Hacker News new | past | comments | ask | show | jobs | submit login

The title is misleading. This isn't some generic torrent tracker. The repo in question is the source of a specific site - nyaa.si [0] which according to the complaint is preconfigured to share copyrighted material:

> The identified files and code are preconfigured to find and provide infringing copies of our Members’ film and tv content to Nyaa.si users in violation of copyright law.

This is probably what makes this complaint possible.

[0]: https://nyaa.si/




> This is probably what makes this complaint possible.

What makes this complaint possible is that the MPA blatantly expands the scope of copyright, counting on their large legal budget and the skewed possible consequences to discourage people from fighting back.

The MPA doesn't care whether this repo actually infringes on their copyright, nor whether the DMCA allows them to issue a takedown request. They only care that this will hurt people who oppose their interests and that there is little practical recourse.


Please don't, however, baselessly propagate their claims. As far as I can tell it's bullshit. The source code may be for nyaa.si but it just contains normal torrent tracker code.

Take a look at the files listed:

> https://github.com/nyaadevs/nyaa/blob/master/utils/api_info....

Online here: https://gitlab.com/SIGBUS/nyaa/-/blob/master/utils/api_info....

See anything remotely suggestive? Not I.

> https://github.com/nyaadevs/nyaa/blob/master/utils/api_uploa...

Nothing interesting here either.

> https://github.com/nyaadevs/nyaa/blob/master/nyaa/static/sea...

Why did they pick the search entry XML for the porn section??? (I can only assume their lawyers have no idea what 'sukebei' means.) Anyway, nothing interesting in this file either.

> https://github.com/nyaadevs/nyaa/blob/master/nyaa/api_handle...

Again, looks ENTIRELY generic.

> https://github.com/nyaadevs/nyaa/blob/master/.docker/nyaa-co...

It's a config file with some credentials in it. Again, nothing.

> https://github.com/nyaadevs/nyaa/blob/master/nyaa/torrents.p...

More generic looking Flask code.

> https://github.com/nyaadevs/nyaa/blob/master/nyaa/templates/...

Again... it's a very basic looking HTML template.

> https://github.com/nyaadevs/nyaa/blob/master/migrations/vers...

Just about the least interesting database migration ever.

> https://github.com/nyaadevs/nyaa/blob/master/config.example....

I'm not even sure why they would pick this file, seriously. At this point they're just making shit up.

> https://github.com/nyaadevs/nyaa/blob/master/.docker/es_sync...

And another random config file.

If there's even a smidgen of actual copyright infringement in this repo, they failed to find it.


They DCMA doesn't say that the repo was used to host copyrighted material:

>The identified files and code are preconfigured to find and provide infringing copies of our Members’ film and tv content to Nyaa.si users in violation of copyright law.

They are basically saying "this is the codebase for nyaa.si, and as proof here are a bunch of files that prove that assertion". What I don't understand is the legal grounds they have to compel GitHub to remove the repository. It would be like if I removed ffmpeg from github because freestreams.tv was using it to power their backend.


I understand that, but those files do not show that the repo is preconfigured to host copyrighted files. I don't get what the search XML or database migration is supposed to prove for example.

All they have demonstrated is that it is the source code for nyaa.si. I don't get how that is enough to do anything.


"preconfigured to find and provide infringing copies" goes way beyond a mere claim of being the codebase.


I'd say that first one (api_info) is semi-plausible in terms of being "configured" for piracy, in that it hard codes the production API of the site, and I assume the data accessible therein is largely piracy.

It is also true that one would need to modify quite a few places to make this codebase sensibly usable for anything non-piracy. Even if an Anime studio decided to make their back catalog available over torrent (for some reason), this site really is not designed for their needs.

However, I strongly suspect this list of files is roughly the set that have 'nyaa' as a stand alone word (i.e. not part of a snake case word or other compound). I suspect they loaded the code into a search engine, and searched for "nyaa.si", and flagged all results. Many search engines will treat the '.' like a space, will filter out "si" as being too short, and thus search for "nyaa", and only find results for that when it exists as a separate word.


> See anything remotely suggestive?

Maybe. I see:

    NYAA_HOST = 'https://nyaa.si'
    SUKEBEI_HOST = 'https://sukebei.nyaa.si'
so this is pre-configured to get torrents from a website which offers mainly/mostly/only infringing torrents.

Caveat: IANAL.


how about trackers.txt ?


See, I still think that's innocuous since it does not list any particular infringing links. For example, Google has no issues showing us opentrackr.org.

And yet, it is completely absent from the DMCA takedown, which signals to me that they really did no work at all before making this takedown.


a link to a list of trackers it no more illegal than publishing the public street address of a business that sells products that are illegal in certain jurisdictions (texas dildo law, anyone?)


You've got a point, https://news.ycombinator.com/item?id=25836800 also points this out.

My eye was caught by

> The Project blatantly infringes the MPA Member Studios’ copyrights and countless other copyrights. Indeed, copyright infringement is so prevalent within the Project that infringement plainly is its predominant use and purpose.

when submitting, and you could argue that does constitute a blanket attack against dual-use products, but the specific allegations you cite are also relevant. Since I could still edit the title I changed it to "for source code of torrent tracker nyaa.si"; hope that's clearer.


Possibly due the inclusion of the tracker URLs in the code? https://web.archive.org/web/20200917190718/https://github.co...




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: