Alcohol, tobacco, drugs, sex, driving and many other things are heavily regulated for children. Should a parent be able to allow his 12yo daughter to drink, smoke or marry 40yo guy? We're living in a society. Social media is just another item in the list.
I’ve seen many people make comparisons between these things and the social media ban, but I actually think they’re quite poor comparisons.
The main concern is around enforcement, which is still TBD for this ban. I assume you mean illicit drugs which are banned for everyone so I’ll skip that as it’s not relevant. For sex/marriage, these are basically not enforced until people notice or they cause a problem so again are not relevant. Driving is an interesting one but I’ll come back to it. The closest comparison here is alcohol and tobacco.
For alcohol and tobacco sales the rule of thumb staff are trained on in Australia is if the customer looks younger than 25 they ask for ID. The customer present their ID, the cashier visually checks the DoB, and the sale goes through. This does not affect the majority of Australians who either do not smoke or drink, or look old enough that no ID check is done. Enforcement for a social media ban would be onerous on all Australians who use social media (I don’t have numbers but I’m sure it’s more more than (smokers ∪ drinkers)) plus the scope for potential abuse or infringement of rights is far greater. Compare this to social media. How would such a ban be enforced? The kids are not stupid, they will find a way around whatever the enforcement mechanism is. So either the enforcement will be a) trivially circumventable to the point where the legislation is completely useless for its ostensible purpose, or b) devolve into an endless cat-and-mouse game trampling Australians’ rights every step of the way. Depending on how eager the Aus gov is to enforce this it could easily extend to VPN bans, destruction of anonymity online, and yet more means to eliminate free speech. These things are all extremely important for a functional society where people, especially vulnerable and marginalised people, can speak up without fear of retaliation. A cashier checking your ID at the shop is nowhere near the top of a slope as steep, nor slippery as this.
Driving is a really interesting comparison here actually. I’m not sure I would be opposed to a social media license. In the same way the purpose of a license is to ensure that road users can do so in a safe manner, maybe something similarly focused on education would be more helpful here than a ban. I’ve actually long blamed a kind of tragedy of the commons for the sorry state of the modern internet. Most users are simply not savvy enough to know better than to use it in all but the dumbest ways, fall for the dumbest scams, and basically allow themselves to be corralled like cattle into the sterile advertiser-friendly pens big tech companies have constructed for them. So in anger I’ve sometime said we should only allow licensed users online. Anyway that’s a bit off topic but a social media license is an interesting concept.
A lot of commenters here are having their minds blown by this. And while I also love this I get the sense that many others here are maybe too young to remember that this kind of open access to data used to exist for lots websites. It inspired companion sites and loads of creativity. I find it tragic really, what the internet has become. I hope federated, and even more-so p2p, protocols take significant foothold on the internet and help revive this spirit of the web. The corpo-web is so fucking boring.
It’s worth noting that twitter itself owes a lot of its popularity to its openness in the early days. In the early days there were third-party clients, RSS feeds, XMPP support, etc. You could post from a curl command in a cron job, leading to all kinds of interesting automated feeds. Then they walked it all back in the early 2010s.
I like that Bluesky’s federation model makes it harder for them to do an “open platform” bait-and-switch like Twitter did.
> I like that Bluesky’s federation model makes it harder for them to do an “open platform” bait-and-switch like Twitter did.
Why would it? They can still lock everything down and few Bluesky users will even notice. This is similar to what Twitter did, or what Google Chat did, etc. Compare this to other federation platforms where a server that locks itself down loses access to a huge chunk of the network, once the other servers reciprocate.
Since migrating your personal data was a thing they thought about since day one, migrating to another network than the current one would be way easier than any centralized service and also easier than ActivityPub.
Seems there is one piece of the puzzle missing yet ("AppViews") in ATProto to be able to run completely independent, but seems they're currently working on getting that in place now.
You can host your personal data, but as long as Bluesky Social, PBC is running the main Relay and AppView, and there's no easy way for Relays to talk to each other (like e.g. it works in Mastodon), they are in reality a centralized service.
Oh yes, and they are the only ones owning and developing AT Protocol, which makes it much more a currently-open-sourced protocol rather than a standard that is jointly developed by the industry.
The big central place is still the PLC directory that effectively means all accounts are centralized at BlueSky, even if your posts are not. They haven't planned to make it any decentralized in the future.
Nothing technical is preventing any federated platform to stop sharing the content with its peer. Only thing that prevents it is if there are multiple big peer and they can't afford a network partition, which is not the case with bluesky. Eventually VC's money will run dry and they don't have any solution for this.
You could still migrate all _your_ data to another service in Twitter quite easily, and most definitely you could in Google Chat. This did not change things.
> You could still migrate all _your_ data to another service in Twitter quite easily
Yeah? I don't remember being able to migrate from/to Twitter and taking followers/following etc with you without having to ask/request others to do something too.
>In the early days there were third-party clients, RSS feeds, XMPP support, etc.
Right. This is something I keep pointing out in threads about RSS. Some people will say RSS never left. Well, it left Twitter for one. Google News and Craigslist for others.
I almost wonder, to GP's point, if people have just completely forgotten all of this, which is why they think nothing was lost.
The 2000s Internet felt way more innovative than the one we have today, despite all of the WASM, WebGPU, JIT optimizations, and other technologies that have been developed in more recent years.
We had torrents, open data, open protocols, and people were sharing data and remixing it freely. Mountains of stuff like this Bluesky demo was released every single day. We had link aggregators to point to the cool things that were happening, and we even had tools that let you pipe data sources between various APIs to enrich and recombine things easily.
Platforms stopped this. Facebook, Google, and even Apple put an end to the wildly evolutionary behavior by delivering a canned experience to the masses.
We need a return to P2P where single platform silos and their army of product managers don't shape how we interact with technology and the bulk flow of information.
> The 2000s Internet felt way more innovative than the one we have today
Because it seems like this stuff is taught in Management 101 in all of the business schools: once you establish yourself with all this talk about "openness", etc. then the only way to succeed is by creating a walled garden, either through abuse of your monopoly position or by regulatory capture.
Cases in point: OpenAI _and_ Anthropic both pushing for regulation of AI, now that they have a dominant position.
I swear, the moment MBAs get involved, they try the same crap everywhere.
It's a common trope to blame MBAs for all the ills in the world.
But the reality is that having a moat and how to defend it is a fundamental strategy that every CEO is expected to know. Because it will be one of the first things you get asked from YC, investors etc.
And using regulation to lock out competitors definitely did not start with OpenAI and Anthropic.
I see 2 way to do this. A company (and PM) sees demand for the feature and they include it, ot it is forced by regulation.
A lot of these companies that originally had open standards formed with huge amounts of VC money and they prioritized growth over everything else. Then when they reached a certain scale, investors valued profitability and they slowly squeezed and monetized users until all of those open standards features were gone.
We still have all the tools you talk about today. But with the benefit of much simpler languages and SDKs and tools like LLMs to help generate code. I've seen children learn programming far faster with Swift Playground on their iPad than I had to with C++ books.
And these sort of canned experience are what helped bring technology to everyone. Which was always supposed to be the main goal.
Kids are all over Discord, Roblox, Minecraft, and VRchat. They're writing scripts and mods, and that's great. They're probably having a blast and a good number of them are learning a lot.
But they're doing all of these things in someone else's walled garden, on buttoned up platforms that typically constrain what they can accomplish. There are fewer degrees of freedom and a lot less ownership in the work they're doing.
These platforms also cost kids money. They use toxic gotcha mechanics and peer pressure to monetize. This part is strictly worse.
It's a lot different carving out your own clubhouse and culture when you're renting. Especially when you're made to speak a certain language and abide by a rigid set of rules.
They are a public benefit corporation that use that as a selling point but then don't disclose their charter. That seems really shady to me, but less than what twitter has become.
I used to run a cron job that would scrape my university's daily bulletin and post every day right after it was updated. It may have been the first “presence” the university had on twitter. I remember following things like weather stations that had automated accounts, as well.
A long long time ago, LiveJournal used to display all new user-uploaded images in a firehose type page.
It was a fascinating glimpse into the shared lives of people all over the world. It was definitely from a simpler time; there's no way something like that would be available now due to the violence/abuse material that gets uploaded.
Mastodon has about 1 million active accounts, bluesky has 12 and you can also host your own data. Threads (by meta) has 100 million but you can't federate it yet so bluesky is the way to go currently for selfhosters and privacy advocates.
Mastodon is also confusing for a lot of people and when you point it out techies will appear to yell at you and tell you that "no in fact it isn't."
Whereas Bluesky has a familiar consumer app feeling to it.
Each of these sites also has a distinct vibe. Threads, for example, feels a bit like when Time Square turned into a kind of Disney property. It's clean and safe! But it also lacks a bit of soul.
not really. its stalking. it poses a direct threat to someones safety. and people arent even supposed to have that information under the current rules... thats why its only possible to get his jets location by crowd-sourcing it. anonymity for high profile passengers is a legitimate concern, enough for the FAA to bake it into how these flights are tracked. soon this loophole will be closed and then what will you point to? how could you seriously compare this to being banned for saying something that is politically incorrect?
Even if true, which I'm skeptical of, social media sites don't have to be permanent. It's no different from any other service being enshittified, just use it until it sucks and then stop.
I'm thinking that it's a cynical way to put all the data on an open network rather than a gated and paywalled one. As far as I can tell, Bluesky is still entirely funded by Twitter and that will eventually be cut off. They're betting on the Google/Mozilla type relationship to be maintained for now. They need to make the indexing service easier to replicate before that happens or the network will collapse when funds dry up.
I don’t want to frame it positively. In my experience, people who hoard knowledge and/or relationships (as with other teams or customers) are toxic.
I hate all sorts of gatekeeping behavior but I especially despise having to wheedle information or effort out of someone like that. They almost invariably have an ego that is out of proportion to their contributions.
Side note: Occasionally I have seen people who take on the thankless task of being a maintainer of something no one else wants- those people tend to welcome others and share as much as others are willing to listen to, so being the single person who knows something doesn’t make you a gatekeeper.
Notes is a very different app to Notepad though. Nobody uses the Notes app to edit plain text files, which is the sole use case of Notepad. Notes is for writing, well, notes. Notes also has a lot of other editing modes and features like drawing with the apple pencil, scanning documents, cross device syncing, etc. As far as I’m aware Notepad can only edit plain text files.
It's unfortunate that Apple Notes doesn't handle plain text. I'd use it like this if possible.
Anyway, I use TextEdit in plain text and autocomplete, autocorrect and spellcheck all work just fine, as they work in every text box in macOS. That Windows' Notepad got some of that just in 2024[1] is bonkers…
I used Notes to edit a text file once and learned a very important lesson; It changes all your double-quotes to fancy Unicode double-quotes. They may look prettier, but they completely break the code you are working on.
I can write a computer program the spews all manner of profane things. If I were to release a product that does that I’m sure it would be much criticized and ultimately unsuccessful. Yet this doesn’t mean we should cripple the programming language to prevent this. Models are much more akin to programming languages than they are to products. If they are used to make products that do things people don’t like then people will not use those products.
you are comparing ai to programming language but programming language if uncensored doesn't have the ability to wreck humanity but uncensored ai sure does.
I would actually be curious if someone uses it for uncensoring because I am gonna be curious about how different would it be to the original model
But aside from that curiosity , this idea can increase no of cyber criminals , drug suppliers and a hell lot more
IMO the Youtube title isn’t even accurate, it’s just clickbait. The HN title is a more accurate representation of the content. Also to back up my sibling commenter, Derek is known to trial several title and thumbnails within the first few hours/days after a video release.
When teachers or doctors or nurses strike regular people, the general public, suffer. In the case of the NYT tech staff nobody suffers except the NYT leadership. Oh no you can't read NYT during election week. Whatever, read something else.
The megacorps' refusal to negotiate or compromise is what caused the problem. They are the villains, not the working people being understaffed, underpaid, and hated on by selfish third parties who don't understand solidarity.
I agree. I’m just pointing out the reason why those other professions need to carefully time their strikes to avoid adversely affecting innocent third parties. In the case of the NYT the only people affected are the leadership so they can time their strike for maximum impact without regard for impact on third parties.
We use pkarr at iroh.computer for node discovery. It is anything but slow. It is very rare for a lookup to take more than a few milliseconds. It is sometimes faster than our non-p2p node discovery option which is using DNS.
DHTs get a bad rap because of many recent DHTs that were horribly inefficient. But mainline is different. Many of the design decisions of mainline seem very limiting at first, but make a lot of sense for perf.
E.g. a pkarr record can only be 1000 bytes, so the entire message fits into a single non-fragmented UDP packet.
Creator here. It takes around 30ms to resolve a public key DNS entry on the DHT. After the first resolution, the entry is obviously cached according to the configured time-to-live (ttl).
It's not as fast as regular DNS, 30ms is still very quick.
