I used to check out the original version a few times in the past. I loved the overall premise of a uncensored network (for the context: I lived in an authoritorian country that doubled down on censoring online communications), but no system I've tried out (Freenet, ZeroNet, I2P, Tor) grew on me. Too much, uh, weird stuff (or worse), while interesting content was pretty sparse. And I don't produce anything worthy of sharing myself.
The tech could be there, but the society is not. [Not] surprisingly, the social demand seems to be all focused on to work around the Internet censorship with very limited interest in bootstrapping something that would be resistant. At scale, humans always pick the cheapest/lowest-effort option, even if it's obviously sub-optimal and even if it leads nowhere we want (or so I think) to be.
The new Freenet (formerly Locutus) is more of a communication than a storage medium, which should make it less susceptible to the kind of problems you're referring to. That said, even on Hyphanet you really need to look for something bad to find it as the default indexes are curated.
We’re also developing a decentralized reputation system based on the "web of trust" concept. This system is intended to help filter out antisocial content—like spam or worse—without relying on any central authority.
This was my experience with matrix.org, mostly pedos and far right extremists or other weird anti social characters . Turned me off from these kinds of platforms big time.
if you’re talking about the matrix.org homeserver’s room directory - these rooms are strictly against our terms of use (section 6 of https://matrix.org/legal/terms-and-conditions/) and we shut them down, and these days have even frozen the roomdir to stop them appearing.
if you’re talking about the wider network - yes, there’s a subset of abusive users… just like on the web, or the internet, or email, etc. Unfortunately there’s nothing we can do about; it comes with the territory of being an open network where anyone can participate.
My experience of Matrix is more that it’s full of FOSS projects like Mozilla, Ubuntu, Debian, Fedora, GNOME, KDE etc… as well as lots of government users. But ymmv. If you are still seeing abusive rooms as a matter of course, please route details to abuse@matrix.org, where we do actually act on them (if they are on servers we control)
Wow, yeah the parent comment is shocking to me. For me matrix is OSS discussions, like a modern IRC (not that IRC is gone, a lot of matrix rooms bridge to IRC).
Though I suppose if you search for other stuff you can find it.
Though it strikes me as a little odd, I never really associated matrix with anonymity, more with federation, end to end encryption, self hosting.
Perhaps hosting matrix servers as tor hidden services could serve that purpose.
It's a bit strange to even criticize these services for the existence of abusive users. The underlying problem is that abusive people exist in real life: it goes without saying that they'll show up in any context that's accessible to the general public.
Complaining that bad people are using Matrix is about as meaningful as complaining that bad people are driving on the highways or are filling their bathtubs with from the local water supply.
I only ever use a half dozen or so public Matrix rooms (I mostly use it for private chat) but my experience in those is nothing like you describe. Unsavory characters occasionally wander in, just as they do on every other public forum I've used, but they are shown to the door and the normal conversation carries on without them.
I wonder how someone must be using the platform, or how they're looking for rooms that might interest them, to get the impression that it's "mostly pedos and far right extremists or other weird anti social characters."
To be fair, everyone with a commute who had or has to sit in endless traffic is turned off from driving on public roads and wishes there were fewer vehicles or that they were the only one on the road.
Paraphrased, "I was driving home, stuck in traffic, and this thought occurred to me, now I know this is bad, but I thought, if half of everyone in this city died, I'd be home by now." - Paul Reiser
They'd need to specifically die in their own homes (or otherwise not on the roads), or else the resulting traffic jams would make your drive home even longer.
"What about..."? Isn't a very convincing form of argument in general and in this case it's particularly specious. When using the public roads, people's unsavory/illegal behaviour with respect to images (for example) isn't something that affects other road users, but if you're using a chat application an innocent user may well see/download content that cannot be unseen and is a crime to download (even innocently).
In that context the decision to avoid a chat application if you feel it is likely to put you at unnecessary risk where there are alternatives that will not is perfectly reasonable.
I don't mind "eccentric" personalities in my online spaces. I hate people trying to restrict online freedom more. To a degree that I think the threat of far-right people is pretty miniscule in comparison, but people allegedly trying to combat them let themselves be driven by fear. And the result is usually as bad as you could expect.
There are a lot of excuses for bad policy with the deflection of "at least I am not a far-right unperson".
A chat application usually cannot put me in any risk at all. Public personas might have different problems here, but the rules of PR aren't that new and these cases build a poor general case.
Fair to have the choice of a supervised chat app as long as people don't try to enforce all chat apps to adhere to arbitrary rules due to the latest outrage or panic over certain people, opinions, politics, bubble tea or whatever can be the source of ire.
The solution to this, at least in the case of Tor, is for existing sites (e.g., Wordpress, medium, etc…) to provide one-click onion-site publishing support.
I don’t think it’s a technical problem. Hosting or accessing a site on distributed networks is no more complicated than running or hosting a VPN (e.g. https://novayagazeta.eu/vpnovaya), and organizations host them just fine, and layman people were proven to be capable of setting up client software when they were forced to (e.g. Instagram or YouTube bans in Russia).
I see it as a more of a chicken-and-egg issue. Publishers don’t come because there’s no audience, audience doesn’t come because there are no publishers. Plus, there is no recognition of distributed networks as a solution to censorship - the current non-enthusiast view of them ranges from “haven’t heard about it” through “tried it, found it useless” all the way to “it’s only for pedos and nazis”, which is extremely harmful for any meaningful and socially beneficial adoption, of course.
I’m not sure if those are the actual reasons. Certainly not a technical issue, though.
It’s no longer funny to see how developers again and again are trying to fix problems of real world by creating digital hideouts because this is the only thing developers do the best - programming. Whenever we see a problem with banks, governments, corporations, we immediately try to fix the problem with “own digital alternative”. But hiding from the real problem in a new network doesn’t fix remaining initial problem for the rest of the world. Complex socio-economic problems shall be fixed with socio-economical instruments. No new network could solve problems of monopolies and profit-greedy corporations.
It's a technical solution to a technical problem, that problem being the client-server architecture which inherently centralizes control and disempowers users.
The Internet started out peer-to-peer, but client-server became dominant during the 90s when the Internet became mainstream - because at the time it was the only option for building services that scaled.
Freenet provides an alternative to the client-server architecture that doesn't require compromising on user experience or scalability.
> Complex socio-economic problems shall be fixed with socio-economical instruments.
Missing the forest for the trees.
The socio-economic problems are caused by the limitations & drawbacks of modern tech: It's too easy to centralize control over a platform, whilst keeping things decentralized requires an upfront desire to do so, along with the nightmare that is backwards compatibility.
> No new network could solve problems of monopolies and profit-greedy corporations.
It is a solvable problem. Monopolies & rent-seeking behaviors happen because the tech involved has chokepoints that favor centralized entities.
We know the business processes that should happen for running service X, based on our daily interactions with them and the idealized form of how they should work. Breaking the processes down into components & putting them onto a decentralized compute platform would significantly relax their control over common chokepoints. Dispute processes would be made as public as current government processes, ideally Estonian-like.
Society and governance are gradually getting transformed and integrated into a global communication network run by software and hardware. The Earth is becoming a giant computer within which we live.
It's too early to declare that complex socio-economic problems cannot be solved by a new network/infrastructure/etc. Software is one of the "socio-economical instruments" and we shouldn't dismiss its potential for social transformation beyond the control of banks, governments, corporations.
While Locutus is an improvement on the previous Hyphanet model, I'm still concerned about the caching model. In Hyphanet, data would be cached the more it was used, and peers connected over multiple hops. This had the potential to cause huge headaches as authorities began investigating issues such as CSAM, because their understanding of how caching and requests work did not match up with how it actually worked. While, certainly, one could build a subset darknet, all it takes is one bad peer to expose the whole darknet to prosecution.
I played with Freenet when it first came out, but when I realized the implications of this -- which later turned out to be true -- I destroyed the hard drive I had been using to run Freenet, just in case.
Locutus seems less aggressive about caching data, but it still does some caching. Without really digging into the documentation or source code, I'd still be nervous about running it.
Locutus (now Freenet) is more of a communication than a storage medium. It's focused more on allowing people to build decentralized tools like group chat[1] for realtime communication than be a content distribution network like bittorrent.
We're also planning a decentralized reputation system based on the concept of "web of trust" that should do at least as good a job of ensuring users aren't exposed to unwanted content like spam or worse.
That is reassuring. I hope that your team can build this web of trust, because it really is crucial.
It is great to have a tool that allows ideas to be shared, but I think that consent regarding which ideas are shared or promoted is important, not only from a legal perspective, but also from an ethical one.
Agreed, any system that allows people to discover content needs to protect them from stuff they don't want to see, whether it's annoyances like spam, malicious attacks like DDoS, or extremely harmful content like CSAM.
Centralized services don't do a great job of this but I think we can do better with a decentralized approach.
We won't have a perfect solution overnight but we've already built important components of such a system (eg. ghost keys[1]).
Under EU rules (Digital Services Act) I believe an automated cache is treated similarly to a pipe - you're no more responsible for holding content in your cache (especially if encrypted!) than a Tor node operator is for someone accessing it through Tor.
German police did raid the home of a guy who ran a lot of Tor nodes. Authoritarians don't need any technical excuse to raid your home - they can just do it if they want to scare you away from these platforms, regardless of how they work.
Unfortunately, US case law is murkier. There have been some notable convictions of people using the old Freenet to search or host CSAM. However, the digital forensics used in these cases was pretty sketchy. If someone simply caching data from downstream queries were held to the same standard, they'd also be considered guilty.
I have no doubt that there was other evidence that led to the conviction of these individuals. But, I can only go on what I know, and what I know is that the standard of digital forensics evidence in those cases was subpar.
Fascinating! I thought this was cool as I clicked through, I'm glad to see that it's been a dramatic topic for 25 years, not just some new idea. Hyphanet seems more openly political, which makes sense but is kinda a hard sell for me after seeing the impact Bitcoin had on US citizens vis-a-vis scams, hacks, and misc. bad actors. But maybe that's a sacrifice worth making in the name of residents of less free nation states?
Either way, pretty funny when a software project has to tackle questions like "what is property in the modern era?" and "what is free speech?" in their FAQ. That's how you know you're really pushing against the status quo, I guess...
Those scams are not isolated to Bitcoin. You see them in many content creator industries. They are giveaway scams and templated to fit a niche. You can do them for gambling, gaming like cs-go skin giveaways, makeup giveaways and so on. They are really easy scams to do because you can just wait for the next trendy thing to happen and then target that trend with bot accounts. On youtube they will often just recycle what accounts don't get reported.
Platforms do their best to take care of this, but these scams are so easy to run that you can automate them and spin up new accounts very quickly. Even if there are financial barriers it can still be madly profitable to accept the costs.
There is no clear way to deal with the concerns you mentioned, but I think playing too safe and not exploring technology will quickly put you behind the curb. Most of it can be resolved with good financial regulation.
I mean, I was more referencing the big epidemic in the US recently, “pig butchering” — basically catfishing with more money involved. Certainly no system alone can make it impossible, but I think even the biggest crypto fans in the world should acknowledge that it makes it much easier, and thus much more common. “Go to this website” is a lot easier/more reasonable ask than “scratch off 20 gift cards and send me the codes”, or whatever western mutual is
Sure, crypto exposes you to potential bad actors in the form of scammers, criminals, etc. Fiat exposes you to potential bad actors in the form of banks, governments, police, etc.
So, whichever represents a bigger threat to you is largely a question of circumstance.
It’s about rates, accessibility, incentives, and other society-level statistics. There are scams that have become much more common thanks to the affordances of untraceable, unmanaged online currencies, which shouldn’t really surprise anyone.
Would love to see some network like Freenet start to see more adoption - it seems obvious that we need something like this. GNUnet seems to have some interesting ideas and good motivations too.
I will admit that I didn't follow the renaming or possibly forking or whatever happened to freenet / hyphanet / etc back last when I was reading about this. If someone could explain it clearly that would be stellar.
Hyphanet is great but requires some basic skills and hardware for running. Using Nostr with Primal as a client works exceptionally well as a Twitter alternative. It's privacy-respecting, censorship-resistant, and truly an outstanding tool. I'm grateful it exists to counter the current wave of regulatory overreach threatening free speech, even in Western countries.
> I will admit that I didn't follow the renaming or possibly forking or whatever happened to freenet / hyphanet / etc back last when I was reading about this. If someone could explain it clearly that would be stellar.
The reasons for the renaming are addressed directly in Freenet's FAQ[1]:
# Why was Freenet rearchitected and rebranded?
In 2019, Ian began developing a successor to the original Freenet, internally named “Locutus.” This redesign was a ground-up reimagining, incorporating lessons learned from the original Freenet and addressing modern challenges. The original Freenet, although groundbreaking, was built for an earlier era.
This isn’t the first time Freenet has undergone significant changes. Around 2005, we transitioned from version 0.5 to 0.7, which was a complete rewrite introducing “friend-to-friend” networking.
In March 2023, the original Freenet (developed from 2005 onwards) was spun off into an independent project called “Hyphanet” under its existing maintainers. Concurrently, “Locutus” was rebranded as “Freenet,” also known as “Freenet 2023,” to signal this new direction and focus. The rearchitected Freenet is faster, more flexible, and better equipped to offer a robust, decentralized alternative to the increasingly centralized web.
To ease the transition the old freenetproject.org domain was redirected to hyphanet’s website, while the recently acquired freenet.org domain was used for the new architecture.
It is important to note that the maintainers of the original Freenet did not agree with the decision to rearchitect and rebrand. However, as the architect of the Freenet Project, and after over a year of debate, Ian felt this was the necessary path forward to ensure the project’s continued relevance and success in a world far different than when he designed the previous architecture.
Thank you very much, I've been browsing a bit since your thorough reply, looking forward to having more of an investigation one of these days when I get a few hours free in a row to have a good go.
I wonder how much Freenet would be considered “web3” and “blockchain” by the HN crowd, considering it explicitly uses smart contracts and transactions signed by self-custodied private keys
Guess it’s all in how you present things and what terms you avoid :)
For what it’s worth, I am interested very much in decentralized systems and smart contracts, having built them and also running a YouTube channel where I interviewed people behind the projects… including Ian Clarke and freenet: https://www.youtube.com/watch?v=JWrRqUkJpMQ
Personally, I don’t consider Freenet and Hyphanet to be "blockchain" in the modern sense, and given how much the meaning of "web3" has changed in recent years, I think it might evolve further.
Freenet stands apart with goals and ideals that are quite different from today’s distributed applications, with a stronger focus on privacy and access to information.
Neither Freenet nor Hyphanet are linked to cryptocurrencies or financial speculation. I see them as decentralized networks created to ensure freedom of expression, privacy, and access to information in an anonymous and censorship-resistant way, without any intrinsic connection to cryptocurrencies or financial systems. And that's great!
I also believe the project has gained a certain credibility over time, thanks to the consistent work and vision of its developers.
Freenet and Hyphanet aren't blockchain, although the original Freenet did pioneer the cryptographic contract back in 2000 (we called them "signed subspace keys"), which a few years later would form the basis for Bitcoin.
Freenet is designed to be a general-purpose platform for building and distributing decentralized systems like group chat[1], social networks, search, really anything that people use the Internet for today. Some people think of blockchain this way but blockchain is much more specialized (eg. group chat on blockchain wouldn't scale).
Noted. It's true that there are a few projects using the name River - which is unsurprising for such a common word, but nothing with significant traction so far as I'm aware.
Interesting that the new freenet is written in rust.
I always thought the worst part of freenet was that it required java.
I haven't used freenet in a long time. Last time I did I'm pretty sure satoshi was posting the og bitcoin releases on the message board (freechat?). Crazy times.
I interviewed the founder of the original freenet and the current freenet, Ian Clarke (among others) on our channel about decentralized systems, might be interesting for the people here:
It is on our channel https://youtube.com/Intercoin where I tried for a few years to have really deep dives with some of the too people on technology, sociopolitics and regulations around decentralized systems.
In the late 1980s early 1990s, FreeNet was not (just) a web domain name.
Freenet was decentralized idea of how the public (me) could get on the internet without already being a member of a privileged institution (or rich and sophisticated).
I believe the idea of what the freenet was prior to becoming a brandname
is worthy of being remembered.
I'm not much of a tube clicker here,
so apologies if this is redundant with the video.
The new Freenet is more of a communication than a storage medium which makes content moderation less of an issue, but we're planning to build a decentralized reputation system based on the idea of "web of trust" that should do at least as good a job of protecting people from stuff they don't want to see as today's centralized services.
Good question - computationally expensive contracts will be removed automatically as part of a mechanism that keeps track of cost/benefit to the peer for each contract.
Leave the field to the censors, create an alternative, and hope that they will not pursue you? Or stand your ground and fight them wherever they are? That is the question.
Theoretically you can run a web server at home but you'll have a problem if you start to get a lot of traffic or you have trouble with your internet connection. Your website will be trivially easy to DDoS.
Services you create on Freenet will scale automatically and are immune to DDoS.
it doesn't really scale in the traditional sense, it just pushes the 'damage' out to the entire net.
freenet overall is one of the most bandwidth and storage intensive platforms out there. I understand why -- but I say this as a means to say that it doesn't really deserve direct comparison to the open web. it's not the same thing, even if the work overlaps -- it's a lot more work.
"Damage" is misleading. A DDoS only harms when concentrated on a few peers. Freenet distributes load across the network, autoscaling in response to demand and preventing overload.
As for bandwidth and storage, I think you’re referring to the old Freenet (now Hyphanet). The new Freenet is optimized for lighter services like group chat and isn’t designed for heavy data sharing like BitTorrent. It should be much less of a bandwidth hog.
However to host something yourself you need a lot of things, for example FTTH to host it at your home, or a hosting provider; then a domain name and other things. These can be taken away from you.
Pretty sure that freenet also requires some sort of ISP to provide a ethernet connection to your machines, which can still be taken away from you. This doesn't change the fact that the www is decentralized already
Eventually Freenet will work over mesh networks like Lora and 802.11ah - meaning it will be entirely decentralized including the communication infrastructure.
On Freenet you own things cryptographically, typically by possessing a private key.
This is similar to a Bitcoin wallet although Freenet isn't a cryptocurrency, it's a general-purpose platform for building and distributing scalable decentralized services.
When many people have the ability to publish independently without relying on a central service, then it’s decentralized. The World Wide Web was initially designed to be decentralized, with the idea that anyone connecting to the internet could host a web server.
In practice, however, this didn’t quite work out. Most people publish through centralized services like Instagram, to name just one.
There are two main obstacles to achieving decentralization. The first is technical difficulty: not everyone wants to learn how to run a web server. The second is reliance on foundational services like domain names and hosting, which can be revoked. For example, if the authorities think you did something illegal, boom, your domain name got confiscated.
So, no, in practice, the World Wide Web isn’t truly decentralized. But at least there remains some possibility for it.
Thanks. I think we've always had the ability to publish content on the internet without relying on external services, and we still have. It's just not something that most people are willing to do, because it takes expertise and resources, and corporations can do it better by specializing in providing these kind of services. This is why people use them. For the same reason people don't usually make their own clothes, furniture, etc., either. Specialization and division of labour allows us to be much more productive and as a result enjoy a higher standard of living than we would if every individual had to produce everything they consume themself.
Eh, besides Cloudflare, you're pouring money down the drain if you're using any of those. Remember that recent thread where some open source project was spending $100/month on a Vercel database handling 3 requests per minute that could have been a $4 VPS?
Non-hyperscaler server hosting is a pretty competitive business that doesn't need further decentralization at this time, though it's not a bad thing either.
> The World Wide Web was initially designed to be decentralized, with the idea that anyone connecting to the internet could host a web server.
Yeah not really. If the Web was designed to be decentralized it would have used URNs as content identifiers instead of URLs. A URL is specific to a scheme (means of access) and authority (where to access).
A decentralized system would use URNs and any host that could service a request could return the resource. Once a resource was in "the Web" it would be accessible to future requests even if the original source went offline.
This sort of mirroring can be built on top of the Web (CDNs, traditional mirrors, etc) but it is not a foundational component. The authority providing a resource needs to online for that resource to be available.
ARIN and ICANN are centralized. WWW can run over alternative number and naming spaces but in practice doesn't do this except for Tor and to a much lesser extent i2p.
"www is already decentralized", okay but not in reality, not even a little bit.
you can spin up as many nodes on the internet to have as many platforms as you want, nothing is stopping you. getting booted from some company's platform for violating some company's tos does not mean the world wide web is not decentralized
PKI and DNS are centralized and any of the _three_ major browsers rely on them and actively deny features if they are not present.
I mean, sure, packet routing is decentralized, and if you're a military operator, this might matter to you, if you're someone who wants a public voice it's not significant.
That's fair, in December last year we realized that we'd need to implement our own transport layer after discovering that the off-the-shelf library we planned to use wasn't flexible enough for our needs.
But the new transport layer is done, unit tests pass, simulations are working, and we're now cleaning up loose ends so this time I think we really are days or weeks away.
We haven't launched yet but we're very[1] close, hopefully just days away as we tie up loose ends.
This[2] diagram hopefully gives a big-picture view of where Freenet fits in. You install the Freenet software (which is tiny, less than 10MB) and then you can access Freenet through your web browser just like with the world wide web. The difference is that there are no servers or datacenters, it's all decentralized.
Ha, looks like I missed a 2023 drama when the original Freenet was renamed to Hyphanet, and Locutus became Freenet (https://www.hyphanet.org/freenet-renamed-to-hyphanet.html; https://freenet.org/faq/#how-do-the-previous-and-current-ver...)
I used to check out the original version a few times in the past. I loved the overall premise of a uncensored network (for the context: I lived in an authoritorian country that doubled down on censoring online communications), but no system I've tried out (Freenet, ZeroNet, I2P, Tor) grew on me. Too much, uh, weird stuff (or worse), while interesting content was pretty sparse. And I don't produce anything worthy of sharing myself.
The tech could be there, but the society is not. [Not] surprisingly, the social demand seems to be all focused on to work around the Internet censorship with very limited interest in bootstrapping something that would be resistant. At scale, humans always pick the cheapest/lowest-effort option, even if it's obviously sub-optimal and even if it leads nowhere we want (or so I think) to be.
The new Freenet (formerly Locutus) is more of a communication than a storage medium, which should make it less susceptible to the kind of problems you're referring to. That said, even on Hyphanet you really need to look for something bad to find it as the default indexes are curated.
We’re also developing a decentralized reputation system based on the "web of trust" concept. This system is intended to help filter out antisocial content—like spam or worse—without relying on any central authority.
This was my experience with matrix.org, mostly pedos and far right extremists or other weird anti social characters . Turned me off from these kinds of platforms big time.
if you’re talking about the matrix.org homeserver’s room directory - these rooms are strictly against our terms of use (section 6 of https://matrix.org/legal/terms-and-conditions/) and we shut them down, and these days have even frozen the roomdir to stop them appearing.
if you’re talking about the wider network - yes, there’s a subset of abusive users… just like on the web, or the internet, or email, etc. Unfortunately there’s nothing we can do about; it comes with the territory of being an open network where anyone can participate.
My experience of Matrix is more that it’s full of FOSS projects like Mozilla, Ubuntu, Debian, Fedora, GNOME, KDE etc… as well as lots of government users. But ymmv. If you are still seeing abusive rooms as a matter of course, please route details to abuse@matrix.org, where we do actually act on them (if they are on servers we control)
Wow, yeah the parent comment is shocking to me. For me matrix is OSS discussions, like a modern IRC (not that IRC is gone, a lot of matrix rooms bridge to IRC).
Though I suppose if you search for other stuff you can find it.
Though it strikes me as a little odd, I never really associated matrix with anonymity, more with federation, end to end encryption, self hosting.
Perhaps hosting matrix servers as tor hidden services could serve that purpose.
It's a bit strange to even criticize these services for the existence of abusive users. The underlying problem is that abusive people exist in real life: it goes without saying that they'll show up in any context that's accessible to the general public.
Complaining that bad people are using Matrix is about as meaningful as complaining that bad people are driving on the highways or are filling their bathtubs with from the local water supply.
I only ever use a half dozen or so public Matrix rooms (I mostly use it for private chat) but my experience in those is nothing like you describe. Unsavory characters occasionally wander in, just as they do on every other public forum I've used, but they are shown to the door and the normal conversation carries on without them.
I wonder how someone must be using the platform, or how they're looking for rooms that might interest them, to get the impression that it's "mostly pedos and far right extremists or other weird anti social characters."
No need to imply hostility like that, it could just have been a factor of when they tried it.
No hostility implied or intended.
[flagged]
if every bar in town bans nazis or punks, even if the last bar has high minded ideals you kinda know which clientele they'll get
I visited matrix.org once and I somehow only stumbled into the high minded clientele.
I haven't seen any pedos or far-right extremists. Quite the contrary really
[flagged]
NOT (any pedos or far-right extremists) != Gerontophiliacs and Marxists
Joke != formal logic
Those people also use the public roads. Are you also turned off from driving on public roads?
To be fair, everyone with a commute who had or has to sit in endless traffic is turned off from driving on public roads and wishes there were fewer vehicles or that they were the only one on the road.
Paraphrased, "I was driving home, stuck in traffic, and this thought occurred to me, now I know this is bad, but I thought, if half of everyone in this city died, I'd be home by now." - Paul Reiser
When combined with "You aren't stuck in traffic, you are the traffic", the two quotes sum to "Go kill yourself."
Darker than mine, nice.
They'd need to specifically die in their own homes (or otherwise not on the roads), or else the resulting traffic jams would make your drive home even longer.
"What about..."? Isn't a very convincing form of argument in general and in this case it's particularly specious. When using the public roads, people's unsavory/illegal behaviour with respect to images (for example) isn't something that affects other road users, but if you're using a chat application an innocent user may well see/download content that cannot be unseen and is a crime to download (even innocently).
In that context the decision to avoid a chat application if you feel it is likely to put you at unnecessary risk where there are alternatives that will not is perfectly reasonable.
I don't mind "eccentric" personalities in my online spaces. I hate people trying to restrict online freedom more. To a degree that I think the threat of far-right people is pretty miniscule in comparison, but people allegedly trying to combat them let themselves be driven by fear. And the result is usually as bad as you could expect.
There are a lot of excuses for bad policy with the deflection of "at least I am not a far-right unperson".
A chat application usually cannot put me in any risk at all. Public personas might have different problems here, but the rules of PR aren't that new and these cases build a poor general case.
Fair to have the choice of a supervised chat app as long as people don't try to enforce all chat apps to adhere to arbitrary rules due to the latest outrage or panic over certain people, opinions, politics, bubble tea or whatever can be the source of ire.
The solution to this, at least in the case of Tor, is for existing sites (e.g., Wordpress, medium, etc…) to provide one-click onion-site publishing support.
I don’t think it’s a technical problem. Hosting or accessing a site on distributed networks is no more complicated than running or hosting a VPN (e.g. https://novayagazeta.eu/vpnovaya), and organizations host them just fine, and layman people were proven to be capable of setting up client software when they were forced to (e.g. Instagram or YouTube bans in Russia).
I see it as a more of a chicken-and-egg issue. Publishers don’t come because there’s no audience, audience doesn’t come because there are no publishers. Plus, there is no recognition of distributed networks as a solution to censorship - the current non-enthusiast view of them ranges from “haven’t heard about it” through “tried it, found it useless” all the way to “it’s only for pedos and nazis”, which is extremely harmful for any meaningful and socially beneficial adoption, of course.
I’m not sure if those are the actual reasons. Certainly not a technical issue, though.
It’s no longer funny to see how developers again and again are trying to fix problems of real world by creating digital hideouts because this is the only thing developers do the best - programming. Whenever we see a problem with banks, governments, corporations, we immediately try to fix the problem with “own digital alternative”. But hiding from the real problem in a new network doesn’t fix remaining initial problem for the rest of the world. Complex socio-economic problems shall be fixed with socio-economical instruments. No new network could solve problems of monopolies and profit-greedy corporations.
It's a technical solution to a technical problem, that problem being the client-server architecture which inherently centralizes control and disempowers users.
The Internet started out peer-to-peer, but client-server became dominant during the 90s when the Internet became mainstream - because at the time it was the only option for building services that scaled.
Freenet provides an alternative to the client-server architecture that doesn't require compromising on user experience or scalability.
> Complex socio-economic problems shall be fixed with socio-economical instruments.
Missing the forest for the trees.
The socio-economic problems are caused by the limitations & drawbacks of modern tech: It's too easy to centralize control over a platform, whilst keeping things decentralized requires an upfront desire to do so, along with the nightmare that is backwards compatibility.
> No new network could solve problems of monopolies and profit-greedy corporations.
It is a solvable problem. Monopolies & rent-seeking behaviors happen because the tech involved has chokepoints that favor centralized entities.
We know the business processes that should happen for running service X, based on our daily interactions with them and the idealized form of how they should work. Breaking the processes down into components & putting them onto a decentralized compute platform would significantly relax their control over common chokepoints. Dispute processes would be made as public as current government processes, ideally Estonian-like.
> The socio-economic problems are caused by the limitations & drawbacks of modern tech...
Could you please provide some examples?
Society and governance are gradually getting transformed and integrated into a global communication network run by software and hardware. The Earth is becoming a giant computer within which we live.
It's too early to declare that complex socio-economic problems cannot be solved by a new network/infrastructure/etc. Software is one of the "socio-economical instruments" and we shouldn't dismiss its potential for social transformation beyond the control of banks, governments, corporations.
While Locutus is an improvement on the previous Hyphanet model, I'm still concerned about the caching model. In Hyphanet, data would be cached the more it was used, and peers connected over multiple hops. This had the potential to cause huge headaches as authorities began investigating issues such as CSAM, because their understanding of how caching and requests work did not match up with how it actually worked. While, certainly, one could build a subset darknet, all it takes is one bad peer to expose the whole darknet to prosecution.
I played with Freenet when it first came out, but when I realized the implications of this -- which later turned out to be true -- I destroyed the hard drive I had been using to run Freenet, just in case.
Locutus seems less aggressive about caching data, but it still does some caching. Without really digging into the documentation or source code, I'd still be nervous about running it.
Locutus (now Freenet) is more of a communication than a storage medium. It's focused more on allowing people to build decentralized tools like group chat[1] for realtime communication than be a content distribution network like bittorrent.
We're also planning a decentralized reputation system based on the concept of "web of trust" that should do at least as good a job of ensuring users aren't exposed to unwanted content like spam or worse.
[1] https://github.com/freenet/river
That is reassuring. I hope that your team can build this web of trust, because it really is crucial.
It is great to have a tool that allows ideas to be shared, but I think that consent regarding which ideas are shared or promoted is important, not only from a legal perspective, but also from an ethical one.
Agreed, any system that allows people to discover content needs to protect them from stuff they don't want to see, whether it's annoyances like spam, malicious attacks like DDoS, or extremely harmful content like CSAM.
Centralized services don't do a great job of this but I think we can do better with a decentralized approach.
We won't have a perfect solution overnight but we've already built important components of such a system (eg. ghost keys[1]).
[1] https://freenet.org/news/introducing-ghost-keys/
Under EU rules (Digital Services Act) I believe an automated cache is treated similarly to a pipe - you're no more responsible for holding content in your cache (especially if encrypted!) than a Tor node operator is for someone accessing it through Tor.
German police did raid the home of a guy who ran a lot of Tor nodes. Authoritarians don't need any technical excuse to raid your home - they can just do it if they want to scare you away from these platforms, regardless of how they work.
Unfortunately, US case law is murkier. There have been some notable convictions of people using the old Freenet to search or host CSAM. However, the digital forensics used in these cases was pretty sketchy. If someone simply caching data from downstream queries were held to the same standard, they'd also be considered guilty.
I have no doubt that there was other evidence that led to the conviction of these individuals. But, I can only go on what I know, and what I know is that the standard of digital forensics evidence in those cases was subpar.
The old Freenet project is now named Hyphanet and is available at https://www.hyphanet.org/index.html
Fascinating! I thought this was cool as I clicked through, I'm glad to see that it's been a dramatic topic for 25 years, not just some new idea. Hyphanet seems more openly political, which makes sense but is kinda a hard sell for me after seeing the impact Bitcoin had on US citizens vis-a-vis scams, hacks, and misc. bad actors. But maybe that's a sacrifice worth making in the name of residents of less free nation states?
Either way, pretty funny when a software project has to tackle questions like "what is property in the modern era?" and "what is free speech?" in their FAQ. That's how you know you're really pushing against the status quo, I guess...
Those scams are not isolated to Bitcoin. You see them in many content creator industries. They are giveaway scams and templated to fit a niche. You can do them for gambling, gaming like cs-go skin giveaways, makeup giveaways and so on. They are really easy scams to do because you can just wait for the next trendy thing to happen and then target that trend with bot accounts. On youtube they will often just recycle what accounts don't get reported.
Platforms do their best to take care of this, but these scams are so easy to run that you can automate them and spin up new accounts very quickly. Even if there are financial barriers it can still be madly profitable to accept the costs.
There is no clear way to deal with the concerns you mentioned, but I think playing too safe and not exploring technology will quickly put you behind the curb. Most of it can be resolved with good financial regulation.
I mean, I was more referencing the big epidemic in the US recently, “pig butchering” — basically catfishing with more money involved. Certainly no system alone can make it impossible, but I think even the biggest crypto fans in the world should acknowledge that it makes it much easier, and thus much more common. “Go to this website” is a lot easier/more reasonable ask than “scratch off 20 gift cards and send me the codes”, or whatever western mutual is
Sure, crypto exposes you to potential bad actors in the form of scammers, criminals, etc. Fiat exposes you to potential bad actors in the form of banks, governments, police, etc.
So, whichever represents a bigger threat to you is largely a question of circumstance.
> after seeing the impact Bitcoin had on US citizens
That's an odd way of putting it, as if crypto scams were limited to the US.
Fiat currencies are also used for scams, hacks, etc. Does that make fiat currencies a hard sell for you as well?
It’s about rates, accessibility, incentives, and other society-level statistics. There are scams that have become much more common thanks to the affordances of untraceable, unmanaged online currencies, which shouldn’t really surprise anyone.
That's two "what about" arguments in a single topic. You're on a roll.
Would love to see some network like Freenet start to see more adoption - it seems obvious that we need something like this. GNUnet seems to have some interesting ideas and good motivations too.
I will admit that I didn't follow the renaming or possibly forking or whatever happened to freenet / hyphanet / etc back last when I was reading about this. If someone could explain it clearly that would be stellar.
Nostr (https://nostr.com/) is doing reasonably- though I think most apps are still in the social media realm.
Not judging technically, but there are reasons I wouldn't touch it with a 10 meters pole.
https://archive.ph/TLwch
[flagged]
Hyphanet is great but requires some basic skills and hardware for running. Using Nostr with Primal as a client works exceptionally well as a Twitter alternative. It's privacy-respecting, censorship-resistant, and truly an outstanding tool. I'm grateful it exists to counter the current wave of regulatory overreach threatening free speech, even in Western countries.
> I will admit that I didn't follow the renaming or possibly forking or whatever happened to freenet / hyphanet / etc back last when I was reading about this. If someone could explain it clearly that would be stellar.
The reasons for the renaming are addressed directly in Freenet's FAQ[1]:
# Why was Freenet rearchitected and rebranded?
In 2019, Ian began developing a successor to the original Freenet, internally named “Locutus.” This redesign was a ground-up reimagining, incorporating lessons learned from the original Freenet and addressing modern challenges. The original Freenet, although groundbreaking, was built for an earlier era.
This isn’t the first time Freenet has undergone significant changes. Around 2005, we transitioned from version 0.5 to 0.7, which was a complete rewrite introducing “friend-to-friend” networking.
In March 2023, the original Freenet (developed from 2005 onwards) was spun off into an independent project called “Hyphanet” under its existing maintainers. Concurrently, “Locutus” was rebranded as “Freenet,” also known as “Freenet 2023,” to signal this new direction and focus. The rearchitected Freenet is faster, more flexible, and better equipped to offer a robust, decentralized alternative to the increasingly centralized web.
To ease the transition the old freenetproject.org domain was redirected to hyphanet’s website, while the recently acquired freenet.org domain was used for the new architecture.
It is important to note that the maintainers of the original Freenet did not agree with the decision to rearchitect and rebrand. However, as the architect of the Freenet Project, and after over a year of debate, Ian felt this was the necessary path forward to ensure the project’s continued relevance and success in a world far different than when he designed the previous architecture.
[1] https://freenet.org/faq/#why-was-freenet-rearchitected-and-r...
Thank you very much, I've been browsing a bit since your thorough reply, looking forward to having more of an investigation one of these days when I get a few hours free in a row to have a good go.
I wonder how much Freenet would be considered “web3” and “blockchain” by the HN crowd, considering it explicitly uses smart contracts and transactions signed by self-custodied private keys
Guess it’s all in how you present things and what terms you avoid :)
For what it’s worth, I am interested very much in decentralized systems and smart contracts, having built them and also running a YouTube channel where I interviewed people behind the projects… including Ian Clarke and freenet: https://www.youtube.com/watch?v=JWrRqUkJpMQ
Personally, I don’t consider Freenet and Hyphanet to be "blockchain" in the modern sense, and given how much the meaning of "web3" has changed in recent years, I think it might evolve further.
Freenet stands apart with goals and ideals that are quite different from today’s distributed applications, with a stronger focus on privacy and access to information.
Neither Freenet nor Hyphanet are linked to cryptocurrencies or financial speculation. I see them as decentralized networks created to ensure freedom of expression, privacy, and access to information in an anonymous and censorship-resistant way, without any intrinsic connection to cryptocurrencies or financial systems. And that's great!
I also believe the project has gained a certain credibility over time, thanks to the consistent work and vision of its developers.
I’ll check out your interview with Ian Clarke!
Freenet and Hyphanet aren't blockchain, although the original Freenet did pioneer the cryptographic contract back in 2000 (we called them "signed subspace keys"), which a few years later would form the basis for Bitcoin.
Freenet is designed to be a general-purpose platform for building and distributing decentralized systems like group chat[1], social networks, search, really anything that people use the Internet for today. Some people think of blockchain this way but blockchain is much more specialized (eg. group chat on blockchain wouldn't scale).
[1] https://github.com/freenet/river
When I was ~14 I downloaded Freenet after finding it on Stumbleupon.
After wondering why my laptop was loosing RAM for a few months, I realized I had been running a node continuously.
Great to see it's still alive though.
That's pretty funny 8-)
Their declaring their freedom and independance in a Matrix chat channel 8-)
This reminds me of the recent Guardian article headlined:
"Americans who believe in democracy have no choice but to vote for Harris"
https://www.theguardian.com/commentisfree/2024/oct/26/vote-k...
How can it be democracy if there's only one choice?
We're working on a decentralized group chat app for Freenet called River[1].
We'll migrate our channel from Matrix to River once it's working reliably.
[1] https://github.com/freenet/river/
in terms of namespace collisions, beware https://www.river.build is already a thing
Noted. It's true that there are a few projects using the name River - which is unsurprising for such a common word, but nothing with significant traction so far as I'm aware.
Interesting that the new freenet is written in rust.
I always thought the worst part of freenet was that it required java.
I haven't used freenet in a long time. Last time I did I'm pretty sure satoshi was posting the og bitcoin releases on the message board (freechat?). Crazy times.
Yeah, there was a decent argument for Java in the early 2000s before it was obvious that Java would fail to gain traction on the desktop.
Rust was the obvious choice when I started work on the new Freenet in 2019.
I interviewed the founder of the original freenet and the current freenet, Ian Clarke (among others) on our channel about decentralized systems, might be interesting for the people here:
https://www.youtube.com/watch?v=JWrRqUkJpMQ
It is on our channel https://youtube.com/Intercoin where I tried for a few years to have really deep dives with some of the too people on technology, sociopolitics and regulations around decentralized systems.
On the mention "original freenet"
In the late 1980s early 1990s, FreeNet was not (just) a web domain name.
Freenet was decentralized idea of how the public (me) could get on the internet without already being a member of a privileged institution (or rich and sophisticated).
I believe the idea of what the freenet was prior to becoming a brandname is worthy of being remembered.
I'm not much of a tube clicker here, so apologies if this is redundant with the video.
Did Freenet ever get over its "content moderation" problem? Don't want to summon the FBI.
The new Freenet is more of a communication than a storage medium which makes content moderation less of an issue, but we're planning to build a decentralized reputation system based on the idea of "web of trust" that should do at least as good a job of protecting people from stuff they don't want to see as today's centralized services.
resisting censorship is fundamentally at odds with supporting censorship
> Did HTTP over get over its "content moderation" problem? Don't want to summon the FBI.
HTTP is not a protocol that automatically distributes content to participating nodes.
If it is a distributed computer: how to prevent crypto mining?
Good question - computationally expensive contracts will be removed automatically as part of a mechanism that keeps track of cost/benefit to the peer for each contract.
Leave the field to the censors, create an alternative, and hope that they will not pursue you? Or stand your ground and fight them wherever they are? That is the question.
Let's do both, since one fosters the other.
www is already decentralized
Theoretically you can run a web server at home but you'll have a problem if you start to get a lot of traffic or you have trouble with your internet connection. Your website will be trivially easy to DDoS.
Services you create on Freenet will scale automatically and are immune to DDoS.
it doesn't really scale in the traditional sense, it just pushes the 'damage' out to the entire net.
freenet overall is one of the most bandwidth and storage intensive platforms out there. I understand why -- but I say this as a means to say that it doesn't really deserve direct comparison to the open web. it's not the same thing, even if the work overlaps -- it's a lot more work.
"Damage" is misleading. A DDoS only harms when concentrated on a few peers. Freenet distributes load across the network, autoscaling in response to demand and preventing overload.
As for bandwidth and storage, I think you’re referring to the old Freenet (now Hyphanet). The new Freenet is optimized for lighter services like group chat and isn’t designed for heavy data sharing like BitTorrent. It should be much less of a bandwidth hog.
To some extent yes.
However to host something yourself you need a lot of things, for example FTTH to host it at your home, or a hosting provider; then a domain name and other things. These can be taken away from you.
Pretty sure that freenet also requires some sort of ISP to provide a ethernet connection to your machines, which can still be taken away from you. This doesn't change the fact that the www is decentralized already
Eventually Freenet will work over mesh networks like Lora and 802.11ah - meaning it will be entirely decentralized including the communication infrastructure.
You don’t need fiber to host basic services, and once you do, it’s not really a problem.
You need a public IP, though, and many home contracts put customers behind NAT.
How does freenet let you own something outright?
On Freenet you own things cryptographically, typically by possessing a private key.
This is similar to a Bitcoin wallet although Freenet isn't a cryptocurrency, it's a general-purpose platform for building and distributing scalable decentralized services.
Your point being?
When many people have the ability to publish independently without relying on a central service, then it’s decentralized. The World Wide Web was initially designed to be decentralized, with the idea that anyone connecting to the internet could host a web server.
In practice, however, this didn’t quite work out. Most people publish through centralized services like Instagram, to name just one.
There are two main obstacles to achieving decentralization. The first is technical difficulty: not everyone wants to learn how to run a web server. The second is reliance on foundational services like domain names and hosting, which can be revoked. For example, if the authorities think you did something illegal, boom, your domain name got confiscated.
So, no, in practice, the World Wide Web isn’t truly decentralized. But at least there remains some possibility for it.
Thanks. I think we've always had the ability to publish content on the internet without relying on external services, and we still have. It's just not something that most people are willing to do, because it takes expertise and resources, and corporations can do it better by specializing in providing these kind of services. This is why people use them. For the same reason people don't usually make their own clothes, furniture, etc., either. Specialization and division of labour allows us to be much more productive and as a result enjoy a higher standard of living than we would if every individual had to produce everything they consume themself.
The WWW is also not really decentralized anymore because 90% is hosted on one of a handful of hyperscalers. Microsoft, Google, Amazon, Cloudflare.
Eh, besides Cloudflare, you're pouring money down the drain if you're using any of those. Remember that recent thread where some open source project was spending $100/month on a Vercel database handling 3 requests per minute that could have been a $4 VPS?
Non-hyperscaler server hosting is a pretty competitive business that doesn't need further decentralization at this time, though it's not a bad thing either.
> The World Wide Web was initially designed to be decentralized, with the idea that anyone connecting to the internet could host a web server.
Yeah not really. If the Web was designed to be decentralized it would have used URNs as content identifiers instead of URLs. A URL is specific to a scheme (means of access) and authority (where to access).
A decentralized system would use URNs and any host that could service a request could return the resource. Once a resource was in "the Web" it would be accessible to future requests even if the original source went offline.
This sort of mirroring can be built on top of the Web (CDNs, traditional mirrors, etc) but it is not a foundational component. The authority providing a resource needs to online for that resource to be available.
ARIN and ICANN are centralized. WWW can run over alternative number and naming spaces but in practice doesn't do this except for Tor and to a much lesser extent i2p.
"www is already decentralized", okay but not in reality, not even a little bit.
Yet monopolized enough that this decentralization does not prevent deplatforming.
you can spin up as many nodes on the internet to have as many platforms as you want, nothing is stopping you. getting booted from some company's platform for violating some company's tos does not mean the world wide web is not decentralized
PKI and DNS are centralized and any of the _three_ major browsers rely on them and actively deny features if they are not present.
I mean, sure, packet routing is decentralized, and if you're a military operator, this might matter to you, if you're someone who wants a public voice it's not significant.
[dead]
It's taking a while for the new version to come out. It has been "weeks away" for like a year now.
That's fair, in December last year we realized that we'd need to implement our own transport layer after discovering that the off-the-shelf library we planned to use wasn't flexible enough for our needs.
But the new transport layer is done, unit tests pass, simulations are working, and we're now cleaning up loose ends so this time I think we really are days or weeks away.
Glad to hear that progress is being made. I can't wait to try it. Thanks!
It seems interesting, I've seen this before, but i'm lost. How do I get it, does it incorporate with the OG internet or what?
We haven't launched yet but we're very[1] close, hopefully just days away as we tie up loose ends.
This[2] diagram hopefully gives a big-picture view of where Freenet fits in. You install the Freenet software (which is tiny, less than 10MB) and then you can access Freenet through your web browser just like with the world wide web. The difference is that there are no servers or datacenters, it's all decentralized.
[1] https://freenet.org/news/weekly-dev-meeting-2024-10-11/
[2] https://docs.freenet.org
First step in checking out Freenet - Click on link to see some video about Freenet - https://freenet.org/news/ghost-keys-ian-interview/
Google video saying "Sign in to prove that you are not a bot"
Ian? Freenet? Really?
Am I a joke to you?
https://pbs.twimg.com/media/GbE3OkwWwAA25N_?format=jpg&name=...
Look under the video and you'll find links to the same video on Rumble and X, take your pick.