[kmeisthax]

This article is making the engineer's mistake regarding reasoning about copyright infringement and the law.

The courts do not care how the copy was made, they care about what markets the copying would allow someone to get into. "Cloud DVRs are OK but only if the kernel, filesystem, and hardware take great pains to ensure separate physical storage locations for and no compression on each customer-created copy" is absurd and no judge is going to go for that.

No, the courts aren't saying "if you waste a bunch of money on extra hard drives, you can infringe copyright", either. Their concern is providing a demarcation line between "things the customer has done with your service" and "things your service provides on it's own". Yes, this line is going to be fuzzy, but it's fuzziness has nothing to do with how the bits are stored. It has to do with the context of the markets in which works are ordinarily sold.

>I originally thought the strangeness of digital copyright outcomes reflected a lack of technical literacy in the courts. But for the most part, I find the Aereo discussion shows general digital competency, and an appropriate aesthetic disgust for the “identical bits are different” problem.

Remember how after the Napster lawsuit, everyone was parroting the thought-terminating cliche "the law needs to catch up to technology"? Yeah... no. In reality the law is almost always three steps ahead of technology, because the law is written in a programming language that executes what you intended to write, not what you actually wrote.

[bluecheese33]

Thanks for the read (post author here)!

As I mentioned, I am willing to admit that I came into this expecting to find technical illiteracy, and I didn't find much. I agree that the mindset to look for hacks and oversights in laws is a naieve engineer tendency.

> No, the courts aren't saying "if you waste a bunch of money on extra hard drives, you can infringe copyright", either.

I agree no court wants this, and I didn't intend to imply otherwise in the post. Regardless, as a result of these cases, this is the current state the DVR industry is in as I understand. Wasting money on storage does insulate you from infringement, and people do it to be safe.

> a programming language that executes what you intended to write, not what you actually wrote

This is a great analogy. It does clearly get more complex when the court is executing "what you would have intended had you known about the internet" though.

Edit: On,

> The courts do not care how the copy was made

In Cablevision, they did for two reasons:

1. To figure out whether buffering was copying, which is a very technical discussion. See the footnote on MAI Systems

2. To figure out WHO was making the copy, for the volition based infringement test

My point here is that it really does get into the technical weeds. I know your point was mostly to just dismiss the deduplication discussion, which is reasonable. If one of my posed problems made it to court, the court would probably just do the "right thing". However, since they haven't made it to court yet, companies don't necessarily want to be the first to gamble on it.

[zerocrates]

Engineer-types certainly do try to loophole their way around the law in a way that's not how the law actually works, but this article is not really engaging in that I don't think (for one, it calls out the practice).

Cablevision is absolutely by its terms hugely reliant on the nitty-gritty technical details of how the DVR service there worked. The court had to kind of wind itself up in knots to work around the existing MAI v. Peak precedent (copies to RAM are actionable infringing copies) and the fact that the service existed just to make these copies, so it gets very in the weeds on how things are stored and the amount of time things are in buffers and so on.

Ultimately, it's probably true the most important thing was that Cablevision were an established player in an existing, uncontroversial market and they were making an iteration on the already allowed and understood "time-shifting" recording systems. But other players in the market shouldn't really be faulted for taking the court at its word that the details actually mattered. Sometimes they really do!

MAI v. Peak, mentioned in the article and above in my comment is a great example of that: the 9th Circuit holds that a computer copying the OS into RAM is a "copy" for the Copyright Act, thanks to the statutory definition of a work being "fixed." Result: a repair technician violates copyright by turning on the computer because he doesn't have a license. This is kind of the polar opposite type of decision to, say, Aereo's case: it's actually quite disruptive but hinges more on the literal definitions in the law and things like the computer not being on already, thus requiring the "copy" to be made. Congress actually changed the Copyright Act to counter this decision but in an extremely specific way, so the general "stuff in RAM is fixed and therefore a copy" principle remains and comes up often.

Aereo lost because they would upset the applecart of retransmission fees, Locast similarly though for nominally quite different reasons. It can be quite difficult to tell in advance if you're going to get a "letter of the law" decision from the courts or something more results-oriented, even from the same court.

An often-unappreciated wrinkle is the more or less total dysfunction of Congress leading to court decisions taking on ever more importance. The courts themselves aren't blind to this, leading probably to more results-oriented decisionmaking than there might be otherwise.

[Asooka]

> In reality the law is almost always three steps ahead of technology, because the law is written in a programming language that executes what you intended to write

So why is it always interpreted in favour of what content distributors intend and never in favour of common sense or the public? Laws are a public contract and as such the public, including you, has the right to push for its interpretation one way with judges having the ultimate authority to resolve disagreements. Just because you look at things from a logical perspective doesn't make you unfit to argue for your rights, and claiming otherwise strikes me as a really self destructive mix of learned helplessness and impostor syndrome.

[laserlight]

> "Cloud DVRs are OK but only if the kernel, filesystem, and hardware take great pains to ensure separate physical storage locations for and no compression on each customer-created copy" is absurd and no judge is going to go for that.

Reminds me of Aereo, the company that rented remote antennas to customers so that they can stream over the Internet what their antennas captured. Supreme court decided that they were violating copyright law [0].

[0] https://arstechnica.com/tech-policy/2014/06/supreme-court-pu...

[Nasrudith]

To put it in other terms it didn't fall, it was slammed to the ground.

[laserlight]

I just noticed that Aereo was already part of the discussion.

[rektide]

It's possible that the courts/law are so absurd that they could be wrong. Rarely is that considered.

[jjoonathan]

In theory, that's what the legislature is for.

In practice, the process for this is completely broken, so we have a system where the judicial branch pretends to divinate intent from the tea-leaves of legislation, even when it's clear that there could not possibly have been any informed intent because major relevant details were simply not known at the time.

It's not great, but it's better than a system where the judicial branch just does anything it wants.

All this said, instant communication and computers almost certainly introduced better forms of judicial and legislative process that haven't been experimented with because of inertia. I wouldn't volunteer our system as the guinea pig, but I hope that somebody gets around to experimenting with this, because our system sucks hard in a bunch of ways that seem like they are probably fixable.