While I use it in LaTeX I tend to avoid it in non-LaTeX contexts BUT even though it's much used by LLMs having switched to EurKey layout years ago I can type it on my keyboard as well as × and many others, so it's not such a perfect AI indicator.
We simply need open, decentralized or distributed platforms. Nostr is not that excellent, but it works, and have a potential to became "unified personal web platform", take a look at it.
- Nostr is generally much lighter; it can even be served behind Tor without having a public IP, there's no need to maintain a web server etc. There are compact, self-contained relays like Haven, for example, which are a single go install-able app that includes everything needed on the server side, with practically zero setup.
- There are various clients, including mobile and web as well as desktop, which is enough to satisfy pretty much everyone's tastes.
- There's also an economic model that could be the future of journalism: everyone publishes what they want, and those who enjoy it can make micro-payments, if they wish, to support the publisher.
For now, it's a toy with many abandoned experiments, while Mastodon is a walking dead, having never really taken off. In other words, as they stand, neither of them is working. But Nostr has the potential to become the communication hub for many; for instance, there's already a Matrix-like service (0xchat and potentially whitenoise) that supports chat, audio, and live video, requiring only Coturn and a Nostr relay. There's also "long form" support, meaning personal blogs all on the same technology.
In other words, in a short space of time, on Nostr you can have:
- A personal blog-style site
- A personal Twitter/X
- Personal chat with audio and video
- Private notes if you need to jot something down on the go
- A search engine and address book that could allows with different access levels, a real address book usage for personal contacts.
Potentially all in a single, complete and lightweight deployment. There isn't the burden of federation, which makes many hesitate to activate it because, depending on who they federate with, they find a massive amount of resources consumed. It's essentially text, binary blobs, and near-real-time communications all in one. Haven is the first piece of the puzzle, MOAR is the successor in the making, but eventually, there will be one that integrates 0xchat and a web client, all-in-one.
The Fediverse hasn't achieved this and doesn't have the characteristics to do so.
Then again, if we're honest, the old Usenet did it better, but it's dead to most people, whereas Nostr is alive. People only dislike it because it comes from a crypto community, and many are biased against anyone from that world regardless.
And what exactly would be the purpose of age verification? Because defining someone "mature" based on their age is pretty hit-and-miss: we have plenty of adults, even of a certain age, who it's hard to imagine have ever finished adolescence, for instance. On paper, they are absolutely of age. We also had a certain Alexander the Great, emperor of a large part of the planet at 20. We had 13-year-old Pharaohs active in government.
We also have gazillions of examples of apparently innocent rules being used to boil Chomsky's frog, one small temperature rise at a time. For the first time in a long while, I'm starting to sense a certain fanaticism on this topic here on HN, which sounds very much like the molecular agitation when water starts to boil.
> And what exactly would be the purpose of age verification? Because defining someone "mature" based on their age is pretty hit-and-miss: we have plenty of adults, even of a certain age, who it's hard to imagine have ever finished adolescence, for instance. On paper, they are absolutely of age. We also had a certain Alexander the Great, emperor of a large part of the planet at 20. We had 13-year-old Pharaohs active in government.
That's really no different than age of consent laws. In the majority of US states (33+DC) that age of consent for sex is 16, 17 in 6 states, and 18 in 11 states.
In Europe it is 14 in 14 countries, 15 in 12 countries, 16 in 20 countries, 17 in 2 countries, and 18 in 3 countries.
All of those are somewhat arbitrary. There are many people over 18 who lack whatever maturity age of consent laws are trying to ensure people have before they can consent.
Going the other way there are people who are under the age of consent in most of those countries or states who are mature enough that there would be no harm in letting them consent.
Any particular population wide age of consent in a state or country then cannot simultaneously protect everyone who needs protection and avoid forcing protection on people who do not need it.
It would in theory be possible to make the age of consent an individual thing where you have to be psychologically evaluated and if you pass you get your consent license. (A hybrid approach might also be possible--a high automatic age of consent like 21, with people under that able to apply for a lower age. Probably also combined with "Romeo and Juliet" laws so people under 21 who just want to fool around with people close to their own age can do so without having to be psychologically evaluated first).
I expect that very very few people would be in favor of replacing the one size fits all approach to age of consent with such an individualized system.
I prefer no filters instead, for one simple reason: who watches the watchmen? If we had a digital identity on a national blockchain run by open-hardware home servers and FLOSS software, where every node exists by virtue of digital identity, meaning there's no risk of a 51% attack and everyone is forced to play with their cards on the table, I might accept a ZK proof. But that's not the case, and the privacy guarantees of private entities and the very subjects pushing for this verification make me say, quite simply, NEVER.
Because we know perfectly well that it's the precursor to mandatory SSO for everything, South Korea style, which is unacceptable and incompatible with Democracy.
I've tested WriteFreely, Mastodon, Nostr, ... but all lack the basic to succeed IMVHO:
- being a single, simple application, without much deps, maybe go-get-able, pip-able, cargo build-able etc, WF actually is one of them
- offer a platform, meaning a blog, comments per posts, distributed identity, optional chat
We have many different projects who do so, but not a single integrated one.
Nostr is good for the infra, have a sufficiently complete relay (Haven) and a future one a bit more complete (MOAR), but lack a built-in client and a decent chat support (0xchat is nice, and very hard to deploy in a sovereign manner).
WF is nice but limited as a blog and have no comments
Matrix is nice for chatting, with a very complex audio/video support, with very little documentation, I manage to get it running, with LiveKit as well, but it's a pain. XMPP is even worse because it lack a complete client for all platforms and it's very touchy on DNS setup.
The defunct ZeroNet was very nice to host personal websites without a domain name and also behind NAT, but offer nothing ready made to use with it.
...
Long story short we have the wrong tech stack underneath. We need to rediscover the old Xerox model of the OS as single integrated app, where anything can be combined at the user will, with ease. Emacs/LispM do better pushing anything in the config instead of relaying on a live image. But that's what we need. We have one mind, we need to combine out digital companion.
It's not impossible; it's their centralised model that is. It's unthinkable to have private platforms on modern mainframes (data centers) instead of distributed, decentralised services where everyone holds a piece (DHT) or whatever they want (e.g. Nostr/Blossom), and is responsible for what they do.
It's impossible to imagine having democratic societies where four fat cats know everything about everyone and most people know almost nothing about them, where information, instead of being scattered everywhere for resilience, is concentrated in just a few hands.
Coming from a company that profiles children with Classroom, sure. Coming from those pushing for age verification, just to shift the Overton window of acceptability towards mandatory logins for everything, the end of the open web and free discussion, all to better feed on slaves.
No thanks, ChatControl is a THREAT to Democracy, and the companies pushing it must be ELIMINATED from the market for reasons of national and human security, along with the politicians lobbying for them.
Considering that office suites are software from a bygone era, born from the idea of letting untrained secretarial staff use a PC as an advanced typewriter and calculator, the business and the squabbles surrounding them, which have absolutely nothing to do with FLOSS, are frankly laughable, if they weren't so pathetic.
LibreOffice (and any office suite) is a piece of software as massive as it is absurd, and those who use it don't even realise it, which is why there's so much business built around it. It's 2026; information shouldn't be managed in scattered files designed for printing and then used on screens anyway. It's high time people were taught how to actually use a computer, rather than playing around with software that hoped to make computers usable for those who don't know how to use them, and has done more harm than good in the process.
If you like drink interested PRs dreams on sale yes, but reality is quite different.
The OEMs who have lied https://unctad.org/system/files/official-document/der2024_en... through their teeth about the lifespan of a well-made desktop to compare it to a smartphone are simply on their last financial legs; their Nazi wet dream doesn't scale. A society like that doesn't work. That's why they'll collapse just like their German-speaking counterparts did a few years back.
The West is collapsing to save four kleptocrats and their model of exploiting us all; nature doesn't give a toss, it carries on her way and so does technology. Dead wood can stay standing for a long time, in some cases it even becomes a fossilised trunk, but it doesn't evolve; it's still a dead branch.
If you think the current Office model will be replaced by some LLM, you're dreaming just like those who hope for it and are working to make it happen. Sure, plenty of jobs are no longer necessary and gradually mankind, if we don't commit suicide first, will end up only creating new knowledge while machines apply it, but this model only works if it's open, distributed, and diverse enough to be resilient and evolve. The current monochrome model can only flare up and die.
reply