Hacker News Reader

Comment by jrochkind1 | Hacker News Reader

That could be, cite to the law or description of it? I'm interested in learning more. What I see in the law (https://www.law.cornell.edu/uscode/text/17/512) is that a "notification of claimed infringement" must include "Identification of the copyrighted work claimed to have been infringed, or, if multiple copyrighted works at a single online site are covered by a single notification, a representative list of such works at that site."

Github's own guidelines to DMCA takedown notices say "Identify the copyrighted work you believe has been infringed." https://help.github.com/articles/guide-to-submitting-a-dmca-...

If the notice did not identify a work believed to have been infringed, then why did Github respond to it? Why don't they follow their own process? If the notice did identify a work being infringed, then they are probably wrong because you can't infringe a copyrighted work by listing a URL in a list. I guess actually getting damages might require proving they _knew_ that, and be generally infeasible/cost prohibitive.

There are some problems with the DMCA regime.

View on HN

Part of the challenge here is that "the DMCA" is a number of clauses. People who are "against the DMCA" often turn out to have an incoherent position, at least from a legal standpoint, when it turns out they just disagree with one of the contentious clauses.

This would in fact be the most contentious clause, the one about being able to take down tools that enable circumvention, the one that is historically the one that perturbs techies and HN-types the most. I think what we see here isn't so much a DMCA takedown of a single line, but a single line modification in an attempt to prevent someone trying to take down the entire ad blocker, by making it so this particular person doesn't have any standing (in the legal sense) to make claims against the ad blockers anymore.

The copyright takedown clause would be number two, but it has a mitigating factor; the DMCA copyright takedown process that you might see on a hosting site or HN itself [1] has a positive element as well, which is that by conforming to the DMCA a site like HN is able to host user content like our comments while discharging from themselves the responsibility of having to pre-filter every comment for copyrighted content. This clause has certainly been abused, and there is a justifiable case that the Feds have not been adequately aggressive about chasing them down, but on the net I still approve of this clause, personally.

(You also have to distinguish between "the DMCA" and a site's policy, which may go above and beyond. Many or most of the things that people complain about for YouTube, for instance, are their own elaborations on the theme, not the legal requirements themselves. Not all of them, though; YouTube tends to favor the big media companies very strongly when it comes to defining "fair use". But things like taking away your monetization and giving it to somebody else is a YouTube policy, not the DMCA. Or at the very least, it's a penumbric emanation of the DMCA and not the DMCA itself.)

Were I the developers or anyone with any ownership in this software, I would hesitate on putting too much stock in the idea that this was an improper use of the DMCA claim process. It was. But the reward for aggressively pushing back on that may be a proper lawsuit for violation of the anti-circumvention clauses, for which there is not a notification process but simply a legal basis for lawsuits granted, IIRC. Your reward for armchair-lawyering this DMCA takedown request could be a true lawsuit.

[1]: https://news.ycombinator.com/dmca.html - have a look at the footer of this page

The DMCA does make distributing circumention measures illegal, but I can't find anything in the law, or discussions of it, saying that the takedown process applies to circumvention measures. The takedown process says an ISP like Github is not liable for copyright infringement if they respond to takedown notices, which is what creates the takedown process. It doesn't say anything about circumvention measures and liability with regard to notices though.

What the DMCA says about circumvention measures:

> (2) No person shall manufacture, import, offer to the public, provide, or otherwise traffic in any technology, product, service, device, component, or part thereof, that—

> (A) is primarily designed or produced for the purpose of circumventing a technological measure that effectively controls access to a work protected under this title;

> (B) has only limited commercially significant purpose or use other than to circumvent a technological measure that effectively controls access to a work protected under this title; or

> (C) is marketed by that person or another acting in concert with that person with that person’s knowledge for use in circumventing a technological measure that effectively controls access to a work protected under this title.

It seems obvious to me that Easylist is not such, I don't think it meets A, B, or C. But it'd be an expensive court process and who knows what the court would end up doing.

In this example, it seems like Easylist has no desire to include that URL anyway, as it is not an ad server.

It may be true that Github can decide to refuse to host the thing anyway, but it's not a DCMA takedown notice process.

The whole situation is indeed a mess.

As I added after you replied, I agree that this is an improper use of the takedown procedures. However, your reward for successfully armchair-lawyering that may be a full-on Federal lawsuit.

Also, I agree that Easylist itself may still not be a lawsuit target. It simply makes a claim about a certain domain, it doesn't do any access circumvention itself based on that claim. But if Easylist isn't, the ad blockers using it certainly would be. And that alone would change the dynamics of the situation quite a bit.

I hate to say it because I like adblockers too, but it is frankly very likely that when the advertising industry finally makes the push against them that by current law, it will indeed turn out that it is illegal to use ad blockers on sites that take active measures to ensure you view ads [1]. Again, don't mistake me saying this for endorsing it, but I think it's a very plain and obvious reading of not just the DMCA, but even something as fundamental to our legal system as common-law contracts... if a website wishes to make viewing their content conditional on viewing an ad, they can do that, just as they can make viewing their content conditional on paying them money, joining a club, or anything else that qualifies as "consideration" [2]. They can also place further restrictions on that content as part of that content. Any argument in favor of ad blocking that would also explain why either Netflix customers have the legal right to retain copies of the movies, or in the most extreme cases, explain why everybody has the legal right to retain copies of Netflix movies, should be discarded as an argument that proves too much [3].

I'd suggest the ad blocking community and the interested tech community at least wargame out the plan for if they lose the legal cases, because I would personally put that at somewhere around 90% probability if any of them ever go to court. Even with the EFF supporting it, I fear the EFF would pretty much be reduced to making very handwavy arguments about fundamental rights and basically pounding on the table, because in my considered opinion they really won't have much else. And even if they are correct, courts tend not to take much account of those arguments.

[1]: If the judgment goes really bad, it could even be illegal to bypass sites that don't try anything explicit. However there is a very good argument here that there is a history on the internet of assuming more rights rather than fewer if you don't assert yourself, such as the fact that browsers generally render things differently anyhow, the long history of search engines, the needs for accessibility software to render pages fundamentally differently anyhow, etc. I think there's a good chance no judge would want to overturn that consensus as it's now around 25 years old.

[2]: https://en.wikipedia.org/wiki/Consideration

[3]: http://slatestarcodex.com/2013/04/13/proving-too-much/

I understand I'm arm chair lawyering, but the purpose of a user agent is to display what the user wants it to, not what the site wants it to. I can't be sued for not watching commercials.

Actively circumventing access restrictions is another issue entirely, but if I'm sent data, there is no reason or guarantee it must be displayed as intended. What about blind folk, or those who don't run JavaScript? Is turning off just now illegal? What about not downloading images?

If a site can't make me pay, and doesn't want it's content to be viewed unless I pay, simply don't send me the content.

Again, circumventing access control, even terrible ones, is one thing, but if you send me the data, then you sent me the data, what the issue?

> If a site can't make me pay, and doesn't want it's content to be viewed unless I pay, simply don't send me the content.

I'm fully on your side here, but to play devil's advocate, I think it's fair to consider an analogy like "if a restaurant doesn't want its food to be consumed unless I pay, simply don't serve me the food."

You ask a server (of the web or the hospitality variety) to serve you the usual. The server gives it to you and reminds you that the deal hasn't been fully executed yet: you're to next [ask the cashier by the door to ring you up || ask the ad server to serve you an ad], and then [pay when asked || render the ad amongst the rest of the content]. Sure, you could forego talking to the [cashier || ad server] instead.

Again, I don't like that one bit, but I think it's the kind of "reasonableness" that holds up in court. IANAL.

So, they serve you the food, and you pay for it.

But, you choose not to eat the raw onions they've served with the meal.

This would be the analogy of choosing not to render html/javascript content.

At the risk of destroying my efforts at being reasonable, suppose it's an "if you eat the whole thing, it's free" situation, except there is no "$19.99 if you can't finish" else clause. You know that they didn't offer an else, yet you eat anyway, and you slip your onions into the plant in the corner.

Then they can make other ridiculous rules too, for example "you should pay 100x the price in menu unless you can stand on your head for an hour". That would probably increase the income dramatically.

But there is no expectation that a restaurant will serve you without paying. There is an expectation that a website will ask for payment/authorization if required, otherwise I'm not required to pay.

Moreover, there is no way to know if the content you're requesting will require a transaction (unlike a restaurant where the prevailing expectation is payment for service, even if prices are left off the menu). It has always been the case that I need to request the resource and then be told if it costs money, otherwise it's given to me.

Likewise, there has never been, and I would argue can't be, an expectation that a user agent render all content as expected. Would custom style-sheets violate the law? Do Lynx, Links, Links2, w3m, mutt, and pine all of a sudden become illegal? How does a screen reader render an ad? How does a braille interface render an ad? Am I now legally required to run a graphical interface otherwise I'm playing legal roulette?

What happens if the adserver malfunctions and doesn't send me an ad? Am I now put in a legally bad spot? What if an ad is sent in swf and I don't have flash installed? I also feel like there are legal implications to forcing someone to execute code sent to them. Do ad servers all of a sudden become responsible for drive-by malware? Can we sue them for damages?

I feel that the crux is that there is no way to know if "payment" is required before requesting a resource. You can't send me something and then say, "oh, yeah, hey, you need to pay me for that" when the (vast) majority of the time I'm sent things without any expectation of payment.

Perhaps the better analogy, then, is that it's kind of like an unattended farm stand with an honesty box. Except the honesty box isn't visible on the way in, it's located on the back of the enter sign so you only see it on the way out, and the driveway is so long that surely you've already started munching on the fruit while starting to leave. Since I'm still on a food kick the whole expectation-of-paying thing is still clouding the analogy a bit, but at least this is closer to the situation than a sit-down restaurant analogy. Oh, and of course our hypothetical farmer needs to have primarily fixed costs, little or no variable costs based on the amount of fruit taken.

It may very well be more common for such hypothetical farmers to forego having an honesty box hiding where you don't see it until you've consumed the fruit, but for those who do choose to have one, are you stealing the fruit if you don't drop in a few bucks?

More on topic: I also think courts would see quite a difference in intent between using a mainstream graphical browser with an ad blocker vs using things like a text mode interface, a screen reader, a braille display, or libcurl. The former is like driving past the honesty box while chuckling; the latter is like not even knowing it was there.

> Perhaps the better analogy, then, is that it's kind of like an unattended farm stand with an honesty box

The issues is that there is normally a expectation of paying for things like produce. Unless there was explicitly a "Free Produce" sign, I would expect to have to pay.

There is no expectation that you need to pay for the data sent to you later; if payment is required for access, you're told so and need to provide it to continue to the resource (or otherwise provide proof that you had paid, e.g. logging in).

There hasn't historically been and can't be an expectation of payment later because that would be untrue for many, if not the vast majority, of websites. Additionally, there has never been an expectation that the client will render everything you send to them. All browsers have the option to disable JS and images, and always have.

It's these differences in expectation and culture that I believe provide the difference between your examples and the web. Violating these constraints would cause legal issues in the vast majority of systems, would mean running old software would be illegal (Chrome preloads links under certain circumstances, but doesn't render them), and would also end up forcing users to run code they didn't choose to run (there is no expectation or knowledge of what code the server will send and choosing to not run harmful code would be illegal), which would be an interesting thought experiment as a civil rights violation. It would also force me to, say, accept a EULA for Flash, even if I disagree with it because I visited a site that randomly sent me a flash payload. Or what about something without a linux runtime; I would have no ability to avoid committing a crime, because I don't have the choice to accept the rest of the content that came with the content I can't run, but am legally required to run.

Violating the very assumptions of how the web works would have terrible ramifications.

> More on topic: I also think courts would see quite a difference in intent between using a mainstream graphical browser with an ad blocker vs using things like a text mode interface, a screen reader, a braille display, or libcurl.

Why? They're all the same thing. I'm deciding how I want to consume content; I'm not circumventing a access control mechanism.

Because what a court does isn't always logical or even reasonable. Especially on IP stuff, especially on IP stuff involving software. Did you pay attention to Oracle v. Google?

So far in the Oracle v Google case it's been ruled that apis are covered under copyright, but that implementing them from scratch is covered by fair use. I mean, that's not wholly unreasonable.

Oracle has appealed, would we'll see what happens.

I seem to have hit the max reply depth, so sibling posts will have to do.

> disable JS and images

> They're all the same thing. I'm deciding how I want to consume content; I'm not circumventing a access control mechanism.

I guess the difference I am trying to highlight is:

* a lack of ads for hard technical and/or compatibility reasons (JS not enabled, images not enabled, graphics subsystems not existing, not having sufficient eyesight, etc.), versus

* a lack of ads because screw you.

This line in the sand may be stupid, but I'm afraid it's not "the same thing." I'm afraid this difference could be argued successfully in court, and that is my point.

Hn removes the reply for an increasing amount of time per level, I think.

But it's not "screw you" it's "I don't want to waste the bandwidth I pay for and am metered on with things I don't want to download and could potentially harm my computer". Viewing those ads costs me money as well, money which isn't going to the person serving the ads, not to mention the risk of malware.

"I understand I'm arm chair lawyering, but the purpose of a user agent is to display what the user wants it to, not what the site wants it to."

This is an assertion that is commonly made on the internet, but I see no reason to believe it carries any legal force, or even necessarily any moral force. In fact it's not that hard to read it as an argument made solely to come to the desired predetermined conclusion rather than any sort of principled argument. It implies that the sender loses all rights to anything they send to you, which is definitely legally untrue; I gave examples above already.

Also, if you win on this point, you will not experience a glorious utopia in which ad blocking is OK and you can save whatever streams you want and so on... you'll experience a world in which all this content gets removed from the web and locked behind even more proprietary clients that will come with what the publishers want. What may seem to you to be a simple bugbite back in favor of what you believe your rights to be may cause a much larger allergic response than you'd anticipate.

"I can't be sued for not watching commercials."

You haven't signed a contract saying you will. That may not be the case online.

The questions about whether such contracts should be something that even can be offered, or whether simply clicking through a EULA or accessing a bit of content can bind one to a contract, or the nature of what such a contract may be allowed to be, are all separate matters of interesting discussion. However I don't foresee any world arising in which the "the purpose of a user agent is to display me what I want to see and therefore any manipulation of the content other people own the rights to is within my rights" is going to hold up. There's too many rights and rights-holders that won't stand for it, and even if you did somehow win that case, they'll simply retreat and retrench in whatever it takes to recover those rights for themselves. If you rewrite the terms of the contract, you have to account for the other side of the contract reacting to it, not just passively sitting back and going "Oh, gosh, I guess I'm stuck then, I'll just keep doing what I'm doing without changing anything."

> The questions about whether such contracts should be something that even can be offered, or whether simply clicking through a EULA or accessing a bit of content can bind one to a contract, or the nature of what such a contract may be allowed to be, are all separate matters of interesting discussion.

But you can't accept a contract just by visiting the site. Especially since the in the same action as becoming aware of the exist of the contract also makes you breach the contract.

> However I don't foresee any world arising in which the "the purpose of a user agent is to display me what I want to see and therefore any manipulation of the content other people own the rights to is within my rights" is going to hold up.

Why? This has always been the purpose of the user agent and it's difficult to impossible to actually make sure things will always look the same in all browsers. Could viewing a site in FireFox or Edge become illegal? Again, how would I know that _before_ taking the action. What about systems such as links2, w3m, elinks, and lynx?

> f you rewrite the terms of the contract, you have to account for the other side of the contract reacting to it, not just passively sitting back and going "Oh, gosh, I guess I'm stuck then, I'll just keep doing what I'm doing without changing anything."

Which terms? The UA has always been the agent of the user, not the site whose content is being displayed.

I just find it very difficult to believe that the court will accept that I've broken a "contract" I can't know exists without breaking it.

What you are describing applies only to US where copyright laws are biased towards the interests of publishers. Developing adblockers in other countries might be the solution I think. It is unbelievable that a publisher might decide what I do with content on my computer. No, he cannot or at least should not be able to decide. If he doesn't want me to block the ads then he should not serve the pages to me in the first place.

> However, your reward for successfully armchair-lawyering that may be a full-on Federal lawsuit.

I assume Github has a whole bunch of non-armchair lawyers.

But this is indeed the problem with the whole system, it comes down to who can pay the legal bills.

It seems obvious to me that Easylist is neither "primarily designed or produced for the purpose of circumventing a technological measure that effectively controls access", "has only limited commercially significant purpose or use other than to circumvent", nor "is marketed.. for use in circumventing a technological measure that effectively controls access."

But it could take a whole lot of legal fees to determine that in court, and as we saw in Oracle v Google, the courts don't always decide what seems obvious to us. For better or worse, any sane person or entity wants to stay out of court regardless of whether their lawyers think they have a great case. Unless they have a whole lot of money to burn.

At the least, I think Github should make it's policies clear about what it's doing. If they say a DMCA takedown notice must "Identify the copyrighted work you believe has been infringed" (as the law indeed says), they should not take action to complaints that don't do this. If they want to respond to other types of complaints, they should say so, and explain how. (And ask their lawyers how it effects their liability under DMCA, if at all).

Github appears to be trying for transparency with their docs and practices on DMCA, which is great and important and greatly appreciated. This is one area where it could be improved. Responding to DMCA takedown notices that are not in fact DMCA takedown notices and do not follow Github's own published instructions/requirements for DMCA takedown notices (cause they aren't DMCA takedown notices)... is not transparency. The DMCA regime has plusses and minuses; mis-educating people about the DMCA law doesn't help us evaluate what these may be in order to be engaged citizens.

> penumbric emanation

Vocabulary much? I'll bet there's a blip on Google Trends today.

It's a legal term: http://legal-dictionary.thefreedictionary.com/penumbras

Reply on news.ycombinator.com