Hacker News Reader

Comment by WA | Hacker News Reader

And what do you think can be done somewhat cheaply on every blog out there that has a comment function and allows someone to upload excerpts of Harry Potter?

The bare minimum technical solution seems to be excessive. You need some kind of "Shazam for all the things", which can identify copyrighted material by having a small snippet. And for that you actually need a huge database with fingerprints or whatever of copyrighted material.

Please let us know if you have a simple and cheap solution that every blog owner can implement (without high costs).

View on HN

Let me state first that Article 13 is a very dangerous idea that will have bad consequences.

That said, we're making our Attribution engine (the shazam for all audio-visual content) free to all platforms and rights holders [0]. It's surely not a great situation, but we don't want to stay still and let the small platforms to crumble under the weight of the giants.

[0] https://docs.google.com/document/d/1CLybxCFg_gz4n62UqVr3XEsy...

Is this actually helpful? It seems like there are really two options here.

One is that you can have your own database of copyrighted works, let anyone submit one, and the law considers that sufficient. But then everyone has their own database that starts with nothing in it, none of them really ever have more than 1% of the total copyrighted works in them nor ever really catch anything, and uploading everything to everyone's database is completely intractable so nobody bothers.

The other is that having an incomplete database is insufficient, but having a complete database is still intractable because anyone else can at any time create a new copyrighted work that isn't in it. And then troll you with it because they know it isn't in any database but can have someone upload it to your site just so they can use the law to destroy you. No database will ever contain more than 1% of the total copyrighted works, because they're created by the millions every minute of every day.

So does pooling together like that actually buy you something? It doesn't seem to change the "are you screwed" calculus much one way or the other.

The way we read the article 13 is that platforms have to offer tools to identify copyrighted work, but also the rights holders have to register the work for them to be identified.

That would imply that one database with an obvious way for rights holders to register their content should be the right way to go.

But we will not know up until an EU country actually writes the law.

> That would imply that one database with an obvious way for rights holders to register their content should be the right way to go.

Having a single database seems unusually problematic. If there is only one operator, who is that supposed to be? What happens if/when they give up or go out of business? What happens when they charge monopoly rents or their system is unreliable or full of bugs?

On the other hand, if there is one submission process that submits works to everybody's database and people have their own copies, that seems like it would be of particular interest to pirates -- go sign up as a database recipient and the copyright holders will send you copies of all their works.

It's as if no one has even attempted to think through how this would work.

Limit the comment to 1000 characters.

Any blog or forum needs administration, as part of which off-topic or copyright-infringing messages should be deleted. "Stay on topic and post only content that is your original work".

What is being targetted here are websites that rely on infringement for their popularity and traffic and ad revenue.

Any blog or forum needs administration

I'm not sure why are you stating this like some sort of undisputable truth.

Article 13 will not let Amazon sell books that were pirated (ie. re-uploaded) and claim everything is fine. They do didn't wanted to fix it so now they will be forced to.

For Joe blogger there is proportionality (https://juliareda.eu/wp-content/uploads/2019/02/Art_13_unoff... - 4a). And if Joe blogger decides to open comment section then he should read it every so often.

If not then this Harry Potter excerpt would be hard to find among spam overflowing comment section.

So, merely deleting a comment is enough? But that’s what we had before. Aren’t you required to have some kind of upload filter so that copyrighted material never ends up in your comments in the first place?

You are ignoring "proportionality" part and "not liable if" part. Yes EU regulators wouldn't be chasing small blogs just like they do not do it with GDPR regulations.

Big boys like Amazon wouldn't be able to get away with this shit (ends up on HN regularly):

https://www.extremetech.com/internet/268109-an-update-on-my-... https://www.forbes.com/sites/wadeshepard/2017/01/02/amazon-s...

Now you tell me how do you make Amazon and similar not to profit off scam? Ask them nicely? Doesn't work.

In debates like this, I tend to mention that UK law has prison time for commercial copyright infringement.

Not being a lawyer, I have no idea how hard it is for a prosecutor to get that outcome.

(So far as opinion, I would like copyright duration reduced to 7 years from publication, but suspect it is easier to make a human-level-creative AI then to change the law).

The Amazon example is not exactly related to the current issue and copyright violation already exists in law.

> Now you tell me how do you make Amazon and similar not to profit off scam?

The fact that they're selling a plagiarized book is grounds for a lawsuit from the author.

How would you automatically distinguish between copyrighted material and parody/satire/fair use?

Google seems to have had such a hard time doing so that it has the appearance that they've given up

For the GDPR, only the regulator can sue. Everyone else can only complain to the regulator. This minimizes the number of possible suits.

For article 13, everybody can sue. And copyright already has lows like grannies without internet ending up in a court for alleged file sharing.

the 2 laws are not even remotely comparable.

>So, merely deleting a comment is enough? But that’s what we had before.

This, and the successfulness of GDPR, is why I'm so doubtful over the problems presented about Article 13.

Reply on news.ycombinator.com