Bypass Paywalls Add-On Takedown Notice | Hacker News Reader

Bypass Paywalls Add-On Takedown Notice

github.com

87 pointshongspike50 comments

I can't believe Github processed this incredibly spurious DMCA request that didn't even detail what was being infringed beyond some vague "anti-circumvention technology". Plus it checks for archived and cached versions of an article so that somehow makes it circumvention?!

The author has already been kicked off Gitlab, so I'm not sure where else realistically they can go that wouldn't fold to copyright holders. Sourceforge? Launchpad? It's almost like it needs to be self hosted on IPFS or published on Bittorrent.

They haven't posted any updates on Twitter yet, but this is it: https://x.com/Magnolia1234B

The reason the author is having such a hard time is because they want to stay anonymous and there is no reliable way of self hosting something (outside the US) at a reasonable cost while staying anonymous. So, they keep hopping from one domain to another and get hit with DMCA everywhere. There is no winning move here.

Maybe the new Russian service will work for them. Although how hard is it really to host files in the <1MB range?

How will you host any files anonymously, no matter the size ? I don't mean bittorrent, ipfs, tor and the likes. Plain old website needs you to pay a hosting provider and a registry for a domain, both of which require your name and some form of payment. If you have the resources, you can create an LLC or find some other workarounds, but it's a hassle and not cheap.

IME for most TLDs and hosting providers there's no KYC to speak of. You can just fill in anything, and keep doing so. Unless I'm completely misremembering things, but I signed up to a hosting provider quite recently.

That works, to varying degrees, until the provider gets a letter from law enforcement. And unless you find some provider willing to take crypto, you are doxxing yourself when you pay.

yes it's better to keep a digital paper trail in a public permanent log.

Several domain registrars / hosting providers (including Njalla) accept Monero, which is as untraceable as you can be. Or Monero can be used to indirectly pay a bitcoin invoice very easily (even Namecheap accepts bitcoin)

can you buy monero with cash?

It doesn't matter. You can buy Bitcoin with cash, convert the Bitcoin to Monero and then as soon as you send that Monero to your own wallet, you're done. You have digital cash to spend anonymously.

Ideally, cash with no INSERT IDENTIFICATION so there's no flow of value to audit. ATMs are out!

which limits it to very small amounts or the proceeds of crime.

The way Monero works, you can simply have two monero addresses/IDs. The one you get the initial coins (cash, bank transfer, doesn’t matter), then transfer to the second one. Then you can buy whatever you want anonymously from the second one. That’s the whole point of monero, that it is not possible to trace transfers, to the contrary of bitcoins where every transfer is in the ledger.

Or am I missing something?

> Or am I missing something?

blinding hatred of crypto? Otherwise, nothing as far as I know.

Who hates crypto on this thread? I rather saw hatred of ATMs =)

the opposite of blind faith is suspicion/disbelief, not blinding hatred.

That's fair, I take back my comment. I don't think it was very fair to you and the dialog in the thread. Goes for the other replier as well.

isn't this the point that the people in power start to look into your finances and asking questions? especially if copyright is involved, there is effectively infinite resources to be spent on chasing you. you need to remain careful not to slip up every single time - they just have to be lucky once.

Yes, this is the crucial point. It's probably not even that hard to get a pile of semi-clean Monero. Cashing it out, paying or avoiding taxes on it, and avoiding lifestyle signs. Best I've come up with is that I'd spend it all donating crypto to OSS projects that I want to see succeed.

neocities?

There are services like https://njal.la/ which can be used anonymously, though they cost more.

There are a variety of services like Tor2Web that make accessing .onion sites on the plain internet trivial. For MOST .onion sites you may well not want to do this, but when all you want to do is hide the identity of the publisher, rather than worry about the privacy of the visitor, it's not really an impediment.

There is the matter of bandwidth -- this is where being pulled into other repositories and copied would be helpful. The code is small, but tor is notoriously slow.

Really? It seems fair to me.

Like I love bypassing paywalls but the extension is clearly away of circumventing a system that's meant to make you pay for content... Isn't it?

I think the fact that they publish the site in a readable state first and adding the paywall after some time (after getting indexed / shared a bit) and services like archive.is just provide a copy of the original state makes it debatable if this is a circumvention.

If I publish a project on GitHub with a BSD license and after some time switch to a proprietary license the old code would still be licensed under BSD, right?

The content published on the site in the open was never licensed to you though.

Lets say I make a movie and do a free screening of it in a public park to get people talking about it. That doesn't mean that it's ok for someone to record that screening and for people later to go looking for that recording rather than paying to watch the movie.

And being sent to a computer and stored in memory to render it to you doesn't change that either. That's necessary for the work to be displayed to you over the Internet and the law understands that (see: "What Color Are Your Bits?")

I like your movie screening example.

Let's say you want to watch a movie about a certain topic. You ask a couple of friends if they can recommend one. They respond by telling you about this free movie screening at the local park - the movie is exactly about the thing you're interested in. You go to the park. Five minutes into the movie they start asking for money or you have to leave.

- I think your friends might not have recommended the movie if it wasn't free - you might not have gone to the movie if it wasn't free - nothing on the outside of the movie screening area indicated that they switched to a paid screening

I feel like there is a difference between a free screening to get some publicity and a simple bait and switch.

But is that illegal? Or is that violation of copyright?

You're accessing content that can be viewed by Googlebot, or by disabling Javascript or some other simple browser tweaks.

Who cares if it's a "violation of copyright" ? Copyright has become a set of overbearing corrupt laws bought and paid for by corporations. Sure there are some vestigial bits in there that an individual can kind of use if they do everything exactly right, but the overall dynamic is skewed very hard in favor of corporate control.

Apart from the blatant abuse of the legal system to computationally disenfranchise individuals via the DMCA et al, the main issue here is using centralized websites as watering hole for authoritative development/distribution. It would be understandable if Github were merely a mirror for exposure etc, but that should be the extent of it.

I think it's time for anonymously hosted .onion and .i2p git repositories, ideally with IPFS and Hyphanet backup in the event the black helicopters find the hosting systems.

Given the arrest of the Telegram CEO in France, I tend to agree. We've grown too complacent about trusting the public internet to host content that the censors will want to do something about. And the more gets done on tor, presumably, the more bandwidth the network will have, given the way it functions.

Gitea hosted in Russia or Iran or similar jurisductions that would SHIFT+DELETE a DMCA request.

The author has a Russian mirror

https://gitflic.ru/project/magnolia1234/bypass-paywalls-chro...

Cool, thanks for sharing

DMCA is a weapon abused by corporations. These laws are written solely for corporations.

What I do an end-user is my discretion; if I want to remove or alter data sent to my endpoint, I should be able to without some corporation dictating what can be done.

Everything on the web needs decentralisation as we're nearing some sort of dystopia where corporations dictate how we live. Git itself it decentralised, but the hosting of files shouldn't be so easily removed.

I'm curious how much time it will take for most FLOSS devs to understand that relaying on third party giant code hosting instead of put back usenet (it's cheap enough) to have visibility and keep the code REALLY distributed and REALLY developed in a distributed workflow...

We have witnessed various GH and co bans for USA orders, often for no reasons (like a Belarus game dev who have nothing to do with foreign gov or intelligence, a USA citizen who have been in Iran for a little time and so on), DMCA and so on. It's astonishing how many develop FLOSS without understanding such basic consideration on depending on third parties opposing the classic "ah, I'm here only for visibility" and than start using PR, CI, ... not even understanding the threat of that and the scale effect.

> put back usenet (it's cheap enough) to have visibility and keep the code REALLY distributed and REALLY developed in a distributed workflow

You absolutely don’t need usenet for that. Git is already distributed and any mailing list is fine for discussion. No need for nntp and the associated complexity.

The real issue is that most people have no idea about how to send and receive patches, share git bundles or use request-pull (or really use Git if I judge by the standard of my office) outside of GitHub.

Oh no, that's absolutely needed, because yes, we can develop with git sending patches via and ML, IF people know your project exists. That's why we need usenet where we can have a freshmeat.net and co equivalents to makes new projects visible. This very visibility is the reason why GH and co are popular, because devs attract others devs who casually found their project, and Usenet have a proven history of making knowledge travel fast.

About people knowing how to develop with an ML and patchsets, the issue it's not much in people but in the substantial absence of modern MUAs though ALSO for this simple automation. I use notmuch-emacs, but I can't propose it to most others simply because deciding to keep maildirs on their own homeservers, most do not even have, muchsync-ing against it, maybe also adding a web-MUA and a dovecot to serve the mails on the go also on mobile, ... it's simply way too much work.

That's why again we need NixOS/Guix System to simply say "hey, import this code, change few parameters and there you go your homeserver FULL setup with anything, just choose other modules to discover the FLOSS world.

This damn need usenet to allow newcomers discovering what we have, learning how to use, and learning a model almost all have never known or forgot for some crappy colorful WebUI/walled garden.

We do not need copies of the big tech stuff, we need to show better stuff from a model where big tech can't even exists economically. That's the FLOSS superiority and strength no company can dream to match if it's done on scale.

Here's a mirror [1], linked from the bpc-clone/bpc_updates repo [2]

[1] https://gitflic.ru/project/magnolia1234/bpc_uploads

[2] https://web.archive.org/web/20240813013640/https://github.co...

I'm not sure I would trust a Russian mirror if the author didn't post it. The main issue with this extension is it has to be updated very regularly. To continue to bypass the paywall requires constant tweaks & edits. So, unless someone makes a fork & wants to take over the project, mirrors aren't particularly helpful.

The author did post it. I'm pretty sure the bpc_updates repo was managed by magnolia, and that's where it was linked

They didn't post it on their Twitter, which as far as I know, is their only other social media presence.

https://x.com/Magnolia1234B

... Magnolia1234B/status/1823073077867287028

You're right! It hasn't been updated in a month, but maybe that's where the updates will be coming.

It's typically updated every couple days to once a week.

https://x.com/Magnolia1234B/status/1823073077867287028

So what is/was the "official" repo/source then? Because the repo at Github [1] points to the Russion repo [2], see here [3].

[1] https://github.com/bpc-clone/bypass-paywalls-firefox-clean

[2] https://gitflic.ru/project/magnolia1234/bypass-paywalls-clea...

[3] https://github.com/bpc-clone/bypass-paywalls-firefox-clean?t...

The Gitflic link you posted is broken. Here is the working one (looks like it was updated recently, but my browser doesn't want to translate)

https://gitflic.ru/project/magnolia1234/bypass-paywalls-chro...

I wasn't aware of the bpc-clone, but it's working and was updated 4 hours ago. I would imagine Github would catch onto that one rather quickly. The parent folder is private.

Host it on gitee dot com to avoid takedown nonsense. I sometimes mirror repos on there.

There is also radicle dot xyz.

I was just wondering if radicle would be a good answer to this problem. I haven't tried it yet, but read about it with great interest and couple months ago.

According to their FAQ, they are still in beta & haven't launched yet.

https://radicle.xyz/faq

The publishers are fully aware their website paywalls are bypassable in order for the content to be indexed by search engines. Why are they not held responsible for this choice? If they truely wanted a paywall they could implement one but they want their cake and eat it too.

thanks to our glorious intellectual property laws protecting the rich and punishing the poor.

radicle + fleek for IPFS hosting

Reply on news.ycombinator.com