Is this an abuse of the ServerHold status? Was this the same mechanism used to delist a Gaza video archive recently?
> This status code is set by your domain's Registry Operator. Your domain is not activated in the DNS.
> If you provided delegation information (name servers), this status may indicate an issue with your domain that needs resolution. If so, you should contact your registrar to request more information. If your domain does not have any issues, but you need it to resolve in the DNS, you must first contact your registrar in order to provide the necessary delegation information.
To add a little context, this suspension comes immediately after Anna's Archive publicly implicated themselves in the Spotify scraping "hack" in which they downloaded nearly the entire content library of Spotify and was preparing to release it publicly (~300TB worth) via torrent.
Archiving it and publishing it are different things.
More importantly, they may sabotage their mission: If Spotify shuts them down, their exiting archives and especially future archives may be effectively lost.
I guess I should say more accurately: Their mission is to both archive it and publish it. They seem to be explicitly against copyright, on principle. Which I greatly respect.
Yeah, it seems to only be a problem when you're a human being remixing the culture you grew up with.
Meta can admit to soullessly scraping books they don't own for their for-profit AI datasets [1], and it's not a problem because they're Meta. But if you're an artist? Nope. Sampling in hip hop songs, for example, is in a "complex legal gray area" (translation: "it's illegal but we don't want to admit that out loud") [2].
Fortunately, Spotify does not have that power. Annas Archive is not based in US or EU jurisdictions. They can make access for normal people a bit harder, but not shut it down.
Annas archive is not based in the EU (sorry for being not clear). So the law in EU is limited to enforce a ban. In germany it is already "banned" via ISP but just DNS.
But the real servers are hosted in kazachstan or russia I think. And they do not cooperate so much with EU courts.
So unless the EU installs a great firewall like china, they cannot really shut it down.
Presumably the opposing party is residing in non-US-or(and? depends on the order of evaluation)-EU territory, but I might be mistaken. "They" refers to both sides in the parent comment.
They are, but archiving without publishing is pointless.
I occasionally wonder how many enormous collections of culture like that of Marion Stokes[1] have been lost because their curators made no effort to realize the value of their collection.
Most archives - the ones in libraries, etc. - are not published, except they are available to qualified people who physically travel there. Most are not even fully indexed - nobody knows all of what's there.
My perspective is compatible with this fact. An archive that approximately nobody can access and/or nobody knows what it contains has no value to society at large, except the potential that it may some day be published.
The good news is I'd guess the number of (nonreligious/nonproprietary) institutionally managed pointless archives is dwindling.
> They are, but archiving without publishing is pointless.
One may collect/archive now (when the data is, well, "available"), and publish later, when copyright expires and the material will likely be harder to obtain.
They stated that they would pass the information on to other archivists and public/private trackers no? They obviously have backups, since there are multiple users seeding Gbs and even TBs of data. Mirrors can be created as well, like TPB.
They didn’t come anywhere close to the entire content library, the 300TB represents about 33% of Spotify, though it is close to 100% of the played music.
To be completely fair, I am not certain what it means for a track to be "virtually unplayed".
First off, it was striking to me how little of the "top 10 000" they published back on Christmas I recognize. I'm not sure what I expected, but 10 000 sounds like a big number, so it seemed likely to me, that if I get a random song from my playlist I could find it there. It turned out I hardly can find an artist I recognize. Ok, I can recall a song from Lady Gaga and even Billie Eilish, I've heard of Bruno Mars (cannot recall any song), but I have no idea what is "Bad Bunny", "Doechii", "Drake". I mean, I think I do have a pretty good idea what these things are (abstractly), and I probably wouldn't want to listen that. And while I knew that all this stuff is very popular, I didn't quite realize how little place in the top-10000 it leaves for the music I (and everyone I know) actually listen to.
I didn't download the metadata they released (it would be hard to process it on my laptop anyway), but now I wonder how much of my 3 TB music collection is in top 100 000, or heck, even top 1M Spotify, or on Spotify at all.
I also am sometimes surprised how little scrobbles some tracks get. I didn't bother to find out what this means, how many people still scrobble to Last.fm or ListenBrainz, but it is just surprising when I see that a track that I didn't consider to be obscure was scrobbled like 50 times this year.
So I'm saying that music worlds seems to be terribly fragmented, even more than I imagined. So the very premise of AA backing-up 97% of Spotify (by the number of plays) may be much lesser achievement at "preserving culture" than it may sound. And of course we are about 8 years too late to backup everything, since by now half of it must be generative NN bullshit. And I'm not even sure it's in those leftover 3% (bots listen to bot-generated music too, right)?
You might not listen, but surely you have heard of Taylor Swift, Justin Bieber, Ariana Grande, Ed Sheeran, Coldplay and of course Christmas Staples of Mariah Carey and Wham?
First off, this is not the top we are talking about, since there is one that AA provided[0]. I am not sure what it matters which names exactly I've heard of, but if you are that curious: I don't know what is Ed Sheeran and Wham (but cannot vouch I've never heard their music in a supermarket), but I definitely remember "Coldplay" being mentioned in a joke onstage by a NIN member[1], but I didn't bother to check out what they are. I can imagine the faces of Taylor Swift & Justin Bieber, but cannot name any song, and I'm sure I've heard Mariah Carey somewhere, since that name is around longer than Rihanna. I have a song or two of Ariana Grande in my playlist though.
Are you sure? See, my point is a conjecture (based on a reasonable assumption that I cannot be that special), that there must be really a lot of us "outliers" out there (so I'm not even sure it's reasonable to call us that).
Let's reiterate. I am well aware that more people listen to that Bad Rabbit, Taylor Swift or Justin Bieber than they listen to <random name from my playilist>, it's not really a surprise. There even is a special name for people like that, it's "celebrity". In fact, that's probably how most people who are into music (including myself, I might say) would categorize them, as "celebrities", not as "musicians" (though, mind you, of course they are musicians, as everyone who ever sang a song is, it's just that when I hear the word "musician" I don't necessarily think of Taylor Swift). Hence these people indulge themselves for not knowing who these guys are, explaining it that "they are not into celebrities".
And it's no surprise that a lot of people listen to celebrities. I mean, if Trump would release a song right now, it would become #1 on Spotify in no time (for a very short time, but still). Well, maybe not #1, but close.
But I also suppose there are a lot of people who are into music. Maybe not so many, as there are people who are into celebrities, but it's still a lot. And after seeing that top-10 000 I suddenly find it very plausible, that a lot of tracks these people call "massive hits" may turn out to be "virtually unplayed". And hence not in those "97% of Spotify (by # of plays)" that AA archived. I am not even claiming it, I'm just saying that this doesn't seem to be impossible.
For instance, any DnB fan would say that "everyone knows Noisia and Black Sun Empire". It would be absolutely laughable attempt at "preserving human culture" not to include them. Surely all of their tracks must be at least in top-5M, right? Well, after seeing top 10K I'm not so sure anymore.
Maybe you've never heard of them, but surely you've heard of Prodigy. Not a single track from Prodigy on top-10K. Or Chemical Brothers. Or Burial, or Placebo, or Nighwish, or King Crimson. These are very famous names in respective circles. There are 2 tracks from Massive Attack — both featured in super-famous movies and trending on TikTok right now. For God's sake, there are only 8 tracks from Madonna in top 10K. Versus 26 from Imagine Dragons and 124 from "Bad Bunny", whatever it is. How do you like Madonna for an obscure artist?
So, my point is that there may be a lot of people listening almost exclusively to "virtually unplayed" music. Entire discographies of (niche) cult-artists may turn out to be buried in these 66% of "virtually unplayed" tracks.
I guess I should just get the metadata and check, but I'm pretty sure that would be outside of capabilities of the hardware I have on hand, so I'm not sure how to go about that.
1) 10,000 tracks really is not a lot. It sounds like a lot, but isn't. My own - relatively small - collection is nearly double that.
2) 10,000 tracks... out of 256,000,000 that AA archived.
I'd be very interested to see some more analysis done on this, particularly as it relates to, say, Last.fm statistics - but I suspect the missing music is not as significant as you think.
In any case, even if every one of those "niche" artists you list are missing from this collection, I don't think it's fair to say it's a "laughable attempt" - it's certainly better than nothing, even if it's not perfect.
The funny thing is, since the advent of streaming I no longer listen to the radio. I listen to new music, but little pop music, and I have never heard a single track from Swift, Bieber, Grande or Sheeran. Coldplay is the only act I like on that list, and the streaming services are pretty good at only playing what I like.
If they were pre-streaming artists I probably would have heard a lot of their catalog because radio played it over and over. Unfortunately you just can’t get away from the Christmas music.
> For now this is a torrents-only archive aimed at preservation, but if there is enough interest, we could add downloading of individual files to Anna’s Archive. Please let us know if you’d like this.
If it is torrents only, what relevance does unregistering the domain make?
An while back, another site started with a pile of pirated music, and that was allofmp3.com Remember those peeps?
Their business model was to sell music by selling bandwidth. Basically is was all the music you want charged by the megabit download.
Pop titles were $0.10 to $0.25. A whole album at 256mbps was roughly $3 give or take.
What got me really thinking was how great the UX experience was. At the time, few came close.
The end of that site was packaged up with Russia's entry into the WTO.
I seem to remember hearing about huge torrents out there too. The right infohash can point a person to huge archives of various kinds, books, video, academic papers, music, the WikiLeak insurance files, which is password protected, as perhaps all of these are.
Used to work for a registry. Registries do not just arbitrarily serverHold things they don't like. Even in blatant abuse cases, they'll reach out to the registrar to clientHold it or delete it instead.
I suspect this came from a court order. The only time I remember serverHolding things were from court orders or other legal requests(FBI/DHS/etc). Though the latter would often just ask for the nameservers to be changed instead.
Not my experience. My personal .us domain was placed in serverHold in April 2025, after Godaddy Registry sent my registrar an email including the following:
"If we do not receive a response and the abusive content remains active, the referred domain will be suspended (-ServerHold-) twelve (12) hours after this notification, to limit any potential damage."
Yes, only 12 hours' notice. And this was in response to what looked like an automated scan, with a screenshot of a directory listing showing some files with (accurate) modification times in 2011. This would not have passed more than a quite cursory human review. And there is no chance that anyone was using the files, as they were at an unused path (which happened to be exposed via directory listings).
(How do I know that the files haven't been infected without my knowledge? Because I checked, but also because I know why they were flagged as malicious. The files contain an exploit, and will try to exploit your browser if you open them. But the exploit only works on 2011-era iOS! The iOS version numbers are right there in the filenames! The files are not real malware, but part of a jailbreak, JailbreakMe, that installs software of your choice on your phone… if your phone is an iPhone from 2011. I continue to host the jailbreak on purpose, for the sake of a handful of people that still use these devices. I guess you can call it retrocomputing. This particular subset of files was unused though.)
.US is a bit of an oddball, at least compared to gtlds. The US DoC puts rather strict requirements on the registry under contract, including this but also spot checks that owners are US citizens, and makes it the responsibility of the registry to take action on it.
Those scans are automated but supposed to have human-in-the-loop, though that doesn't really mean much either. Unfortunately, once the process starts, receiving no response means an automatic suspension.
In that case, it was the registrar (not the TLD owner) that put the domain under clientHold and clientTransferProhibited, etc (which disables DNS lookups).
The Genocide.live site is now back up (just yesterday) after they raised a fuss on social media, and were able to get the domain unlocked to transfer it to Trustname (out of Estonia) as their new registrar.
(The Namecheap founder/CEO Richard Kirkendall surprisingly came across as surprisingly unaware of how anything to do with domain name registries, TLDs, and DNS works on Twitter in an exchange where he thought the entity running the .live gTLD was the archive's new registrar [1], claimed the domain was unlocked for transfer when it was still on clientHold [2], and a raft of other silly mistakes.)
I'm pretty sure serverHold being applied to domains targeted by court orders is normal. It also has all of the xProhibited statuses which ICANN notes are "usually enacted during legal disputes".
After publicly shaming NameCheap on Twitter, NameCheap backed down a little. They've still de-platformed genocide.live, but released the DNS name so the owner could transfer it out. IIRC, they are using a "free speech" provider in Estonia now.
The fascist website Stormfront lost their .org in 2017 with some very unclear procedural action, but I don't think it was a ServerHold. It was offline for quite awhile and I think genuinely hurt the site. Although it looks like it's back now?
I think this was genuinely due to the Norwegian Massacre. The guy who did it had an account on stormfront. The whole thing exploded and the owner had to explain that he was not in contact with the mass murderer.
Thanks for linking this article! However, the full description of the incident seems that Namecheap blocked the domain, the CEO got into tweet arguments justifying it (and taking inconsistent positions), and after the backlash said it was a hasty block by the mod team, without an apology.
I generally give NC the benefit of the doubt (because of their principled stance on Ukraine, which to me means they can be principled elsewhere too), but in this case it's very difficult to explain it away.
> Was this the same mechanism used to delist a Gaza video archive recently?
The mechanism used to shut down a website whose videos were fake and that was funded by actual terrorists? I don’t think Anna’s Archive is in the same category, no.
Anna Archive is a notable project. Wikipedia displays links to projects. It's not about "being used as DNS" but about providing basic info about the topic, URL being an important part of it.
I'd like to see the intersection of people that think AA is some last bastion of Internet freedom, while simultaneously trying to get all of KF's routes blackholed.
The problem with playing internet police is there is always inherent bias.
But this mechanism is used to circumvent DNS blockade. Wikipedia may be next to moderate if they can force DNS providers and even the org registrar to give in, wikipedia could fold too.
DNS is another layer. The URLs shown on Wikipedia will still have to be resolved to IP addresses, which is where DNS comes in. Referring to Wikipedia for the URLs/domains does nothing to circumvent DNS blockades.
Evenn though its onelayer down - the same tactics that were used to suspend/takeover domains would still apply , at the end of the day one still has to get the IPv4/IPv6 address from someone(who can be coerced).
When Trump pressures RIPE NCC or APNIC to deregister an IP address block, that's the end of the internet as we know it, and the return to national networks with very limited interconnection. Even Russia still has address registrations despite being sanctioned.
Alternatively they pressure USA ISPs to block the addresses. That's already regularly done but it probably won't be enough to satisfy the extortion industrial complex which is out for blood.
A quick look at the last few administrations is all anyone needs to see how this one interprets the powers and duties that come with the office.
One of my favorite phrases coined during the last Trump administration was something like, "not just wrong, but wrong beyond normal parameters." It basically meant exactly what we are discussing here; namely, being an outlier of some sort.
I specifically mentioned foreign policy. There, I don't remember a single US government that was not a net negative for the rest of the world (Israel excluded).
It circumvents the purpose of the DNS block which is meant to prevent people from easily finding the site. Anna's Archive can easily register new domain names and put them on Wikipedia, thus allowing people to easily discover the new location of the site.
Of course many sites can serve as "DNS" - Reddit, Github, X, basically anywhere you can put a URL. So DNS blocking is relatively useless.
> Linking to illegal services can be illegal, that’s why.
What is illegal in one country can be illegal everywhere.
I don't remember Wikipedia removing LGBTIAQ++ articles just because that's illegal in Iran.
If a government thinks Wikipedia is illegal in their country, they can force local ISP providers to block it, but it's not Wikipedia's responsibility [1] to censor itself.
The Wikimedia Foundation is a US corporation. There are national chapters in some other countries, which are corporations in the respective country. The internet isn't the Wild West; websites are subject to the laws of the countries they operate in.
Countries can pressure them for many reasons, fairly or not. Under pressure, Wikipedia founder Jimmy Wales interfered with a page about the war in Gaza (though I don't know th outcome of that).
I still remember the hiatus around the pirate bay . org going down back in the day. They updated the landing page on the alternative domains to include a hydra above the pirate ship where above each hydra head there was a domain name[1]. I thought that was a great comeback by the maintainers.
Those guys should get a prize for resilience alone, they put many government run services to shame. The only thing all this legal stuff seems to have done is to harden them even further.
Do not go there without ublock! The ads on there, especially porn ads, have gotten out of control. Who is it that's making money off of thepiratebay these days?
I would urge people to not go there at all. Their moderation is imo pretty lacking, and malware sometimes slips by. There are much safer pirate websites out there (although you should still only visit them with uBlock Origin installed)
Well! Not really a surprise, is it? We knew that DNS censorship was coming sooner or later. The real surprise is that it lasted so long. X509 (TLS) PKI is also probably being abused right now. We know how differently the administration sees services like these that are actually beneficial to humanity. Choosing to rely on services like DNS and PKI exclusively for the function they provide is a very bad idea going forward for us normies.
We should have considered these centralized and corporate driven core infrastructure components as interim measures while more independent alternatives were being developed. We have a few different alternatives right now. Can't we just choose one and switch over? (something not based on blockchains.) Something like GNU Name System, may be?
PS: They will probably block the IP if Server hold/DNS block is not useful anymore. That's a different problem though.
I wonder if in the interim period between DNS blocks and full IP bans, some websites will start distributing their IP addresses directly as the "official" way to connect to them.
The question is, how will you obtain that IP address from them? DNS was the answer to that question. So when you ask that question again, you're effectively asking for a DNS alternative.
I mean, all of this is a hypothetical, but wouldn't you find the IP in the same way how you'd find the domain now? If there's means among communities to disseminate new TLDs as these websites are forced to move back and forth, then there's nothing stopping them from sharing IPs instead.
I always wonder why sites like Anna's Archive (and a lot of torrent tracker sites) don't provide .onion addresses. I imagine a significant proportion of their traffic comes from the Tor network, and onion addresses don't have this same weakness that regular DNS addresses have as they're just the key. I can't imagine if they have the servers already up and running, running an .onion address is that much more work considering the resilience it would add and it's existence may even encourage users to access through Tor if they promote it as the primary address.
It has been a few years since I last used Tor, but even back then you could download at megabit speeds over the network. I would assume that the situation has improved, I’m surprised that you seem to report the opposite.
If they're going to have .onion and .p2p addresses, they should also add Yggdrasil. Yggdrasil is easier to have always running in the background so I don't have to toggle it on just to access AA.
EDIT: Never mind, I forgot how the Tor software works. I still think also having Yggdrasil would be nice, though.
Seems like great publicity for Anna's Archive. I've heard an increasing amount about Anna's Archive over the last 12 months. It has popped up a lot. I wonder if they've seen their traffic spike a lot.
I recommend Anna's Archive get a Nostr account. Once they finally have a solid court order to seize domains, generally the rate at which they get seized accelerates greatly. Nostr is the only decentralized manner (no, Mastodon/fediverse is dependent on domain names, which are getting seized by courts in relation to this -- it is not decentralized at all when it comes down to it) that people can reliably use to have a latest content feed distributed.
"Check Wikipedia to evade the court order" just encourages legal action against Wikipedia. Even linking to copyright violations is, under current court precedent, able to bring civil liability upon third parties. It is draconian and our framers would have considered it a clear First Amendment violation, but unfortunately the current jurisprudence says that is the law.
Tor and I2P is a better fit. Nostr is very weird. It sells itself as decentralized, but basically all frontends provide same several relays.
When those relays get subpoenas and remove your resource, you're done. You can use some unknown relays to publish, but who's gonna use them as clients outside of the defaults? It's effectively designed for shadowbans.
Many clients automatically seek, or prompt an action in 1 click to retrieve content from additional relays that a Nostr pubkey announces if said content is referenced but not available on already subscribed relays. As a publisher, you announce what relays you are currently publishing to in your identity metadata. So even if you don't specifically subscribe to a smaller relay, you can still access the content on it.
Tor and I2P are great technologies. ZLibrary, for example, runs an excellent Tor hidden service and it is usually the most reliable way to access news from the site. However, this did not remain true for a while when two of the operators were arrested. Tor and I2P require you to have infrastructure online. The point of "check Wikipedia for news" is that you can have something persist even if you do not have your servers online. Nostr is the best technology available in this category.
You announce all the relays you publish to to the relays you are publishing to. If someone quotes your post in a post of their own, in many clients subsequent readers will be able to retrieve your content from the relays which they don't currently follow. Relay discovery mechanisms grow progressively better. I don't know of any pubkey ever banned from major relays. The operator of Damus, one of the more successful clients in history and one of the main default relays, openly engages with dissident personalities in a welcoming manner. You could probably get a "filtering transparency report" from him and others and ask if they've blacklisted any specific pubkeys and why. I am unaware of any that are currently blacklisted, and generally the network seems to defer to WoT mechanisms by clients to blacklist content.
Regardless, this remains a far more resilient persistent source of information that you don't operate than "check Wikipedia".
I still don't get how to bypass a theoretical block, you need to access at least one of the current major relays or a side channel to find a follower, only аfter that you can re-translate the new set of relays. The autosub is good, but IMO the current major operators have an Elusive Joe situation because nostr is very small. Things will change as soon as people with money and government connections see it as a problem.
I totally agree with the Wikipedia argument though.
I think relay discovery by clients has improved by leaps and bounds and will continue to do so. Previously you had to give some complicated "follow me at such and such relay" thing. These days you can just tell people your npub and things seem to "just work". I don't know what's going on under the hood, I can only say that clients figuring out which relays have content seems to have improved drastically.
To improve user experience, Nostr clients typically pre-load several large relays. In fact, Nostr also supports using NIP-19[1] pointers to pass custom relay hints to the client, similar to the tracker in BitTorrent and Magnet. Furthermore, I believe that with Let's Encrypt now offering free and widespread IP certificates, domain dependency issues will be further alleviated on Nostr.
Tor, sure. i2p requires some proxy config in your browser and you need to run a service in the background explicitly. I wish they'd release a dedicated client like Tor does.
I'm not sure why that would require more friction or why my comment was downvoted: i2p is harder to use than Tor. Adoption will be hindered by it, even in tech circles. I want to fire up an executable and maybe click through one prompt but not have to remember all the configuration steps the next time I want to use i2p.
The current focus are new rust and go implementations, but embedding lib for applications is also in the roadmap.
I also agree on hindrance. I don't understand why they don't provide a simple docker-compose at least for daemon deployment for immutable management with controlled scopes. There is an image in the dockerhub, but no proper instructions. People have to spend several hours to ensure that everything works correctly.
Freenet (aka Hyphanet) is 25 years old... it's been "stable" for decades. Biggest problem is that content is sort-of hosted locally and considering how much child porn and other criminal content is in its darker corners tha can be risky.
Better have a .onion. It's almost impossible to seize and you control the keys. YOU ARE THE OWNER, not some for-profit registrar. Onions should be the default, it's secure (you own the keys), decentralized and far better than relying on CAs for encryption.
Tor and I2P are great technologies. ZLibrary, for example, runs an excellent Tor hidden service and it is usually the most reliable way to access news from the site. However, this did not remain true for a while when two of the operators were arrested. Tor and I2P require you to have infrastructure online. The point of "check Wikipedia for news" is that you can have something persist even if you do not have your servers online. Nostr is the best technology available in this category.
> It's almost impossible to seize and you control the keys. YOU ARE THE OWNER
This also remains true for Nostr.
But furthermore, as an operator of several Tor hidden services corresponding to public web services. I can assure you that many users, especially those on mobile devices, will stop using your service in large numbers if you direct them to a hidden service. iPhones don't allow background processes without special dispensation from Apple so the Tor/I2P circuit dies every time someone switches between apps. It's also an extreme development challenge, as they don't allow subprocesses either, and then of course your app will have to abide by the GPL at least for I2P (nonstarter for some). "Just ruin your experience for all iOS users and switch to the GPL for all your client code" is not a realistic suggestion. Not that Annas-Archive has a their own client app.
If they can't figure it out anything else, I think Tor is the most plausible tech to be used.
What are the alternatives if these other services don't provide enough traffic to sustent the download speed of the files? Something old like USENET certainly can't be used anymore.
I hope they follow the same pathway of The Pirate Bay or Rutracker.
Operational excellence is of course dependent on the operator but I would still think it's far easier to bring up onion as it's disposable and works behind NAT'ed VMs which makes it further easy to run.
I don't know anything about Nostr since it does not focus on anonymity and isn't as old as Tor (more than 2 decades of research and application), I wouldn't rely on Nostr for anything serious.
Laundering the code through OpenAI to see if the GPL sticks through training, would make for an interesting court case if you asked ChatGPT to write an I2P clients "from scratch" for a closed source iOS client.
>It's almost impossible to seize and you control the keys. YOU ARE THE OWNER, not some for-profit registrar.
You may own the keys but the non-profit The Tor Project owns the network. And when they decide to shut it down your "ownership" of the domain keys doesn't matter in the slightest. You might think this is a silly scenario but actually it happened in 2021/2022 when the tor project unilaterally decided to kill the entire Torv2 network and all domains were made inoperable. All links between sites, everything that made .onion a web, was lost.
The Tor Project does this whenever they feel that there's a security issue. It will happen again.
As someone that spent 10 years building completely legal community sites on the .onion network with the delusion of ownship it really hurt me. I'm never using .onion again. It is not a place to try to build communities. It is only for people that need 'security' as a highest priority and don't care if everything gets wiped out.
They don't own the network. The people who run the relays do. v2 wasn't shutdown in an instant. It was necessary and you could have just redirected your users to v3 and tell them to use it instead but you had to whine about your short-commings on Tor?
It's not only for high-security. It's for the state-of-the-art anonymity.
All the links between .onion sites broke when the relay and other infrastructure operators started running the broken (no Torv2 support) releases the Tor Project put out. All the writings of sites about each other. Everything that made it a web.
It doesn't matter that it was technical possible to try to manually reach out to random visitors of my sites and try to tell them that the entire domain name was changing. That didn't fix the web or links aspect at all.
They did not. And many apps (Ricochet Messenger comes to mind) were not visited by a web browser. So it isn't like you could announce an HTTP 302 and just seamlessly transition.
I’ll never use Tor because I have no idea what the Tor client is actually doing. Is it enabling someone to use my network connection for cybercrime without my knowledge? No thanks.
Clients are never used as relays in TOR. You never route anyone's traffic until you setup it yourself. And you can't miss that part, and it's not a default, and requires additional configuration.
Also relays (not exit nodes) are pretty safe to operate and running them is a decent thing, supporting free internet instead of a corporate ads machine, let's not frame it as a "crime support".
> Also relays (not exit nodes) are pretty safe to operate and running them is a decent thing, supporting free internet instead of a corporate ads machine, let's not frame it as a "crime support".
Well the purpose of using Tor is to prevent any network operators from knowing who you're talking to. Which AIUI is primarily a concern if either you're not allowed to talk to whoever ("great firewall" type things), or you risk getting in trouble for talking to whoever (Silk Road etc, or disfavored politics).
I guess if you're worried about hacks and doxxing rather than LE? Or if you only call things crime when they should be illegal rather than when they formally are?
Using Tor browser and running a Tor node are different things, by using the browser you are not contributing to the network, you're just accessing it. If you're worried about someone using your network connection for cybercrime you shouldn't run a Tor node (although there is significantly less risk with a relay node), but you shouldn't worry about using regular Tor.
> by using the browser you are not contributing to the networK
That's false to some extent. Tor's promise comes from it's vast population of users. The more users it has, the better it is to improve everyone's anonymity. So in a way, even by using it, you are helping Tor network. And please, save the "criminal" bs (meant for the original comment).
Why? The utility of any network grows with the number of participants, even that of inherently asymmetric networks that strictly distinguish "producers" and "consumers". (More eyeballs make the network more valuable to content providers.)
This might not be how courts determine culpability of redistributing any potentially illegal content, of course.
>This might not be how courts determine culpability of redistributing any potentially illegal content, of course.
Which is precisely the point of this discussion.
Might as well argue "By protecting the environment you're supporting the drug trade, because people that a climate catastrophe would wipe out will be able to be drug users".
This here response continues to stretch "pedantic correction" to new levels.
What's "literally outlined" I'd guess is that the utility of the Tor network increases with adoption which nobody ever doubted.
What is discard is the tenuous over-stretched argument in this thread regarding fears of legality, that went like this:
GP: Using Tor browser and running a Tor node are different things, by using the browser you are not contributing to the network, you're just accessing it.
P: That's false to some extent. Tor's promise comes from it's vast population of users. The more users it has, the better it is to improve everyone's anonymity. So in a way, even by using it, you are helping Tor network.
As others have mentioned, that's not what Tor does by default. Just because you don't know how it works doesn't mean that it's generally unknowable.
And conversely, it's enough to visit a random website running WebTorrent or just a plain HTTP DDoS attack to possibly "use your connection for cybercrime".
Since RFC 3514 unfortunately never gained traction, distinguishing good, bad, and illegal traffic remains difficult.
This is a discussion on whether or not it is better to use it to announce new domain names post-suspension than Wikipedia, not if it can sustain petabytes of data.
Given the archives use of BT, probably better to just use BEP-44 on mainline DHT if you want to store an arbitrary address somewhere - or use something like IPNS, or .onion as others have mentioned.
As nostr relies on gossip, there is no guarantee you will have access to latest address.
I am unaware of an app that sees present day use for communicating short messages that uses BEP-44 or mainline DHT. Nostr works today for millions of people. These services have normal people using them, not Hacker News commentators that are extremely sophisticated -- imagine telling someone who only knows how to use normal web services and apps they install from the app store to "just retrieve a BEP-44 message from the DHT".
I’m not aware of any laypeople who use nostr either.
But if you are going to build some nostr-based “resolver” into a browser you could instead use any of the other protocols - 2 of which are designed specifically for resolution, and the other already having a robust resolver implementation built on top of it?
I quite like nostr, but let’s not pretend it solves every problem.
Trying to understand nostr, I looked up its Wikipedia page...
> In 2024, in an article reporting on the project's funding, Business Insider claimed to have identified fiatjaf, and had found two websites previously published by this person to disseminate the work of Olavo de Carvalho, a far-right conspiracy theorist.
That... seems extremely irrelevant. If fiatjaf is contributing something useful and significant to the commons, why does it matter that he used to spread far-right conspiracy theories in the past?
> As a result of its ability to quickly and discreetly create accounts and publish posts to relays, Nostr can propagate spam much more easily if left unchecked. A notable example includes a case where multiple protocol bridges have been used to conduct spam waves on the Bluesky social network (itself connected to a competing protocol, the AT Protocol) by creating posts on Nostr, bridging the post to ActivityPub and bridging it again to Bluesky.
Surely they also had to create a Bluesky account for that? I don't see how Nostr is to blame here. Perhaps Bluesky forgot to use anti-spam measures when bridging things over from other sources? That's kind of on Bluesky, no?
This reads like a smear campaign against Nostr. I don't think I have the necessary Wikipedia karma to get it amended, but gee do I have opinions on this...
Of course, these days the people paid to do this have learned not to do edits from their own corporation or government office's announced IP blocks. But in times passed finding many of this category of edits was as simple as sorting edits on Wikipedia by the originating IP address and looking for which ones came from institutionally announced subnets.
Point being, massive amounts of capital and intelligence resources have been dedicated to censoring social media. There's nanny employees in every single social media company making sure "important" complainers are heard and their desires to silence voices fulfilled. I follow a large number of people on Nostr that have been banned from every other platform. Facebook. Twitter. Bluesky. "Free speech" sites like Gab and ActivityPub servers that advertise "free speech". But Nostr has the same entrance requirements and cryptographic sovereignty that Bitcoin provides. Generate a keypair and you can publish. People that want to find your content can simply subscribe to your public key. This results in a subversion of countless state and corporate capital expenditures. If people use Nostr, they will permanently lose the ability to moderate content in this oppressive manner. They absolutely do not want this to happen.
> Nostr can propagate spam much more easily if left unchecked.
Nowhere have I had a worse problem with spam than Twitter and Facebook. For all the alleged vulnerability of Nostr to spam, it has not currently materialized as an issue.
Note that filtering out actual spam without a centralized moderator is one of the most solved problems on the Internet. If you've ever installed Spamassassin or other well subscribed to Bayesian filters on an email server, you know that you never see spam ever again. In actuality, spam is a much bigger problem when you are dependent on fickle human moderation.
As a 'left-leaning individual' it's funny because if you look up anti-war left leaning outlets and such on Wikipedia, they don't tend to have exactly glowing entries on there. Wikipedia and the other outlets described as 'left-leaning' are neoliberal institutions. Believe me that there's no love for these on the left.
When it's convenient for smears, neoliberals are left but then at other times it's the communists etc. In other words, 'left-leaning' is a grab bag of what one doesn't like these days, rather than any really meaningful group.
> Wikipedia and the other outlets described as 'left-leaning' are neoliberal institutions.
What exactly do you think 'neoliberal' means?
I do agree Wikipedia is not 'left-leaning', mainly because 'right' and 'left' are bullshit names that don't mean anything. But it doesn't even have the power to act in a situation that would make it neoliberal.
Neoliberal as in prominent decision makers/editors etc, such as Jimmy Wales express the sort of free market and foreign policy philosophy that has been mainstream since about the 80s.
It means that entries on individuals, countries etc. are broadly in line with what you'd read in any mainstream media outlet and so is its outlook on 'Western civilization'.
That doesn't mean it's not a good project, or that it has some great power, just that its 'gatekeepers' are not exactly dissidents of any sort.
Is biased in claiming the consensus is a contentious topic, instead of only a tiny well founded minority ever supporting it. But it's the same bias you will see in any history book.
If we go extreme in another direction, this one has the same bias of representing fringe views as equally represented in a debate:
There's a strong modernist bias, with a secondary classical liberal one. What is about exactly the same bias you would see on the main literature of both subjects.
So, no, except for behaving like an encyclopedia and reflecting the literature biases, I fail to see how the wiki is neoliberal as a whole.
The defining feature of fascism is these kind of tyrannical public-private partnerships. It was the entire basis of Mussolini's fascism. They are the literal fascists.
Redifining fascism as meaning "racism and anti-semitism" (certainly attitudes which by the current definition far predate fascism) has been one of the most clever acts of sleight of hand by the regime, giving it unlimited freedom to enact the most totalitarian form of fascism ever conceived.
> aren't these the people that would want the ability to disseminate information in the face of fascism?
Everybody wants free speech — but only for opinions they agree with. And they are against censorship — unless the "right people" are censored.
Recently, the left has been far more authoritarian, labeling everything they don't like as "far right hate speech", pushing to make dissent illegal, and demanding censorship. I guess the pendulum will swing the other way eventually.
It's not really a left VS right issue, but an authoritarian one. Free speech can be uncomfortable, that is the point. "Free speech, but…" does not work.
> Recently, the left has been far more authoritarian
I'm not sure how a reasonable comparison of authoritarian behavior seemingly assigns more weight to random Wikipedia contributors lumped together as "leftists" compared to the literal government currently controlled by the right that is routinely threatening to pull FCC licenses for critical speech among other intentionally speech chilling threats.
I'd say the pendulum has already swung the other way, while swinging much, much further and more openly than nebulous mob demands for "cancel culture", over zealous Twitter moderation of hate speech or whatever else the previous go-to examples for the left were. Before 2025 showed what a truly authoritarian anti-free speech policy looks like when wielded by those with actual legal power and zero shame.
Some pages have 'semi-protection' or 'extended semi-protection', where a user must have a Wikipedia account which must have made a certain number of edits and be older than a specified time period.
However, this isn't one of them.
You can edit it immediately, either with or without creating an account.
As long as there is centralization, there is always an avenue for abuse with money. The DNS root itself is heavily influenced by a group of allied nations, through the ICANN if I'm right. That can be used to exert pressure on TLD registries, including ccTLD registries. Of course, that cannot be used for surgical control single domains like Anna's Archives'. But DNS blocking is an old technique by now. The copyright cartel needs to get it banned only in a few populous countries to destroy the value of a domain. We can keep finding workarounds. But at some point, they won't have to worry about people who can actually do that.
They could also potentially write their latest address to the Ethereum blockchain instead, assuming there is a client that could scan and resolve. This would be bulletproof.
Some sites like Anna's Archive have .onion site for the Tor network, and others do not. Is there a considerable downside (DDOS?) to providing access to their site by those means?
I am aware of POW but that hasn't stopped the DDoS. If you look at chart from few months back, there was significant hit due to crazy DDoS carried out by DrugHub which was evident in a Dread post.
This happened a couple days ago, not "a few hours ago". I was surprised torrentfreak hadn't posted about it at the time -- maybe took them a while to notice? I tried to hit the site on Jan 2nd, 10:40pm pacific timezone, and the domain was inaccessible then.
<sarcasm>
I thought that Anthropic and OpenAI created precedents that legitimize libgen and Anna archive. They pillaged all the books and scientific papers from those archives and got away with it. Why not the rest of us?
</sarcasm>
Be careful not to confuse using the material and distributing it. There are open legal cases sorting out what fair use means for generative AI. Distribution (seeding in the case of torrents) of this material isn't legal. It got Meta in trouble, and it's getting Anna's archive in trouble.
People always say this like the tech industry wasn't culturally anti-copyright and pro-creative commons before. Those same people probably work at Meta and Anthropic, just like Google's book project which got them in trouble.
> People always say this like the tech industry wasn't culturally anti-copyright and pro-creative commons before.
I completely agree with that. The problem is that the current system is such that only billion dollar players can flout the rules, while everyone else is left in the dust.
Others already mentioned they lost their lawsuit. Should the fines have killed Anthropic? Would have been more fair and a less bad world?
Why not focus energy on being anti-aggressive copyright in general. These system won't ever be fair. It's just rent seeking enabled by the government and some people can afford the rent.
You’re talking past me for no real reason, mate. That’s precisely the point I’m making.
Young Carlos thinks it matters that Anthropic got sued when they can keep flouting the rules anyway, and I disagree: it’s not a fair system until we ditch the rent-seeking entirely.
They paid one of the largest settlements in world history. Should I guess that hackers are only satisfied with the public execution of the company leadership?
To pick a nit: Technically Anthropic didn't loose any lawsuit or pay any fine. They came to an agreement with the authors to pay them a $1,5 billion settlement.
Which was a lot of money per book.
Wasn't the Google project scanning physical books and not distributing them externally? That seems like a very different thing than torrenting or even downloading stuff uploaded by a third party.
Fair point, but I think the Pinkertons would be at my door within the hour if I started re-appropriating the art style of Studio Ghibli or Disney for commercial profit.
Why the negativity? You can also as an individual do the same as Anthropic and get sued for billions. You have that option, don't let anybody hold you back!
And that's why helping torrenting and seeding the content of AA is vital: they can take down a domain name but not block everyone who seeds.
I said this before but if you've got some spare GB/TB on a computer/server, consider "donating" it for culture preservation purposes:https://annas-archive.se/torre nts
Looking at mutable torrents recently and it's ability to circumvent censorship via DHT.
I wonder if it could be revived.
The only problem is the mutable torrents standard is in draft and not adopted widely. I think I saw someone proposing using DHT in a way that allows them to host websites. If feasible becomes very difficult to take down.
That was always such a silly argument, even more so with the rise of cheap and reliable 3D printing. I download, print, and use physical objects every week and don’t know or care what IP issues might exist as long as it works for its purpose. I can’t wait until the day comes when you can download and print a working car at home.
Then I'm surprised. It's much different to the usual situation when even mass made products are either banned or very limited, e.g. unicycles prohibited, e-scooters only up to 20 km/h, e-bikes only up to 25 km/h.
I agree, it's surprising. Even in California of all places, you can make a one-off vehicle at home and register it for legal road use, without even needing modern safety or emissions equipment:
Another even more common strategy is to "restore" a classic car using some extremely small number of parts from some really old pre-emissions and pre-safety equipment car. This is often done for hod rods, dune buggies, etc. where it will be, say a "1930 Ford" but contain only some minuscule amount of that original car it is titled and registered as. There's a sizable industry of homemade "kit cars" that require you to start with a legally registered VW Beetle, but ultimately they often retain nothing except parts of the thin sheet metal floor pan, and somehow that is apparently legal.
I like IPFS but decentralized storage is just awfully expensive to maintain without any reward. Main IPFS mirrors don't tolerate AA content, it's taken down very quickly.
And also IPFS rats our to the whole world all your network interfaces, MAC addresses, and internal network configs.
And even after multiple requests of supporting a Tor mode, have routinely ignored that with "but its too hard!"
And, I quit running IPFS back in the .31 version after adding some chat logs to my local machine's share, and found a google crawler within 1 hour and fully indexed them.
IPFS support interface binding, hard disk quotas, working as a router or not, and bandwidth quotas among several settings. Yes, by default can be a bit intensive.
You are right; it's a bit of a excuse. Tons of small services under Unix (by default) listen
on several interfaces and networks. I think Bitlbee itself listened on 0.0.0.0 instead
of localhost. Ditto with some UPNP/DLNA daemons.
It takes very little to set up IPFS to just listen to tun0, disable routing (let ygg do its job) and throttle the bw a little so it doesn't hog the whole network.
I would be surprised if the Kubo/IPFS developers didn't already configure Ygg for themselves, as both are software written in Go.
The author of NNCP https://nncp.mirrors.quux.org/ (and the other author from Tofuproxy) are pretty much aware of Yggdrasil.
Funny how HN is usually ripe with folks crawling out of the woodwork to defend IP ownership as a fundamentally important principle, both on moral and economic grounds.
Yet, as of 08:37:36 MST Monday, 5 January 2026 there isn't a single comment on this thread complaining about Anna's IP theft.
Don't get me wrong, that makes me VERY happy, I am a firm believer that the very notion of IP ownership needs to die a horrible death, something that AI may very well make happen in short order, yay.
But still, I can't help but wonder why the "this is IP theft" crowd is completely silent when it comes to the like of pirate bay and Anna.
> the very notion of IP ownership needs to die a horrible death, something that AI may very well make happen in short order, yay
I agree with the disdain for IP, especially with what it had grown to become nowadays. But while I was also initially optimistic that AI companies may find a way to make IP go away as a byproduct of their activities, now it seems more like the big businesses will cut deals with one another and leave us commoners with nothing. Entertainment megacorps and AI companies rule the world, and I have no doubts that they'd find a way to become close allies. The AI companies get their near-endless stream of training data, the entertainment industry gets a cut of that juicy AI money and gives away their data willingly, while the IP remains locked away from ordinary people for eons more, just the way they like it. No one but us wants IP reform, or at least no one with real power, so it will probably never happen.
> I am a firm believer that the very notion of IP ownership needs to die a horrible death, something that AI may very well make happen in short order, yay.
The leading AI labs are not killing IP. They are taking IP and reshaping/combining it to produce their own highly lucrative proprietary IP package which they sell to you.
The mirror image of IP defenders are AI boosters who argue against IP when it comes to slurping up media but squirm when you say "ok, then publish all of the inputs that go into making your frontier models, and publish the model weights too."
AA is not stealing every byte of data they can get in order to make billions of dollars, collect personal data about people, and then sell that for even more money.
Yep, let's accept the monstrous industries which lock down culture for money.
I for one support their efforts. The same way we store seeds in vaults deep in the depths of the earth, we should do this for digital content too, and without retaliation from any specific industry.
What would be better is solve the root problem. These (illegal, somewhat legitimate) hoarding sites are most valuable for research literature which, given the public funding nature of these things should not be gated to begin with.
The comsequence of resolving the symptoms is that illegitimate use piggy back on it. Artistic literature that would legitimately deserve protection get hoarded as well.
Sweating authors of clearly copyrightable arts, typically novels, manuals, are seeing their work accessed free of royalties. For the sake of freely distributing scientific literature.
It makes it impossible to make then distinction given the legitimate utility is operating in a dark domain.
Yes, we should archive everything. And we should perhaps reform IP more broadly and re-think how we treat our culture. And we shouldn't expect retaliation.
But retaliation will happen, and I worry that it's going to pull down one of the most incredible archives along with it.
Excuse my FUD, but are they really non-profit as described in FAQ?
I find it a little hard to believe given how aggressive they are at marketing the paid version.
As someone who runs a charity, it surprises me how often people equate "non-profit" with "doesn't get paid" or "does everything for free". Non-profits still need money, they just don't use that money to make profit that is paid to shareholders. Instead, they use the money to further the stated objectives of the organisation.
NB: I'm not trying to imply anything about whether Anna's Archive is non-profit or not. Just that the fact that it aggressively tries to raise money isn't a relevant factor.
Check their gitlab for bounties. They have a lot of money in reserves to be spent on that. There are a few open bounties for hundreds of thousands of dollars. I've also worked on a bounty for them once and can confirm they have plenty more money in their budget. So I believe that they don't pocket donations, or at least not a very large portion of them.
They provide LLM data sets for 'donations'. They collect other open collections and beg for paid memberships to access pirated material. Its a russian project. You can for sure assume that someone is earning something somewhere.
I've seen people claim this a lot, but is there a single proof that supports this? The only potential insights into AA owners was an arrest of a few people suspected of running the site, and they were Latin American, not Russian.
This kind of implies a legitimacy that AA just doesn't have. They're not 'a' non-profit, and there's simply no way to know that they're not taking in loads of money and pocketing it.
There is. Without a legal framework its just an association that claims not to make a profit. This is all word games though as be clearly don't agree on the definition of a non-profit organization.
In UK there is simply no such status as "non-profit", so technically none of what you listed are "non-profit" organizations. These are unincorporated associations, that, by the way, can be making profit (but that would be on shaky grounds, since common law makes it hard not to fuck up, being as vague as it always is). However, there is a "charity" status in UK, and you'd have be a registered organization to obtain it.
Anyway, "non-profit" is (i.e. "only makes sense defined as...") a legal status, it isn't just a way to say "not making money" (after all, we wouldn't call any failing business a non-profit, right?), so it really doesn't make any sense to ask if an illegal underground gang is "non-profit". GP is correct to point that out.
When people talk about a "non profit" entity, they are almost certainly referring to a legal non-profit. Maybe that's not what GP meant.
And yes, you are technically correct in that there's no way to know that they are or are not pulling in money and pocketing it. I'm not sure how that's relevant to the questions asked but okay.
I certainly don't consider taking money from AI companies while giving data to everyone else for free makes them AI company footsoldiers, but that offer is presumably what's being alluded to.
It gives AI companies influence, if they supply a significant amount of AA's budget. It also may give AI companies knowledge of AA's people and assets that could be given to authorities.
Which - now that the various commercial LLM companies have harvested the archive, they will be more than happy to use litigation to raise the drawbridge behind them, so as to block any future competitors from acquiring the same training data...
Is this an abuse of the ServerHold status? Was this the same mechanism used to delist a Gaza video archive recently?
> This status code is set by your domain's Registry Operator. Your domain is not activated in the DNS.
> If you provided delegation information (name servers), this status may indicate an issue with your domain that needs resolution. If so, you should contact your registrar to request more information. If your domain does not have any issues, but you need it to resolve in the DNS, you must first contact your registrar in order to provide the necessary delegation information.
https://www.icann.org/resources/pages/epp-status-codes-2014-...
To add a little context, this suspension comes immediately after Anna's Archive publicly implicated themselves in the Spotify scraping "hack" in which they downloaded nearly the entire content library of Spotify and was preparing to release it publicly (~300TB worth) via torrent.
They published a blog post outlining their plans.
Did the operators _want_ to poke the well connected & well funded bear with a historical anger problem?
No, but they weren't not going to, given that their mission is to archive all cultural content, by hook or by crook.
Archiving it and publishing it are different things.
More importantly, they may sabotage their mission: If Spotify shuts them down, their exiting archives and especially future archives may be effectively lost.
I guess I should say more accurately: Their mission is to both archive it and publish it. They seem to be explicitly against copyright, on principle. Which I greatly respect.
It's time to abolish copyright. It creates more problems (stiffles innovation, creates rents) than it solves (rewards innovation).
It doesn't create problems for large companies that make AI systems.
Yeah, it seems to only be a problem when you're a human being remixing the culture you grew up with.
Meta can admit to soullessly scraping books they don't own for their for-profit AI datasets [1], and it's not a problem because they're Meta. But if you're an artist? Nope. Sampling in hip hop songs, for example, is in a "complex legal gray area" (translation: "it's illegal but we don't want to admit that out loud") [2].
[1] https://futurism.com/the-byte/facebook-trained-ai-pirated-bo...
[2] https://urbanspook.com/copyright-laws-2025-impact-on-hip-hop...
Fortunately, Spotify does not have that power. Annas Archive is not based in US or EU jurisdictions. They can make access for normal people a bit harder, but not shut it down.
(Edited for clarity)
> Fortunately, Spotify does not have that power. They are not based in US or EU jurisdictions.
Perhaps I misunderstood something, but according to my understanding
1. Spotify is registered in Luxembourg and has its operational headquarter in Sweden (Stockholm). Both are EU countries.
2. I guess it won't be Spotify that sues, but the individual music labels (very likely united).
Annas archive is not based in the EU (sorry for being not clear). So the law in EU is limited to enforce a ban. In germany it is already "banned" via ISP but just DNS.
But the real servers are hosted in kazachstan or russia I think. And they do not cooperate so much with EU courts.
So unless the EU installs a great firewall like china, they cannot really shut it down.
> But the real servers are hosted in kazachstan or russia I think. And they do not cooperate so much with EU courts.
I believe the "official" AA servers only host the website + source code. The actual copyrighted content is stored by volunteers who seed the torrents.
Exactly, this is why the 'Hydra' is difficult to take it down.
Presumably the opposing party is residing in non-US-or(and? depends on the order of evaluation)-EU territory, but I might be mistaken. "They" refers to both sides in the parent comment.
I'm not sure archiving and publishing are different things.
They are, but archiving without publishing is pointless.
I occasionally wonder how many enormous collections of culture like that of Marion Stokes[1] have been lost because their curators made no effort to realize the value of their collection.
1. https://en.wikipedia.org/wiki/Marion_Stokes
Most archives - the ones in libraries, etc. - are not published, except they are available to qualified people who physically travel there. Most are not even fully indexed - nobody knows all of what's there.
My perspective is compatible with this fact. An archive that approximately nobody can access and/or nobody knows what it contains has no value to society at large, except the potential that it may some day be published.
The good news is I'd guess the number of (nonreligious/nonproprietary) institutionally managed pointless archives is dwindling.
> They are, but archiving without publishing is pointless.
One may collect/archive now (when the data is, well, "available"), and publish later, when copyright expires and the material will likely be harder to obtain.
Both are illegal, if you just hoard you will never know if what you have is useful. Only way to judge that is by letting people use it.
I can save a copy of my friend's book on my computer, archiving it. Nobody else could see it unless I publish it.
They stated that they would pass the information on to other archivists and public/private trackers no? They obviously have backups, since there are multiple users seeding Gbs and even TBs of data. Mirrors can be created as well, like TPB.
No, because they are all backed up on torrent. Good luck, getting those "shut down" from the DHT
They didn’t come anywhere close to the entire content library, the 300TB represents about 33% of Spotify, though it is close to 100% of the played music.
Kind of nuts that 66% of their library is virtually unplayed. It’s hard to make it as a musician.
It is ridiculously easy to create an album with Suno and push it Spotify. I'm surprised its only 66% TBH
Anna's archive has a great analysis of the Spotify data.
They identify a huge surge in tracks that few listen to after gen AI started.
The analysis is worth reading. The distribution is (Pareto)^3 ~99% of the tracks played are 1% of the catalogue.
1. Generate slop music nobody will ever listen to 2. ???? 3. Profit
It's actually:
1. Generate slop music no _human_ will ever listen to
2. Use a botnet to "play" this music en masse
3. Profit
This is a whole arms race, with companies (such as Beatdapp) specializing in detecting fraudulent plays.
Source: I work for a niche music retailer that struggles with the same issues on a smaller scale.
From a stat I saw years ago, about the same amount of apps on the iOS app store have never been downloaded.
To be completely fair, I am not certain what it means for a track to be "virtually unplayed".
First off, it was striking to me how little of the "top 10 000" they published back on Christmas I recognize. I'm not sure what I expected, but 10 000 sounds like a big number, so it seemed likely to me, that if I get a random song from my playlist I could find it there. It turned out I hardly can find an artist I recognize. Ok, I can recall a song from Lady Gaga and even Billie Eilish, I've heard of Bruno Mars (cannot recall any song), but I have no idea what is "Bad Bunny", "Doechii", "Drake". I mean, I think I do have a pretty good idea what these things are (abstractly), and I probably wouldn't want to listen that. And while I knew that all this stuff is very popular, I didn't quite realize how little place in the top-10000 it leaves for the music I (and everyone I know) actually listen to.
I didn't download the metadata they released (it would be hard to process it on my laptop anyway), but now I wonder how much of my 3 TB music collection is in top 100 000, or heck, even top 1M Spotify, or on Spotify at all.
I also am sometimes surprised how little scrobbles some tracks get. I didn't bother to find out what this means, how many people still scrobble to Last.fm or ListenBrainz, but it is just surprising when I see that a track that I didn't consider to be obscure was scrobbled like 50 times this year.
So I'm saying that music worlds seems to be terribly fragmented, even more than I imagined. So the very premise of AA backing-up 97% of Spotify (by the number of plays) may be much lesser achievement at "preserving culture" than it may sound. And of course we are about 8 years too late to backup everything, since by now half of it must be generative NN bullshit. And I'm not even sure it's in those leftover 3% (bots listen to bot-generated music too, right)?
> It turned out I hardly can find an artist I recognize
I've heard of 9 of the top 10 and 15 of the top 20 at https://chartmasters.org/most-monthly-listeners-on-spotify/
You might not listen, but surely you have heard of Taylor Swift, Justin Bieber, Ariana Grande, Ed Sheeran, Coldplay and of course Christmas Staples of Mariah Carey and Wham?
First off, this is not the top we are talking about, since there is one that AA provided[0]. I am not sure what it matters which names exactly I've heard of, but if you are that curious: I don't know what is Ed Sheeran and Wham (but cannot vouch I've never heard their music in a supermarket), but I definitely remember "Coldplay" being mentioned in a joke onstage by a NIN member[1], but I didn't bother to check out what they are. I can imagine the faces of Taylor Swift & Justin Bieber, but cannot name any song, and I'm sure I've heard Mariah Carey somewhere, since that name is around longer than Rihanna. I have a song or two of Ariana Grande in my playlist though.
Edit: Ok, I've finally googled "Coldplay". Yeah, definitely heard "Clocks" somewhere.
[0] https://annas-archive.li/blog/spotify/spotify-top-10k-songs-...
[1] https://www.youtube.com/watch?v=qboe5CebixA
You're a (waaay) outlier.
Are you sure? See, my point is a conjecture (based on a reasonable assumption that I cannot be that special), that there must be really a lot of us "outliers" out there (so I'm not even sure it's reasonable to call us that).
Let's reiterate. I am well aware that more people listen to that Bad Rabbit, Taylor Swift or Justin Bieber than they listen to <random name from my playilist>, it's not really a surprise. There even is a special name for people like that, it's "celebrity". In fact, that's probably how most people who are into music (including myself, I might say) would categorize them, as "celebrities", not as "musicians" (though, mind you, of course they are musicians, as everyone who ever sang a song is, it's just that when I hear the word "musician" I don't necessarily think of Taylor Swift). Hence these people indulge themselves for not knowing who these guys are, explaining it that "they are not into celebrities".
And it's no surprise that a lot of people listen to celebrities. I mean, if Trump would release a song right now, it would become #1 on Spotify in no time (for a very short time, but still). Well, maybe not #1, but close.
But I also suppose there are a lot of people who are into music. Maybe not so many, as there are people who are into celebrities, but it's still a lot. And after seeing that top-10 000 I suddenly find it very plausible, that a lot of tracks these people call "massive hits" may turn out to be "virtually unplayed". And hence not in those "97% of Spotify (by # of plays)" that AA archived. I am not even claiming it, I'm just saying that this doesn't seem to be impossible.
For instance, any DnB fan would say that "everyone knows Noisia and Black Sun Empire". It would be absolutely laughable attempt at "preserving human culture" not to include them. Surely all of their tracks must be at least in top-5M, right? Well, after seeing top 10K I'm not so sure anymore.
Maybe you've never heard of them, but surely you've heard of Prodigy. Not a single track from Prodigy on top-10K. Or Chemical Brothers. Or Burial, or Placebo, or Nighwish, or King Crimson. These are very famous names in respective circles. There are 2 tracks from Massive Attack — both featured in super-famous movies and trending on TikTok right now. For God's sake, there are only 8 tracks from Madonna in top 10K. Versus 26 from Imagine Dragons and 124 from "Bad Bunny", whatever it is. How do you like Madonna for an obscure artist?
So, my point is that there may be a lot of people listening almost exclusively to "virtually unplayed" music. Entire discographies of (niche) cult-artists may turn out to be buried in these 66% of "virtually unplayed" tracks.
I guess I should just get the metadata and check, but I'm pretty sure that would be outside of capabilities of the hardware I have on hand, so I'm not sure how to go about that.
The metadata torrent is only ~200GB, which should be well within your capabilities.
https://annas-archive.li/torrents/spotify
Anyway, I think you should keep in mind 2 things:
1) 10,000 tracks really is not a lot. It sounds like a lot, but isn't. My own - relatively small - collection is nearly double that.
2) 10,000 tracks... out of 256,000,000 that AA archived.
I'd be very interested to see some more analysis done on this, particularly as it relates to, say, Last.fm statistics - but I suspect the missing music is not as significant as you think.
In any case, even if every one of those "niche" artists you list are missing from this collection, I don't think it's fair to say it's a "laughable attempt" - it's certainly better than nothing, even if it's not perfect.
The funny thing is, since the advent of streaming I no longer listen to the radio. I listen to new music, but little pop music, and I have never heard a single track from Swift, Bieber, Grande or Sheeran. Coldplay is the only act I like on that list, and the streaming services are pretty good at only playing what I like.
If they were pre-streaming artists I probably would have heard a lot of their catalog because radio played it over and over. Unfortunately you just can’t get away from the Christmas music.
Sure, but I'm sure you've heard of Taylor Swift and Justin Bieber.
Traditional radio mostly sucks, but Soma.fm and KEXP are both great for discovering new music.
Very hard if you have little talent..
https://annas-archive.li/blog/backing-up-spotify.html
> For now this is a torrents-only archive aimed at preservation, but if there is enough interest, we could add downloading of individual files to Anna’s Archive. Please let us know if you’d like this.
If it is torrents only, what relevance does unregistering the domain make?
Establishing a position Anna's opponents may consider an advantage.
And there is a site idea!
Annasopponents.news --> Can inform passersby on anything related to Anna's Archive along with activism related material, how to's and the like.
Ideally, if AA doesn't have any public web presence it's a lot harder to publicly disseminate those torrents.
Realistically, it's just a way for someone to say something is being done about this, even if it's not going to actually make a difference.
Yeah, obviously I don't know if it is actually related, but my first thought when I couldn't open it today was "Told you so"...
Spotify was created from a library of pirated music.. the irony
Came here to say that.
An while back, another site started with a pile of pirated music, and that was allofmp3.com Remember those peeps?
Their business model was to sell music by selling bandwidth. Basically is was all the music you want charged by the megabit download.
Pop titles were $0.10 to $0.25. A whole album at 256mbps was roughly $3 give or take.
What got me really thinking was how great the UX experience was. At the time, few came close.
The end of that site was packaged up with Russia's entry into the WTO.
I seem to remember hearing about huge torrents out there too. The right infohash can point a person to huge archives of various kinds, books, video, academic papers, music, the WikiLeak insurance files, which is password protected, as perhaps all of these are.
As someone who grew up poor in an ex-Eastern Bloc country, allofmp3.com was a godsend.
Used to work for a registry. Registries do not just arbitrarily serverHold things they don't like. Even in blatant abuse cases, they'll reach out to the registrar to clientHold it or delete it instead.
I suspect this came from a court order. The only time I remember serverHolding things were from court orders or other legal requests(FBI/DHS/etc). Though the latter would often just ask for the nameservers to be changed instead.
Not my experience. My personal .us domain was placed in serverHold in April 2025, after Godaddy Registry sent my registrar an email including the following:
"If we do not receive a response and the abusive content remains active, the referred domain will be suspended (-ServerHold-) twelve (12) hours after this notification, to limit any potential damage."
Yes, only 12 hours' notice. And this was in response to what looked like an automated scan, with a screenshot of a directory listing showing some files with (accurate) modification times in 2011. This would not have passed more than a quite cursory human review. And there is no chance that anyone was using the files, as they were at an unused path (which happened to be exposed via directory listings).
(How do I know that the files haven't been infected without my knowledge? Because I checked, but also because I know why they were flagged as malicious. The files contain an exploit, and will try to exploit your browser if you open them. But the exploit only works on 2011-era iOS! The iOS version numbers are right there in the filenames! The files are not real malware, but part of a jailbreak, JailbreakMe, that installs software of your choice on your phone… if your phone is an iPhone from 2011. I continue to host the jailbreak on purpose, for the sake of a handful of people that still use these devices. I guess you can call it retrocomputing. This particular subset of files was unused though.)
.US is a bit of an oddball, at least compared to gtlds. The US DoC puts rather strict requirements on the registry under contract, including this but also spot checks that owners are US citizens, and makes it the responsibility of the registry to take action on it.
Those scans are automated but supposed to have human-in-the-loop, though that doesn't really mean much either. Unfortunately, once the process starts, receiving no response means an automatic suspension.
literally 1984
I remember itch.io was put on serverHold by having someone claim there was "fraud and phishing":
https://news.ycombinator.com/item?id=42363727
But .org is managed differently from .io, so maybe a more formal legal action was taken.
Full writeup of the Gaza video archive takedown, if you aren't up to date on it: https://neosmart.net/blog/namecheap-com-revokes-domain-hosti...
In that case, it was the registrar (not the TLD owner) that put the domain under clientHold and clientTransferProhibited, etc (which disables DNS lookups).
The Genocide.live site is now back up (just yesterday) after they raised a fuss on social media, and were able to get the domain unlocked to transfer it to Trustname (out of Estonia) as their new registrar.
(The Namecheap founder/CEO Richard Kirkendall surprisingly came across as surprisingly unaware of how anything to do with domain name registries, TLDs, and DNS works on Twitter in an exchange where he thought the entity running the .live gTLD was the archive's new registrar [1], claimed the domain was unlocked for transfer when it was still on clientHold [2], and a raft of other silly mistakes.)
[0]: https://neosmart.net/blog/namecheap-com-revokes-domain-hosti...
[1]: https://x.com/namecheapceo123/status/2007139737379934559?s=2...
[2]: https://x.com/namecheapceo123/status/2007228146060259365?s=2... yet https://x.com/receipts_lol/status/2007984691476156443?s=20
I'm pretty sure serverHold being applied to domains targeted by court orders is normal. It also has all of the xProhibited statuses which ICANN notes are "usually enacted during legal disputes".
Yes. Genocide.live is under clientHold status by namecheap and they are trying to transfer it out.
I don't know much about the story or that site, but I can just open it from my end? Whois information also doesn't seem to show any special status.
After publicly shaming NameCheap on Twitter, NameCheap backed down a little. They've still de-platformed genocide.live, but released the DNS name so the owner could transfer it out. IIRC, they are using a "free speech" provider in Estonia now.
Hey, that's the same Namecheap that banned Russian citizens for being Russian in 2022? Nice to hear they're still principled.
The genocide will not be televised!
(A judge in the Netherlands settled the question if there is a genocide going on and since then we've been on Israels naughty list lol)
They transfered to a new registrar in 2024. Problem solved for now
https://www.whois.com/whois/genocide.live
(btw, that is an excellent website I didn't know about)
The fascist website Stormfront lost their .org in 2017 with some very unclear procedural action, but I don't think it was a ServerHold. It was offline for quite awhile and I think genuinely hurt the site. Although it looks like it's back now?
I think this was genuinely due to the Norwegian Massacre. The guy who did it had an account on stormfront. The whole thing exploded and the owner had to explain that he was not in contact with the mass murderer.
https://en.wikipedia.org/wiki/2011_Norway_attacks
The Gaza video archive is https://genocide.live, and it was Namecheap that took it down.
https://x.com/receipts_lol/status/2006732606164152651
It looks like it has been restored, and responsibility attributed to a moderator acting hastily due to the extreme nature of the content: https://sunur.substack.com/p/namecheap-stole-the-domain-of-g...
I don't like the idea of registrars doing moderation at all.
Or Cloudflare, for that matter…
That's valid, and I think a point that Cloudflare itself has acknowledged. Cloudflare is, however not quite as fundamental as domain names.
> liberal progressive hypocrisy
zionist mccarthyism isn't progressive
Thanks for linking this article! However, the full description of the incident seems that Namecheap blocked the domain, the CEO got into tweet arguments justifying it (and taking inconsistent positions), and after the backlash said it was a hasty block by the mod team, without an apology.
I generally give NC the benefit of the doubt (because of their principled stance on Ukraine, which to me means they can be principled elsewhere too), but in this case it's very difficult to explain it away.
> Was this the same mechanism used to delist a Gaza video archive recently?
The mechanism used to shut down a website whose videos were fake and that was funded by actual terrorists? I don’t think Anna’s Archive is in the same category, no.
> We recommend checking our Wikipedia page for the latest domains.
I wonder how wikipedia feels being used as DNS?
EDIT: Apparently this is a well known practice. Some interesting discussion here: https://news.ycombinator.com/item?id=40008383
Anna Archive is a notable project. Wikipedia displays links to projects. It's not about "being used as DNS" but about providing basic info about the topic, URL being an important part of it.
Kiwifarms is also a notable web site but they do not show a link to it. This means there is editorialisation going on.
That's a common problem of English, German, and a few more Wikipedia editor gangs seeing themselves as the moral guardians of the internet.
If you switch to some other languages, you'll find the links.
Having a sense of morality is not a problem. Good for them.
reasonably explained by the chasm of difference in intent and impact between anna's archive and kiwifarms
I'd like to see the intersection of people that think AA is some last bastion of Internet freedom, while simultaneously trying to get all of KF's routes blackholed.
The problem with playing internet police is there is always inherent bias.
What a crazy worldview: to think that good things are good and bad things are bad.
Who is the official judge and jury of the Internet?
In this specific case of what’s allowed on wikipedia: the Wikimedia Foundation
But this mechanism is used to circumvent DNS blockade. Wikipedia may be next to moderate if they can force DNS providers and even the org registrar to give in, wikipedia could fold too.
Then pastebin, never ending cat and mouse game.
DNS is another layer. The URLs shown on Wikipedia will still have to be resolved to IP addresses, which is where DNS comes in. Referring to Wikipedia for the URLs/domains does nothing to circumvent DNS blockades.
Maybe with the upcoming IP address certificates, we can use IP addresses again. I remember using IP addresses to access some sites in the past.
https://news.ycombinator.com/item?id=44379034
Using a wildcard DNS service, you can already do this.
https://sslip.io/ for instance.
Evenn though its onelayer down - the same tactics that were used to suspend/takeover domains would still apply , at the end of the day one still has to get the IPv4/IPv6 address from someone(who can be coerced).
When Trump pressures RIPE NCC or APNIC to deregister an IP address block, that's the end of the internet as we know it, and the return to national networks with very limited interconnection. Even Russia still has address registrations despite being sanctioned.
Alternatively they pressure USA ISPs to block the addresses. That's already regularly done but it probably won't be enough to satisfy the extortion industrial complex which is out for blood.
> When Trump pressures RIPE NCC or APNIC to deregister an IP address block
sed "/Trump/US-Govt/g"
Why do people here always casually single out Trump? He's not an outlier, it's just how US foreign policy has worked for centuries.
"Not an outlier"
A quick look at the last few administrations is all anyone needs to see how this one interprets the powers and duties that come with the office.
One of my favorite phrases coined during the last Trump administration was something like, "not just wrong, but wrong beyond normal parameters." It basically meant exactly what we are discussing here; namely, being an outlier of some sort.
I specifically mentioned foreign policy. There, I don't remember a single US government that was not a net negative for the rest of the world (Israel excluded).
It circumvents the purpose of the DNS block which is meant to prevent people from easily finding the site. Anna's Archive can easily register new domain names and put them on Wikipedia, thus allowing people to easily discover the new location of the site.
Of course many sites can serve as "DNS" - Reddit, Github, X, basically anywhere you can put a URL. So DNS blocking is relatively useless.
Why should wikipedia fold? Can any country put any legitimate pressure against them for just listing a domain.
Linking to illegal services can be illegal, that’s why.
> Linking to illegal services can be illegal, that’s why.
What is illegal in one country can be illegal everywhere.
I don't remember Wikipedia removing LGBTIAQ++ articles just because that's illegal in Iran.
If a government thinks Wikipedia is illegal in their country, they can force local ISP providers to block it, but it's not Wikipedia's responsibility [1] to censor itself.
[1] at least should not be
The Wikimedia Foundation is a US corporation. There are national chapters in some other countries, which are corporations in the respective country. The internet isn't the Wild West; websites are subject to the laws of the countries they operate in.
They can also just create a new foundation elsewhere if the US becomes hostile.
What country does Anna’s archive operate in?
No it's not, at least in US. Google fought and won this case.
Countries can pressure them for many reasons, fairly or not. Under pressure, Wikipedia founder Jimmy Wales interfered with a page about the war in Gaza (though I don't know th outcome of that).
Why should registrar/ISP fold?
It's not that they should, they often do though.
https://meta.wikimedia.org/wiki/Wikimedia_DNS
Can’t imagine they care too much given they themselves also run public dns servers.
I still remember the hiatus around the pirate bay . org going down back in the day. They updated the landing page on the alternative domains to include a hydra above the pirate ship where above each hydra head there was a domain name[1]. I thought that was a great comeback by the maintainers.
[1]: https://www.businessinsider.com/the-pirate-bay-has-a-new-log...
Those guys should get a prize for resilience alone, they put many government run services to shame. The only thing all this legal stuff seems to have done is to harden them even further.
Such great moves!
On their way down, the original creators made sure The Pirate Bay would continue to be that gift that keeps on giving.
If anyone hasn’t heard it, the podcast Darknet Diaries episode 92 (2021-05-11) is a great listen.
Do not go there without ublock! The ads on there, especially porn ads, have gotten out of control. Who is it that's making money off of thepiratebay these days?
I would urge people to not go there at all. Their moderation is imo pretty lacking, and malware sometimes slips by. There are much safer pirate websites out there (although you should still only visit them with uBlock Origin installed)
Well! Not really a surprise, is it? We knew that DNS censorship was coming sooner or later. The real surprise is that it lasted so long. X509 (TLS) PKI is also probably being abused right now. We know how differently the administration sees services like these that are actually beneficial to humanity. Choosing to rely on services like DNS and PKI exclusively for the function they provide is a very bad idea going forward for us normies.
We should have considered these centralized and corporate driven core infrastructure components as interim measures while more independent alternatives were being developed. We have a few different alternatives right now. Can't we just choose one and switch over? (something not based on blockchains.) Something like GNU Name System, may be?
PS: They will probably block the IP if Server hold/DNS block is not useful anymore. That's a different problem though.
Tor and monero will save the world.
Tor yes. I even recall seeing a rare use of blockchain for domain names (Namecoin) over a decade ago. But I don't see how Monero helps?
Financial freedom and private transactions. It's also mined on consumer hardware making it far more decentralized than bitcoin.
I wonder if in the interim period between DNS blocks and full IP bans, some websites will start distributing their IP addresses directly as the "official" way to connect to them.
The question is, how will you obtain that IP address from them? DNS was the answer to that question. So when you ask that question again, you're effectively asking for a DNS alternative.
I mean, all of this is a hypothetical, but wouldn't you find the IP in the same way how you'd find the domain now? If there's means among communities to disseminate new TLDs as these websites are forced to move back and forth, then there's nothing stopping them from sharing IPs instead.
Just add the censored IP's to your /etc/hosts file, and don't care.
I always wonder why sites like Anna's Archive (and a lot of torrent tracker sites) don't provide .onion addresses. I imagine a significant proportion of their traffic comes from the Tor network, and onion addresses don't have this same weakness that regular DNS addresses have as they're just the key. I can't imagine if they have the servers already up and running, running an .onion address is that much more work considering the resilience it would add and it's existence may even encourage users to access through Tor if they promote it as the primary address.
Z-Library does, currently it's http://bookszlibb74ugqojhzhg2a63w5i2atv5bqarulgczawnbmsb6s6q...
(as found on https://library-access.sk/#useful_link_tab)
Or IPFS/IPNS
https://docs.ipfs.tech/concepts/ipns/
Onion/Tor has horrendous latency and throughput. Latency is bearable, but dial-up speeds for downloading can drive one insane.
But you can share torrent links via onion hidden service easily.
It has been a few years since I last used Tor, but even back then you could download at megabit speeds over the network. I would assume that the situation has improved, I’m surprised that you seem to report the opposite.
The speeds are usually bad because exit nodes are more rare. But Onion addresses don't go through an exit, and stay within the Tor network.
This wasn't even true 10 years ago...
If they're going to have .onion and .p2p addresses, they should also add Yggdrasil. Yggdrasil is easier to have always running in the background so I don't have to toggle it on just to access AA.
EDIT: Never mind, I forgot how the Tor software works. I still think also having Yggdrasil would be nice, though.
How do you access Yggdrasil? .onion addresses require downloading a web browser.
I access .onion addresses from the same browser I'm typing this comment in. You need a Tor proxy, but other than that it's regular http traffic.
Oh I forgot that running Tor doesn't automatically route traffic through it. My mistake, sorry.
Yggdrasil provides an IPv6 address in a formerly reserved and therefore unused block, so you just run Yggdrasil and visit the IPv6 address.
How do you run Yggdrasil? Explain it like I'm an end-user who has no idea what IP is, much less IPv6.
My point is, how many users would have access to something on Yggdrasil? How many could - what skill is required?
Fair, it's not end-user-ready. But they should support it; it's a cool platform and it doesn't take much work to set up on their end.
Seems like great publicity for Anna's Archive. I've heard an increasing amount about Anna's Archive over the last 12 months. It has popped up a lot. I wonder if they've seen their traffic spike a lot.
I also discovered it via one of the blocking controversies. The attacked is weaponized for discovery. karma at its best.
see streisand effect: https://en.wikipedia.org/wiki/Streisand_effect
Agreed. I'd never heard of it. Now I'll be checking it out with a view to supporting it.
Definitely. I never knew about it. Now it's bookmarked.
https://annas-archive.li/ works and there's a few alternatives listed at the bottom of the page
Doesn't work where I live, is the block on a country level?
Try using the DNS4EU resolver, see https://www.joindns4.eu/
I guess so, could also your provider. Have you tried accessing with a vpn?
Blocked here too and all alternatives.
I recommend Anna's Archive get a Nostr account. Once they finally have a solid court order to seize domains, generally the rate at which they get seized accelerates greatly. Nostr is the only decentralized manner (no, Mastodon/fediverse is dependent on domain names, which are getting seized by courts in relation to this -- it is not decentralized at all when it comes down to it) that people can reliably use to have a latest content feed distributed.
"Check Wikipedia to evade the court order" just encourages legal action against Wikipedia. Even linking to copyright violations is, under current court precedent, able to bring civil liability upon third parties. It is draconian and our framers would have considered it a clear First Amendment violation, but unfortunately the current jurisprudence says that is the law.
Tor and I2P is a better fit. Nostr is very weird. It sells itself as decentralized, but basically all frontends provide same several relays.
When those relays get subpoenas and remove your resource, you're done. You can use some unknown relays to publish, but who's gonna use them as clients outside of the defaults? It's effectively designed for shadowbans.
Many clients automatically seek, or prompt an action in 1 click to retrieve content from additional relays that a Nostr pubkey announces if said content is referenced but not available on already subscribed relays. As a publisher, you announce what relays you are currently publishing to in your identity metadata. So even if you don't specifically subscribe to a smaller relay, you can still access the content on it.
Tor and I2P are great technologies. ZLibrary, for example, runs an excellent Tor hidden service and it is usually the most reliable way to access news from the site. However, this did not remain true for a while when two of the operators were arrested. Tor and I2P require you to have infrastructure online. The point of "check Wikipedia for news" is that you can have something persist even if you do not have your servers online. Nostr is the best technology available in this category.
Where does the nostr pubkey announce it? Let's say you are banned from main default relays. Is there any side channel?
You announce all the relays you publish to to the relays you are publishing to. If someone quotes your post in a post of their own, in many clients subsequent readers will be able to retrieve your content from the relays which they don't currently follow. Relay discovery mechanisms grow progressively better. I don't know of any pubkey ever banned from major relays. The operator of Damus, one of the more successful clients in history and one of the main default relays, openly engages with dissident personalities in a welcoming manner. You could probably get a "filtering transparency report" from him and others and ask if they've blacklisted any specific pubkeys and why. I am unaware of any that are currently blacklisted, and generally the network seems to defer to WoT mechanisms by clients to blacklist content.
Regardless, this remains a far more resilient persistent source of information that you don't operate than "check Wikipedia".
I still don't get how to bypass a theoretical block, you need to access at least one of the current major relays or a side channel to find a follower, only аfter that you can re-translate the new set of relays. The autosub is good, but IMO the current major operators have an Elusive Joe situation because nostr is very small. Things will change as soon as people with money and government connections see it as a problem.
I totally agree with the Wikipedia argument though.
I think relay discovery by clients has improved by leaps and bounds and will continue to do so. Previously you had to give some complicated "follow me at such and such relay" thing. These days you can just tell people your npub and things seem to "just work". I don't know what's going on under the hood, I can only say that clients figuring out which relays have content seems to have improved drastically.
Check NIP-65[1], this is the key to the latest Nostr client's automatic discovery feature.
[1] https://nips.nostr.com/65
To improve user experience, Nostr clients typically pre-load several large relays. In fact, Nostr also supports using NIP-19[1] pointers to pass custom relay hints to the client, similar to the tracker in BitTorrent and Magnet. Furthermore, I believe that with Let's Encrypt now offering free and widespread IP certificates, domain dependency issues will be further alleviated on Nostr.
[1]: https://nips.nostr.com/19
Tor, sure. i2p requires some proxy config in your browser and you need to run a service in the background explicitly. I wish they'd release a dedicated client like Tor does.
Both require daemons that offer SOCKS (other other protocols) proxies for browsers to route through.
If you're talking about the Tor Browser bundle, I2P has this: https://geti2p.net/en/download/easyinstall
Pretty sure the whole concept of i2p is that every user also contributes to the network, which is why the config
I'm not sure why that would require more friction or why my comment was downvoted: i2p is harder to use than Tor. Adoption will be hindered by it, even in tech circles. I want to fire up an executable and maybe click through one prompt but not have to remember all the configuration steps the next time I want to use i2p.
They work on that, https://geti2p.net/el/blog/post/2025/10/16/new-i2p-routers
The current focus are new rust and go implementations, but embedding lib for applications is also in the roadmap.
I also agree on hindrance. I don't understand why they don't provide a simple docker-compose at least for daemon deployment for immutable management with controlled scopes. There is an image in the dockerhub, but no proper instructions. People have to spend several hours to ensure that everything works correctly.
There's always Freenet...
Freenet is promising but it's not even fully done. Experimental stage. I am also skeptic of their more usage of LLM generated code these days.
Freenet (aka Hyphanet) is 25 years old... it's been "stable" for decades. Biggest problem is that content is sort-of hosted locally and considering how much child porn and other criminal content is in its darker corners tha can be risky.
It's also incredibly slow.
Stable and yet never used in live env. It's being rewritten and not at all stable. I don't expect it to become the standard anyway.
Better have a .onion. It's almost impossible to seize and you control the keys. YOU ARE THE OWNER, not some for-profit registrar. Onions should be the default, it's secure (you own the keys), decentralized and far better than relying on CAs for encryption.
Tor and I2P are great technologies. ZLibrary, for example, runs an excellent Tor hidden service and it is usually the most reliable way to access news from the site. However, this did not remain true for a while when two of the operators were arrested. Tor and I2P require you to have infrastructure online. The point of "check Wikipedia for news" is that you can have something persist even if you do not have your servers online. Nostr is the best technology available in this category.
> It's almost impossible to seize and you control the keys. YOU ARE THE OWNER
This also remains true for Nostr.
But furthermore, as an operator of several Tor hidden services corresponding to public web services. I can assure you that many users, especially those on mobile devices, will stop using your service in large numbers if you direct them to a hidden service. iPhones don't allow background processes without special dispensation from Apple so the Tor/I2P circuit dies every time someone switches between apps. It's also an extreme development challenge, as they don't allow subprocesses either, and then of course your app will have to abide by the GPL at least for I2P (nonstarter for some). "Just ruin your experience for all iOS users and switch to the GPL for all your client code" is not a realistic suggestion. Not that Annas-Archive has a their own client app.
If they can't figure it out anything else, I think Tor is the most plausible tech to be used. What are the alternatives if these other services don't provide enough traffic to sustent the download speed of the files? Something old like USENET certainly can't be used anymore.
I hope they follow the same pathway of The Pirate Bay or Rutracker.
Operational excellence is of course dependent on the operator but I would still think it's far easier to bring up onion as it's disposable and works behind NAT'ed VMs which makes it further easy to run.
I don't know anything about Nostr since it does not focus on anonymity and isn't as old as Tor (more than 2 decades of research and application), I wouldn't rely on Nostr for anything serious.
Laundering the code through OpenAI to see if the GPL sticks through training, would make for an interesting court case if you asked ChatGPT to write an I2P clients "from scratch" for a closed source iOS client.
Namecoin has had all the DNS issues fixed since 2011. It is one of the few real and useful applications of bitcoin based technology.
Somehow it never got too the attention it deserved.
It was also the first known "altcoin"
Namecoin sounds very promising, first time hearing about it.
How are we supposed to discover official .onion adresses though ? Is there some kind of DNS for that ?
The same way you discover official clearnet domain names, i.e. by using a search engine, getting url from a friend, etc.
>It's almost impossible to seize and you control the keys. YOU ARE THE OWNER, not some for-profit registrar.
You may own the keys but the non-profit The Tor Project owns the network. And when they decide to shut it down your "ownership" of the domain keys doesn't matter in the slightest. You might think this is a silly scenario but actually it happened in 2021/2022 when the tor project unilaterally decided to kill the entire Torv2 network and all domains were made inoperable. All links between sites, everything that made .onion a web, was lost.
The Tor Project does this whenever they feel that there's a security issue. It will happen again.
As someone that spent 10 years building completely legal community sites on the .onion network with the delusion of ownship it really hurt me. I'm never using .onion again. It is not a place to try to build communities. It is only for people that need 'security' as a highest priority and don't care if everything gets wiped out.
They don't own the network. The people who run the relays do. v2 wasn't shutdown in an instant. It was necessary and you could have just redirected your users to v3 and tell them to use it instead but you had to whine about your short-commings on Tor?
It's not only for high-security. It's for the state-of-the-art anonymity.
All the links between .onion sites broke when the relay and other infrastructure operators started running the broken (no Torv2 support) releases the Tor Project put out. All the writings of sites about each other. Everything that made it a web.
It doesn't matter that it was technical possible to try to manually reach out to random visitors of my sites and try to tell them that the entire domain name was changing. That didn't fix the web or links aspect at all.
They didn't provide any migration path from v2 to v3?
They did not. And many apps (Ricochet Messenger comes to mind) were not visited by a web browser. So it isn't like you could announce an HTTP 302 and just seamlessly transition.
I’ll never use Tor because I have no idea what the Tor client is actually doing. Is it enabling someone to use my network connection for cybercrime without my knowledge? No thanks.
Clients are never used as relays in TOR. You never route anyone's traffic until you setup it yourself. And you can't miss that part, and it's not a default, and requires additional configuration.
Also relays (not exit nodes) are pretty safe to operate and running them is a decent thing, supporting free internet instead of a corporate ads machine, let's not frame it as a "crime support".
> Also relays (not exit nodes) are pretty safe to operate and running them is a decent thing, supporting free internet instead of a corporate ads machine, let's not frame it as a "crime support".
Well the purpose of using Tor is to prevent any network operators from knowing who you're talking to. Which AIUI is primarily a concern if either you're not allowed to talk to whoever ("great firewall" type things), or you risk getting in trouble for talking to whoever (Silk Road etc, or disfavored politics).
I guess if you're worried about hacks and doxxing rather than LE? Or if you only call things crime when they should be illegal rather than when they formally are?
LE relies on opsec failures which is very clear on their busts. They are incompetent hypocrite fools.
Using Tor browser and running a Tor node are different things, by using the browser you are not contributing to the network, you're just accessing it. If you're worried about someone using your network connection for cybercrime you shouldn't run a Tor node (although there is significantly less risk with a relay node), but you shouldn't worry about using regular Tor.
> by using the browser you are not contributing to the networK
That's false to some extent. Tor's promise comes from it's vast population of users. The more users it has, the better it is to improve everyone's anonymity. So in a way, even by using it, you are helping Tor network. And please, save the "criminal" bs (meant for the original comment).
This response stretches "pedantic correction" to new levels.
Why? The utility of any network grows with the number of participants, even that of inherently asymmetric networks that strictly distinguish "producers" and "consumers". (More eyeballs make the network more valuable to content providers.)
This might not be how courts determine culpability of redistributing any potentially illegal content, of course.
>This might not be how courts determine culpability of redistributing any potentially illegal content, of course.
Which is precisely the point of this discussion.
Might as well argue "By protecting the environment you're supporting the drug trade, because people that a climate catastrophe would wipe out will be able to be drug users".
It's literally outline by Tor Project team, if you care to even read from official sources.
This here response continues to stretch "pedantic correction" to new levels.
What's "literally outlined" I'd guess is that the utility of the Tor network increases with adoption which nobody ever doubted.
What is discard is the tenuous over-stretched argument in this thread regarding fears of legality, that went like this:
GP: Using Tor browser and running a Tor node are different things, by using the browser you are not contributing to the network, you're just accessing it.
P: That's false to some extent. Tor's promise comes from it's vast population of users. The more users it has, the better it is to improve everyone's anonymity. So in a way, even by using it, you are helping Tor network.
As if that what was discussed...
You are being pedantic and annoying. Why would you even respond in off-topic response? Your comments have served nothing to the topic on-hand.
As others have mentioned, that's not what Tor does by default. Just because you don't know how it works doesn't mean that it's generally unknowable.
And conversely, it's enough to visit a random website running WebTorrent or just a plain HTTP DDoS attack to possibly "use your connection for cybercrime".
Since RFC 3514 unfortunately never gained traction, distinguishing good, bad, and illegal traffic remains difficult.
Tor doesn't work like this. i2p, however, does. At least by default.
I highly doubt nostr’s suitability as a petabyte DFS/CDN.
This is a discussion on whether or not it is better to use it to announce new domain names post-suspension than Wikipedia, not if it can sustain petabytes of data.
Given the archives use of BT, probably better to just use BEP-44 on mainline DHT if you want to store an arbitrary address somewhere - or use something like IPNS, or .onion as others have mentioned.
As nostr relies on gossip, there is no guarantee you will have access to latest address.
I am unaware of an app that sees present day use for communicating short messages that uses BEP-44 or mainline DHT. Nostr works today for millions of people. These services have normal people using them, not Hacker News commentators that are extremely sophisticated -- imagine telling someone who only knows how to use normal web services and apps they install from the app store to "just retrieve a BEP-44 message from the DHT".
I’m not aware of any laypeople who use nostr either.
But if you are going to build some nostr-based “resolver” into a browser you could instead use any of the other protocols - 2 of which are designed specifically for resolution, and the other already having a robust resolver implementation built on top of it?
I quite like nostr, but let’s not pretend it solves every problem.
Trying to understand nostr, I looked up its Wikipedia page...
> In 2024, in an article reporting on the project's funding, Business Insider claimed to have identified fiatjaf, and had found two websites previously published by this person to disseminate the work of Olavo de Carvalho, a far-right conspiracy theorist.
That... seems extremely irrelevant. If fiatjaf is contributing something useful and significant to the commons, why does it matter that he used to spread far-right conspiracy theories in the past?
> As a result of its ability to quickly and discreetly create accounts and publish posts to relays, Nostr can propagate spam much more easily if left unchecked. A notable example includes a case where multiple protocol bridges have been used to conduct spam waves on the Bluesky social network (itself connected to a competing protocol, the AT Protocol) by creating posts on Nostr, bridging the post to ActivityPub and bridging it again to Bluesky.
Surely they also had to create a Bluesky account for that? I don't see how Nostr is to blame here. Perhaps Bluesky forgot to use anti-spam measures when bridging things over from other sources? That's kind of on Bluesky, no?
This reads like a smear campaign against Nostr. I don't think I have the necessary Wikipedia karma to get it amended, but gee do I have opinions on this...
> This reads like a smear campaign against Nostr.
It's well known that corporations and governments pay people fulltime to edit Wikipedia. Wikipedia has a whole article detailing the extent of it:
https://en.wikipedia.org/wiki/Conflict-of-interest_editing_o...
Of course, these days the people paid to do this have learned not to do edits from their own corporation or government office's announced IP blocks. But in times passed finding many of this category of edits was as simple as sorting edits on Wikipedia by the originating IP address and looking for which ones came from institutionally announced subnets.
Point being, massive amounts of capital and intelligence resources have been dedicated to censoring social media. There's nanny employees in every single social media company making sure "important" complainers are heard and their desires to silence voices fulfilled. I follow a large number of people on Nostr that have been banned from every other platform. Facebook. Twitter. Bluesky. "Free speech" sites like Gab and ActivityPub servers that advertise "free speech". But Nostr has the same entrance requirements and cryptographic sovereignty that Bitcoin provides. Generate a keypair and you can publish. People that want to find your content can simply subscribe to your public key. This results in a subversion of countless state and corporate capital expenditures. If people use Nostr, they will permanently lose the ability to moderate content in this oppressive manner. They absolutely do not want this to happen.
> Nostr can propagate spam much more easily if left unchecked.
Nowhere have I had a worse problem with spam than Twitter and Facebook. For all the alleged vulnerability of Nostr to spam, it has not currently materialized as an issue.
Note that filtering out actual spam without a centralized moderator is one of the most solved problems on the Internet. If you've ever installed Spamassassin or other well subscribed to Bayesian filters on an email server, you know that you never see spam ever again. In actuality, spam is a much bigger problem when you are dependent on fickle human moderation.
> That... seems extremely irrelevant.
I find it pretty relevant who is behind what.
Definitely a smear campaign, ironically one that seems to be organized by left-leaning individuals on Wikipedia, Business Insider, etc.
Which is bizarre to me because aren't these the people that would want the ability to disseminate information in the face of fascism?
They are attacking their own side (again.) When will idealists learn that this is not the way?
> left-leaning individuals
Here we go again.
As a 'left-leaning individual' it's funny because if you look up anti-war left leaning outlets and such on Wikipedia, they don't tend to have exactly glowing entries on there. Wikipedia and the other outlets described as 'left-leaning' are neoliberal institutions. Believe me that there's no love for these on the left.
When it's convenient for smears, neoliberals are left but then at other times it's the communists etc. In other words, 'left-leaning' is a grab bag of what one doesn't like these days, rather than any really meaningful group.
> Wikipedia and the other outlets described as 'left-leaning' are neoliberal institutions.
What exactly do you think 'neoliberal' means?
I do agree Wikipedia is not 'left-leaning', mainly because 'right' and 'left' are bullshit names that don't mean anything. But it doesn't even have the power to act in a situation that would make it neoliberal.
It absolutely does. It's full of editorial decisions. What content is on there, what is cut. What editors consider germane, etc.
It can absolutely act in a way that makes it neoliberal.
Neoliberal as in prominent decision makers/editors etc, such as Jimmy Wales express the sort of free market and foreign policy philosophy that has been mainstream since about the 80s.
It means that entries on individuals, countries etc. are broadly in line with what you'd read in any mainstream media outlet and so is its outlook on 'Western civilization'.
That doesn't mean it's not a good project, or that it has some great power, just that its 'gatekeepers' are not exactly dissidents of any sort.
> express the sort of free market and foreign policy philosophy that has been mainstream since about the 80s
Let's see:
https://en.wikipedia.org/wiki/Laffer_curve
Right on the introduction it clearly says that any argument based on the curve is pseudo-science.
https://en.wikipedia.org/wiki/Washington_Consensus
Is biased in claiming the consensus is a contentious topic, instead of only a tiny well founded minority ever supporting it. But it's the same bias you will see in any history book.
If we go extreme in another direction, this one has the same bias of representing fringe views as equally represented in a debate:
https://en.wikipedia.org/wiki/Communism
If we go here:
https://en.wikipedia.org/wiki/Management_science
There's a clear neoliberal bias. But if instead we go here:
https://en.wikipedia.org/wiki/Public_administration
There's a strong modernist bias, with a secondary classical liberal one. What is about exactly the same bias you would see on the main literature of both subjects.
So, no, except for behaving like an encyclopedia and reflecting the literature biases, I fail to see how the wiki is neoliberal as a whole.
> What is about exactly the same bias you would see on the main literature of both subjects.
I think you just answered your own question.
Why do out single out the neoliberal one if each subject clearly has a different bias?
The defining feature of fascism is these kind of tyrannical public-private partnerships. It was the entire basis of Mussolini's fascism. They are the literal fascists.
Redifining fascism as meaning "racism and anti-semitism" (certainly attitudes which by the current definition far predate fascism) has been one of the most clever acts of sleight of hand by the regime, giving it unlimited freedom to enact the most totalitarian form of fascism ever conceived.
> aren't these the people that would want the ability to disseminate information in the face of fascism?
Everybody wants free speech — but only for opinions they agree with. And they are against censorship — unless the "right people" are censored.
Recently, the left has been far more authoritarian, labeling everything they don't like as "far right hate speech", pushing to make dissent illegal, and demanding censorship. I guess the pendulum will swing the other way eventually.
It's not really a left VS right issue, but an authoritarian one. Free speech can be uncomfortable, that is the point. "Free speech, but…" does not work.
> Recently, the left has been far more authoritarian
I'm not sure how a reasonable comparison of authoritarian behavior seemingly assigns more weight to random Wikipedia contributors lumped together as "leftists" compared to the literal government currently controlled by the right that is routinely threatening to pull FCC licenses for critical speech among other intentionally speech chilling threats.
I'd say the pendulum has already swung the other way, while swinging much, much further and more openly than nebulous mob demands for "cancel culture", over zealous Twitter moderation of hate speech or whatever else the previous go-to examples for the left were. Before 2025 showed what a truly authoritarian anti-free speech policy looks like when wielded by those with actual legal power and zero shame.
You don't need "karma" to edit Wikipedia.
Some pages have 'semi-protection' or 'extended semi-protection', where a user must have a Wikipedia account which must have made a certain number of edits and be older than a specified time period.
However, this isn't one of them.
You can edit it immediately, either with or without creating an account.
> Once they finally have a solid court order to seize domains, generally the rate at which they get seized accelerates greatly.
Given that most ccTLDs live in different jurisdictions, that's not really a huge problem.
> Given that most ccTLDs live in different jurisdictions, that's not really a huge problem.
The copyright-industrial complex is internationally very well-connected.
That may be, but you can't just keep going to the same US court for warrants.
As long as there is centralization, there is always an avenue for abuse with money. The DNS root itself is heavily influenced by a group of allied nations, through the ICANN if I'm right. That can be used to exert pressure on TLD registries, including ccTLD registries. Of course, that cannot be used for surgical control single domains like Anna's Archives'. But DNS blocking is an old technique by now. The copyright cartel needs to get it banned only in a few populous countries to destroy the value of a domain. We can keep finding workarounds. But at some point, they won't have to worry about people who can actually do that.
For what it's worth, that doesn't seem to ever have worked.
https://annas-archive.se/ Works still
https://open-slum.org/
Why does SLUM say that all the Z-Library links are dead? They work perfectly fine for me?
Seems like SLUM is down.
Just the HN hug of death. Here's another
https://open-slum.pages.dev/
I remember swapping lists of IP addresses of FTP servers ~1990 when there was no DNS/Domains in Germany working on the internet.
There was very little DNS in the early days. I just had big notepads full of IPs next to the computer. Especially for the pirate sites.
They could also potentially write their latest address to the Ethereum blockchain instead, assuming there is a client that could scan and resolve. This would be bulletproof.
Nation-state hackers deliver malware from "bulletproof" blockchains - https://news.ycombinator.com/item?id=45860258 - November 2025
Some sites like Anna's Archive have .onion site for the Tor network, and others do not. Is there a considerable downside (DDOS?) to providing access to their site by those means?
I have seen Tor network take a hit when there's fight between darknet marketplaces. It's certainly an issue.
not anymore. Most recent onion sites have enabled proof-of-work mechanism in place, so DDoS attacks are now expensive to carry out.
I am aware of POW but that hasn't stopped the DDoS. If you look at chart from few months back, there was significant hit due to crazy DDoS carried out by DrugHub which was evident in a Dread post.
link the chart here, so I can take a look.
https://metrics.torproject.org/userstats-relay-country.html look at the spikes.
This happened a couple days ago, not "a few hours ago". I was surprised torrentfreak hadn't posted about it at the time -- maybe took them a while to notice? I tried to hit the site on Jan 2nd, 10:40pm pacific timezone, and the domain was inaccessible then.
You could consider emailing them if you notice things first in the future
https://torrentfreak.com/contact/
<sarcasm> I thought that Anthropic and OpenAI created precedents that legitimize libgen and Anna archive. They pillaged all the books and scientific papers from those archives and got away with it. Why not the rest of us? </sarcasm>
Be careful not to confuse using the material and distributing it. There are open legal cases sorting out what fair use means for generative AI. Distribution (seeding in the case of torrents) of this material isn't legal. It got Meta in trouble, and it's getting Anna's archive in trouble.
> got away with it
Anthropic paid a settlement of $1 500 000 000 to authors.
Doesn't really mean much until individuals can also sacrifice a small percentage of their ARR to completely ignore IP law.
I'd take that deal, but until it becomes and option, we have a clearly broken system. Rules for thee; not for me, etc.
People always say this like the tech industry wasn't culturally anti-copyright and pro-creative commons before. Those same people probably work at Meta and Anthropic, just like Google's book project which got them in trouble.
> People always say this like the tech industry wasn't culturally anti-copyright and pro-creative commons before.
I completely agree with that. The problem is that the current system is such that only billion dollar players can flout the rules, while everyone else is left in the dust.
Worst of both worlds IMO.
Others already mentioned they lost their lawsuit. Should the fines have killed Anthropic? Would have been more fair and a less bad world?
Why not focus energy on being anti-aggressive copyright in general. These system won't ever be fair. It's just rent seeking enabled by the government and some people can afford the rent.
You’re talking past me for no real reason, mate. That’s precisely the point I’m making.
Young Carlos thinks it matters that Anthropic got sued when they can keep flouting the rules anyway, and I disagree: it’s not a fair system until we ditch the rent-seeking entirely.
They paid one of the largest settlements in world history. Should I guess that hackers are only satisfied with the public execution of the company leadership?
To pick a nit: Technically Anthropic didn't loose any lawsuit or pay any fine. They came to an agreement with the authors to pay them a $1,5 billion settlement. Which was a lot of money per book.
And a lot more expensive per book than buying, digitizing, and destroying the physical copy. Which was ruled legal.
Wasn't the Google project scanning physical books and not distributing them externally? That seems like a very different thing than torrenting or even downloading stuff uploaded by a third party.
They can: VPNs.
Fair point, but I think the Pinkertons would be at my door within the hour if I started re-appropriating the art style of Studio Ghibli or Disney for commercial profit.
Why the negativity? You can also as an individual do the same as Anthropic and get sued for billions. You have that option, don't let anybody hold you back!
they should've released the spotify dump anonymously
That would lose the PR advantage.
Isn't Anna's Archive already run anonymously? So they sort of did...
What is the recommended way to use or download/mirror Anna’s Archive now?
alternative tlds you can visit: .se .li .pm .in
We love you Anna!
And that's why helping torrenting and seeding the content of AA is vital: they can take down a domain name but not block everyone who seeds.
I said this before but if you've got some spare GB/TB on a computer/server, consider "donating" it for culture preservation purposes:https://annas-archive.se/torre nts
I wish I knew what I was seeding though
And I don't mean the copyrighted part (I don't care about that) but the actual content which might be questionable
You literally download the items you are seeding. Just open them and see for yourself.
Looking at mutable torrents recently and it's ability to circumvent censorship via DHT.
I wonder if it could be revived.
The only problem is the mutable torrents standard is in draft and not adopted widely. I think I saw someone proposing using DHT in a way that allows them to host websites. If feasible becomes very difficult to take down.
If anything this is a very strong argument for a completely decentralized domain name system.
The DNS is much too fragile in light of all of the recent developments.
> If anything this is a very strong argument for a completely decentralized domain name system.
As you probably know, this is a very old idea, dating back almost from the birth of bitcoin [1] [2].
Real shame it didn't take off.
[1] https://www.namecoin.org/ [2] https://en.wikipedia.org/wiki/Namecoin
There is nothing stopping anyone here from registering a new shiny .org domain (or any tld) and pointing it to Anna’s Archives name servers.
Except they are probably not set up to respond to new domain names and generate HTTPS certificates for them.
They really ought to get an .onion site up.
I don't understand the premise. .ORG was never a safe-haven TLD in terms of abuse protection. Juicy apple falls from tree, story at 11?
thepiratebay.org is still up
Don’t copy that floppy!
Piracy is killing music.
Streaming is killing music piracy. The scene is a shell of its former self.
Hardly. Soulseek has been gaining interest in recent years, for one:
https://trends.google.com/trends/explore?date=today%205-y&ge...
Your link only shows past 5 years, so you're missing the hayday of music piracy, was this intentional?
Try this: https://trends.google.com/trends/explore?date=all&geo=US&q=s...
Try with any music piracy (or hell, piracy in general) related search terms and you'll see a similar trend.
You wouldn't download a car.
(I mean: You wouldn't steal a car. :-) ).
That was always such a silly argument, even more so with the rise of cheap and reliable 3D printing. I download, print, and use physical objects every week and don’t know or care what IP issues might exist as long as it works for its purpose. I can’t wait until the day comes when you can download and print a working car at home.
But such working cars from home will not be certified to participate at the public traffic and remain toys for driving on own private property.
No, it is possible to register a homemade car for use on public roads, pretty commonly done actually.
Then I'm surprised. It's much different to the usual situation when even mass made products are either banned or very limited, e.g. unicycles prohibited, e-scooters only up to 20 km/h, e-bikes only up to 25 km/h.
Regarding 3D made parts recently there was an accident: https://news.ycombinator.com/item?id=46152941. One may expect increasing regulation. At least in the air.
I agree, it's surprising. Even in California of all places, you can make a one-off vehicle at home and register it for legal road use, without even needing modern safety or emissions equipment:
https://www.jalopyjournal.com/forum/threads/registering-as-a...
Another even more common strategy is to "restore" a classic car using some extremely small number of parts from some really old pre-emissions and pre-safety equipment car. This is often done for hod rods, dune buggies, etc. where it will be, say a "1930 Ford" but contain only some minuscule amount of that original car it is titled and registered as. There's a sizable industry of homemade "kit cars" that require you to start with a legally registered VW Beetle, but ultimately they often retain nothing except parts of the thin sheet metal floor pan, and somehow that is apparently legal.
so every .org domain is now subject to US law system ?
Of course they are? PIR is American.
Yes, and they always have been.
They could always use Yggdrasil.
I have no clue why you're getting downvoted but ygg is a solid option but it would overwhelm the network since AA has intense traffic.
IPFS then. It's just static content, so it has sense to be distributed between nodes. Not under ygg of course.
On the YGG network, I've used streaming without many issues, plus BT. And people have been even gaming over YGG.
Altough they could use YGG for a static address and then just use IPFS for the data but outside of YGG.
I like IPFS but decentralized storage is just awfully expensive to maintain without any reward. Main IPFS mirrors don't tolerate AA content, it's taken down very quickly.
And also IPFS rats our to the whole world all your network interfaces, MAC addresses, and internal network configs.
And even after multiple requests of supporting a Tor mode, have routinely ignored that with "but its too hard!"
And, I quit running IPFS back in the .31 version after adding some chat logs to my local machine's share, and found a google crawler within 1 hour and fully indexed them.
No thanks.
IPFS support interface binding, hard disk quotas, working as a router or not, and bandwidth quotas among several settings. Yes, by default can be a bit intensive.
What? That's the laziest argument against IPFS. Bind it on one interface or just use a VM.
Defaults absolutely matter.
And no, I dont my whole internal network enumerable because of 'convenience'.
I don't think the kind of crowd IPFS attracts would wanna use the default config. Most ould choose to configure according to their setup.
You are right; it's a bit of a excuse. Tons of small services under Unix (by default) listen on several interfaces and networks. I think Bitlbee itself listened on 0.0.0.0 instead of localhost. Ditto with some UPNP/DLNA daemons.
It takes very little to set up IPFS to just listen to tun0, disable routing (let ygg do its job) and throttle the bw a little so it doesn't hog the whole network.
I would be surprised if the Kubo/IPFS developers didn't already configure Ygg for themselves, as both are software written in Go.
The author of NNCP https://nncp.mirrors.quux.org/ (and the other author from Tofuproxy) are pretty much aware of Yggdrasil.
Funny how HN is usually ripe with folks crawling out of the woodwork to defend IP ownership as a fundamentally important principle, both on moral and economic grounds.
Yet, as of 08:37:36 MST Monday, 5 January 2026 there isn't a single comment on this thread complaining about Anna's IP theft.
Don't get me wrong, that makes me VERY happy, I am a firm believer that the very notion of IP ownership needs to die a horrible death, something that AI may very well make happen in short order, yay.
But still, I can't help but wonder why the "this is IP theft" crowd is completely silent when it comes to the like of pirate bay and Anna.
> the very notion of IP ownership needs to die a horrible death, something that AI may very well make happen in short order, yay
I agree with the disdain for IP, especially with what it had grown to become nowadays. But while I was also initially optimistic that AI companies may find a way to make IP go away as a byproduct of their activities, now it seems more like the big businesses will cut deals with one another and leave us commoners with nothing. Entertainment megacorps and AI companies rule the world, and I have no doubts that they'd find a way to become close allies. The AI companies get their near-endless stream of training data, the entertainment industry gets a cut of that juicy AI money and gives away their data willingly, while the IP remains locked away from ordinary people for eons more, just the way they like it. No one but us wants IP reform, or at least no one with real power, so it will probably never happen.
> I am a firm believer that the very notion of IP ownership needs to die a horrible death, something that AI may very well make happen in short order, yay.
The leading AI labs are not killing IP. They are taking IP and reshaping/combining it to produce their own highly lucrative proprietary IP package which they sell to you.
The mirror image of IP defenders are AI boosters who argue against IP when it comes to slurping up media but squirm when you say "ok, then publish all of the inputs that go into making your frontier models, and publish the model weights too."
Possibly because context matters to many people.
AA is not stealing every byte of data they can get in order to make billions of dollars, collect personal data about people, and then sell that for even more money.
yeah no shit, this is what happens when you agitate the major music record labels - it's going to get worse
In that case it would be Elsevier and co.
Anna's Archive recently released a dump of the entire Spotify catalogue so they are very much in the music industries crosshairs as well now.
Only metadata has been released so far.
A lot of which was contributed by users typing in CD play lists to CDDB which got corporatised by GraceNote and ended up somewhere in Sony land.
So this effectively re-releases into the public domain a lot of the user contributions during the 1990s.
They only distribute outdated SciHub Records
That's not true. Anna's Archive gets new paper uploads from Z-Lib and LibGen as well. It's much more up to date than SciHub.
Yep, let's accept the monstrous industries which lock down culture for money.
I for one support their efforts. The same way we store seeds in vaults deep in the depths of the earth, we should do this for digital content too, and without retaliation from any specific industry.
What would be better is solve the root problem. These (illegal, somewhat legitimate) hoarding sites are most valuable for research literature which, given the public funding nature of these things should not be gated to begin with.
The comsequence of resolving the symptoms is that illegitimate use piggy back on it. Artistic literature that would legitimately deserve protection get hoarded as well.
Sweating authors of clearly copyrightable arts, typically novels, manuals, are seeing their work accessed free of royalties. For the sake of freely distributing scientific literature.
It makes it impossible to make then distinction given the legitimate utility is operating in a dark domain.
Yes, we should archive everything. And we should perhaps reform IP more broadly and re-think how we treat our culture. And we shouldn't expect retaliation.
But retaliation will happen, and I worry that it's going to pull down one of the most incredible archives along with it.
TPB is still up, 20+ years later, despite plenty of attention from similar groups.
Agree with you and I'd also like to see more abuse by said companies punished
They went after Pirate Bay by literally threatening trade war repercussions with Sweden which is far more destructive than any files downloaded
GitHub did it for source code.
https://archiveprogram.github.com/arctic-vault/
also have the feeling that going for music is going to turn out badly
Surprise to who exactly? Anna’s Archive has been one of the most blatant piracy outlets to come around since the days of Napster.
That seems a tad judgmental.
Anna is welcome here on the Group W bench.
Excuse my FUD, but are they really non-profit as described in FAQ? I find it a little hard to believe given how aggressive they are at marketing the paid version.
As someone who runs a charity, it surprises me how often people equate "non-profit" with "doesn't get paid" or "does everything for free". Non-profits still need money, they just don't use that money to make profit that is paid to shareholders. Instead, they use the money to further the stated objectives of the organisation.
NB: I'm not trying to imply anything about whether Anna's Archive is non-profit or not. Just that the fact that it aggressively tries to raise money isn't a relevant factor.
Check their gitlab for bounties. They have a lot of money in reserves to be spent on that. There are a few open bounties for hundreds of thousands of dollars. I've also worked on a bounty for them once and can confirm they have plenty more money in their budget. So I believe that they don't pocket donations, or at least not a very large portion of them.
They provide LLM data sets for 'donations'. They collect other open collections and beg for paid memberships to access pirated material. Its a russian project. You can for sure assume that someone is earning something somewhere.
> Its a russian project
I've seen people claim this a lot, but is there a single proof that supports this? The only potential insights into AA owners was an arrest of a few people suspected of running the site, and they were Latin American, not Russian.
This kind of implies a legitimacy that AA just doesn't have. They're not 'a' non-profit, and there's simply no way to know that they're not taking in loads of money and pocketing it.
> They're not 'a' non-profit
How do you know?
> and there's simply no way to know that they're not taking in loads of money and pocketing it.
Or that they are, right?
They are not a non-profit because they are not a registered company/organization.
There is no requirement for a non-profit organisation to be registered.
There is. Without a legal framework its just an association that claims not to make a profit. This is all word games though as be clearly don't agree on the definition of a non-profit organization.
> There is.
No. In UK, many an unregistered association is a non-profit e.g. sports clubs, volunteer groups, societies.
In UK there is simply no such status as "non-profit", so technically none of what you listed are "non-profit" organizations. These are unincorporated associations, that, by the way, can be making profit (but that would be on shaky grounds, since common law makes it hard not to fuck up, being as vague as it always is). However, there is a "charity" status in UK, and you'd have be a registered organization to obtain it.
Anyway, "non-profit" is (i.e. "only makes sense defined as...") a legal status, it isn't just a way to say "not making money" (after all, we wouldn't call any failing business a non-profit, right?), so it really doesn't make any sense to ask if an illegal underground gang is "non-profit". GP is correct to point that out.
> In UK there is simply no such status as "non-profit"
Incorrect.
‐--------
https://dictionary.cambridge.org/dictionary/english/non-prof...
noun [ C ]
uk /ˌnɒnˈprɒf.ɪt/ us /ˌnɑːnˈprɑː.fɪt/
(also not-for-profit)
an organization whose aim is to make money for a social or political purpose or to provide a service that people need, rather than to make a profit
(Definition of non-profit from the Cambridge Advanced Learner's Dictionary & Thesaurus © Cambridge University Press)
‐--------
That's Cambridge UK.
> However, there is a "charity" status in UK, and you'd have be a registered organization to obtain it.
Again incorrect.
See https://www.gov.uk/setting-up-charity
I'm not entirely sure what this sort of pedantry is meant to accomplish. Congratulations on being technically correct, I suppose?
Both I and GP have pointed out the crux of the matter, and you are ignoring it.
By pedantry, you mean "pointing out completely misusing terms and falsehoods about UK law", right?
> Both I and GP have pointed out the crux of the matter
"They're not 'a' non-profit".
> and you are ignoring it.
No. I am challenging it.
I've seen no evidence either way and so far it seems nor have you. So let's wait and see.
Literally everybody except you knows what I and both GP were referring to.
Everybody including me, I'd say.
When people talk about a "non profit" entity, they are almost certainly referring to a legal non-profit. Maybe that's not what GP meant.
And yes, you are technically correct in that there's no way to know that they are or are not pulling in money and pocketing it. I'm not sure how that's relevant to the questions asked but okay.
> When people talk about a "non profit" entity, they are almost certainly referring to a legal non-profit.
In UK, a non-profit association does not have to be legally registered to be legal.
knowledge for me but not for thee
It might be a sincere effort in the beginning, but at this point, Anna is merely AI company’s foot soldier.
Do you mean, intentionally? It's an interesting question; is there any evidence?
They do have this section about supplying AI data: https://annas-archive.se/llm
I certainly don't consider taking money from AI companies while giving data to everyone else for free makes them AI company footsoldiers, but that offer is presumably what's being alluded to.
It gives AI companies influence, if they supply a significant amount of AA's budget. It also may give AI companies knowledge of AA's people and assets that could be given to authorities.
Not sure intention but stating the de facto state from my observation.
Which - now that the various commercial LLM companies have harvested the archive, they will be more than happy to use litigation to raise the drawbridge behind them, so as to block any future competitors from acquiring the same training data...
That would be a very interesting and ironic timeline.