Interview with Bluesky CEO Jay Graber
theverge.com
theverge.com
Until the decentralisation experience becomes AOL simple, it will remain a backwater.
I'm not familiar with the ins and outs of decentralization yet, but I can totally see this being an angle companies to take. Especially as data privacy becomes even more important to people.
maybe it's a different story with activitypub and bluesky's AT protocol, but it's not as if decentralized social protocols are a new innovation, the market just isn't interested in "privacy" and "control" - those that do have already discovered they can get by without social media altogether
(I say this as someone building a self hosted home server/appliance, but I think i'll promise more of a fireproof safe - not for keeping secrets, but for making sure you don't lose essential documents / family photos, and a home assistant that doesn't shout "I CANT CONNECT TO THE WIFI" when I ask it to switch a lamp off)
I think it could be aimed well at something like the livestreaming/Twitch/YT community, and build out from that base.
Thanks for sharing Freedombox too, I've never heard of them before.
The problem emerges when companies need to monetize the hosting and delivery of the content. It’s 100% a business model challenge, not a technical one.
if i could open a screen share session with a similar easy shortcut, and broadcast whatever i have open on my screen with my friends, regardless of where the pixels originated, with a full duplex link i could talk over / see my friends with, i think i'd have much more meaningful evenings
"can you hear me? I can hear you but not the game? ok now i can hear the game but not you? ok now i can hear the game bleeding through your mic? did you set up a virtual audio cable? is your mic patched through correctly?" every single time.
sometimes i just want to scroll reddit with friends, or watch a part of a dvd with friends, or play pubg mobile with friends, but i dont want to break out a display splitter and ask them to log in to a centralized service and accept an endless barrage of cookie banners and "Do you accept our new terms?" checkboxes and watch ads just to watch some memes with me (after i spend 30 minutes figuring out how their API changed so i can keep streaming the way i figured out how to already do it without issues just last week)
why cant i just send my pixels to my friends via bittorrent or some other protocol
Back in the day Teamspeak or mumble were hassle free.
teamspeak and mumble were definitely easier to start up and use once you get up and running the first time, i think i remember having to either install drivers or something else to get stuff like global push-to-speak and other core features to work
In my experience, Discord is no more or less buggy than Teamspeak or Mumble
We already have a better working model for this; it's HOW WE DO EMAIL. On one hand, email is a very strong "centralized identity" thing, and on the other it's not absolute. Yes, it has problems, but as twitter has shown -- deeply centralizing the thing has MORE problems.
The main driver behind current social media's shittiness is the user-hostile business model that means the incentives of the user and the platform are opposed.
Federation doesn't actually address any of this, and currently the federated social media world "runs" on effectively charity. "Runs" in quotes because we've already seen instance shutdowns which caused a fair bit of trouble. It only currently "runs" because the scale is small enough that it can indeed be sustained by charitable efforts, and hasn't devolved into a unusable spam-fest because it's too small to actually be worth spamming.
If your "solution" to social media shittiness is effectively turn it into something ran by charities, that's fine but you can achieve this much easier (and more efficiently) by just having one charity run one centralized service with an open API (since the business incentives around lock-in are no longer applicable).
Some of those communities might go on and last a while and accumulate knowledge.
Ah maybe not, sounds like a path the gamergate stuff. But the concept certainly feels nostalgic.
And besides now I have a job and a small amount of extra cash to pay for a small community website.
Would be nice to have federated reddits instead of reddit.
Maybe it's the opposite, a community like reddit if You're banned or maybe shamed or even down voted for an ignorant take, not even a bad faith argument, maybe that user stays on Reddit but joins a different subreddit that new subreddit may be more likely to indulge in that viewpoint, maybe even sarcastically.
While on small websites I think a user in that case might just be dissuaded from engaging in the community outright.
I don't believe gamergate to be a special thing, just oxygen, fuel, and a spark.
Personally I've found reddit karma to be correlated with the obnoxiousness of the user, not the inverse. There's plenty out there who feel the same. Even with that point aside, what is there to prove? I don't understand the assertion. Wrongthink will get you censored on Reddit faster than anywhere else regardless of how much karma you had.
Even good posts on Reddit get shuffled past the first pages after a few days. There's too much negative incentive to put effort into content when it can get censored, downvoted, or buried. Sure, you might potentially reach more users in the right circumstances, but many of those users are addicted to notifications and doomscrolling, not enthusiasts of the subject matter. Plenty of that here on HN as well (more on the side of trend-chasing, manufactured two-minute hate about climate or national news, while rarely challenging the status quo) but it's at least tolerable, and I know the moderation is fair here.
Forums haven't "lost out" when they can still serve to bring people together who have, or want to read unpopular opinions. I could go on about use cases. If there were none, there would be no forums.
The takeaway here is that Reddit can be and is considered by a decent proportion to be a dystopian failure, not a "convenience". A black hole of siphoned search traffic that regurgitates a watered down and sanitized imitation of something resembling a community.
Ignoring all the publication of content bits, I always felt like what the internet really needed to improve portability of accounts and data was just a federated user authentication standard.
https://engineering.fb.com/2024/03/21/networking-traffic/thr...
Probably all of it?
So today you can spin up your own Mastodon server and have anyone follow you from Threads and vice versa.
https://en.m.wikipedia.org/wiki/Embrace,_extend,_and_extingu...
Either way their intentions are pretty easy to understand.
ActivityPub integration keeps regulators off their back and allows external content to appear on Threads which they can then monetise alongside ads.
No doubt that the existing major networks would prefer to be able to enclose the budding commons that the fediverse is starting to create, but I'm hopeful that people will choose community and/or locally operated alternatives rather than continue buying into the mega corporate offerings.
I think social media as we know it today is already at the end of the road. Our data has been obtained to make generative AI products and now we are entering a new phase of the internet. The social media we have has already been poisoned by assholes trying to make everyone addicted for money, or victims of non-kinetic warfare.
There is going to be way to many self-managing sock puppets all with an agenda. Automated psyops agents left to their devices, all talking to each other and you.
That’s not to say there won’t be sociable media, it’s just that regardless of how federated anything is we are all still going to constantly be in contact with AI.
the next big social media platform will be the one that can guarantee the most authentic human connections.
Until we get to that point, we will have what we have not and watch it get worse as smaller federated instances pop up that become selective or invite only.
Not much different then private hidden services
i am on whatsapp, wechat, telegram, matrix, signal, sms, irc, email, even discord...
on each of these services i am mostly in contact with real humans that i met in person.
i am in a few groups mostly made out of these same people, but also some groups where i never met anyone. i am not on facebook, mastodon or ex-twitter or any service that tries to feed me with stuff instead of just giving me a timeline of all messages shared. even on youtube i follow the channels of my choice and mostly ignore recommendations except to discover other interesting channels.
i don't see any of that going away. therefore i don't see anything changing for me
Most of the apps you mentioned are focussed on that second category, whereas (I believe) the GP was talking about the first category.
i agree that this distinction is important, but i am afraid it is a lost battle, and as we criticize the negative effects of the first category of social media we need to also point out how social media can be useful when it is in the second category.
Which is a thing that I don't think is feasible. I could be wrong, of course. But I do agree that authenticity is going to be the issue that social platforms will live or die on.
I think the next evolution of "social media" is more likely to be a return to the old-fashioned notion of interacting with people in person.
But none of our personal experiences is a great indicator of what "most people" do.
84% of people aged 18 to 29 use at least one social media site.
That's what I mean by it not being feasible. Technically, there are many, many ways to make this happen, but they all come with very serious problems that I don't think most people would accept.
On the other hand, people accept things now that just a short time ago were considered outrageous and unacceptable by the majority of people (in the US, anyway). I maintain hope that this descent will stop at some point, and that hope is likely coloring my expectations.
the problem is a lack of ability to recognize worthwhile content, a lack of critical consumption. there is also be a problem to find good content among the mass of junk. but that's already a problem now. i have mostly given up. i find new content mostly through HN.
the problem will be that a lot of good content will remain undiscovered.
so let me make this claim in response: the next big social media platform will be the one that will make it easy to discover actually good content as opposed to popular content.
But is possible that social media as source is already poisoned by Gen AI postings.
There's no way government are going to give up the control they have with big tech and companies that can be regulated and served warrants and stuff. The last thing they want is for individual decentralized citizens to be able to stand up their own systems and discuss dangerous ideas[1] without oversight.
On top of that, big tech will be eager to push things like Web Environment Integrity[2] and Private Access Tokens[3] that can ensure that unapproved (or semi or fully anonymous) actors with dangerous operating systems like linux aren't allowed to participate, at least on a first-class level. It will be packaged as "anti-spam" or "a better user experience for our users" but the fact that it massively empowers the tech companies and enables them to further entrench you into their walled gardens will absolutely not be an opportunity they want to miss.
[1]: The definition of dangerous ideas will depend on who and which party is in power. Unless the corporate overlords end up in control, in which case it will be what their managers consider dangerous.
[2]: https://en.wikipedia.org/wiki/Web_Environment_Integrity
If you want to a shining example of effective content moderation, public utility and centralization, then just look at Wikipedia. What's different about them? Non-profit.