, but good.
joel b
in reply to Kristian • • •All of those big networks fall short though. Most of the people I want to talk to have left them long ago. I am 40 years old this year and the only system I can find most of the people I want to talk to on is the telephone network. Money and influence can keep a system going long after it has lost its intrinsic value but eventually the money dries up and there is nothing to keep it going. I deleted my 15 year old Facebook account when I realized no one was left there that I wanted to talk to.
Apple has not really innovated in over 10 years. The only thing I have from their ecosystem is an iphone and it integrates terribly with everything else in my life. I have to use MS Azure for some of my job and it’s easily 10x slower when I have to iterate development and I have made my managers keenly aware of this. We are only using them as far as we must. Meanwhile linux has continued to gain market share. The other senior devs and I sneak it into the business at any opportunity and it works very well.
Communities, like all things in nature will grow, bloom, and die. Just
... show moreAll of those big networks fall short though. Most of the people I want to talk to have left them long ago. I am 40 years old this year and the only system I can find most of the people I want to talk to on is the telephone network. Money and influence can keep a system going long after it has lost its intrinsic value but eventually the money dries up and there is nothing to keep it going. I deleted my 15 year old Facebook account when I realized no one was left there that I wanted to talk to.
Apple has not really innovated in over 10 years. The only thing I have from their ecosystem is an iphone and it integrates terribly with everything else in my life. I have to use MS Azure for some of my job and it’s easily 10x slower when I have to iterate development and I have made my managers keenly aware of this. We are only using them as far as we must. Meanwhile linux has continued to gain market share. The other senior devs and I sneak it into the business at any opportunity and it works very well.
Communities, like all things in nature will grow, bloom, and die. Just because the open-source attempts follow this same rule but do not have the billions of dollars and marketers behind them does not indicate they are failures nor that the capitalist networks are immune.
Each day we must renew ourselves and our world.
Chris Trottier
in reply to Kristian • • •@z428 Friendica never had the install base that Diaspora had. It never had the funding either. Or even the press.
And yet, despite initially being made in 2010, Friendica has enthusiasm *right now* because they have a fundamentally different approach to protocols from Diaspora.
Friendica will talk to anyone. Friendica uses its own protocol, DRFN. But it also uses Diaspora and ActivityPub.
Friendica is happy to talk to everyone. Diaspora is not. That’s the difference.
Chris Trottier
in reply to Chris Trottier • • •@z428 You allude to the fact that developers aren’t excited about Diaspora anymore. And that’s because developers don’t want to work on something that insists on everything being “just so”.
Developers are excited about working on Friendica, though. You can use it with Fedilab. It just got added to @tootsdk.
Both Diaspora and Friendica are relative the same age. One of them follows your preferred approach, and the other doesn’t.