Hacker News Reader

Comment by brian_cloutier | Hacker News Reader

> All DRM is a non-solution really.

I know that by now it's tradition to say this, but can we please stop?

It seems to be that a law of life is 'everything dies'. Is medicine therefore useless? Are all the people who spend money for nothing but delaying the inevitable getting ripped off?

We all admit perfect security is impossible. Yes, you're right! Controlling the spread of information is a very hard problem. Yes, you're right! It would be very nice if the things I want to watch and listen to were made freely available by their creators.

But no, you're wrong. DRM has a purpose, and it can be successful even without perfectly achieving its goals. People tend to avoid effort, and if you can make pirating content more difficult than obtaining it legitimately, most people will obtain it legitimately.

There is no fundamental reason why information deserves to be free, it's just easier to copy than physical things.

View on HN

DRM has a purpose, and it can be successful even without perfectly achieving its goals. People tend to avoid effort, and if you can make pirating content more difficult than obtaining it legitimately, most people will obtain it legitimately.

I haven't yet met a DRM scheme that doesn't achieve this purpose backwards(1). What it really rests on is that people don't know they could just get it easier on "allmyvideos.net".

There is no fundamental reason why information deserves to be free, it's just easier to copy than physical things.

Its not just easier. A copy has zero marginal cost. That makes it special and different than anything that came before. DRM seems to just be a monkey-patch to try to get that marginal cost to be non-zero. I'm not saying that it deserves to be free. I'm not even saying that I think it should be. I'm saying that it IS free. We don't want it to be, because our economic models don't support it (yet), but that's its natural state.

I stand by "DRM is silly". If that money were spent providing a better, easier product, they'd make more than they do now with DRM. It seems like they're perfectly willing to spend $3 to keep from losing $1 to piracy. It feels like a kind of willful ignorance.

No data, of course, just a strong opinion, held loosely.

(1) It doesn't take long outside the US to find out that there are plenty of things DRM makes impossible to aquire legitimately.

> DRM seems to just be a monkey-patch to try to get that marginal cost to be non-zero.

Yes, that's absolutely what it is. As a society, we seem to believe that information can be owned. We get angry at each other for 'stealing jokes'. Fanboys around the world are up in arms whenever another platform copies some feature from their beloved. Plagiarism is an accusation which can ruin a career. We have always felt entitled to do what we please with the fruit of our labors, and we feel no differently when that fruit is information.

We have all had bad experiences with DRM, but we've all had bad experiences with technology in general. Especially here on HN, why does the conclusion have to be "DRM is intrusive and impossible to do right anyway" instead of "What an opportunity for a non-intrusive alternative"?

I'm not saying the second option is right, but why can't it be?

  > Plagiarism is an accusation which can ruin a career.

There's a difference between ownership of an idea, and plagiarism. An example of trying to 'own' an idea would be:

  1) Person A creates an idea
  2) Person B independently creates the same idea (or
     something extremely similar).
  3) Person A attempts to assert control over the idea
     by dictating what Person B can or cannot do with it
     based on the fact that Person A 'got there first.'

An example of plagiarism would be:

  1) Person A creates an idea.
  2) Person B copies Person A's idea, and attempts to claim
     independent creation (or attempts to claim creation prior
     to Person A).

The real difference is that with plagiarism, Person B is committing a fraud about the source of the idea. It's more about authorship than anything else. Even in a society where you couldn't own an idea, you could still have authorship as the first person to think of something.

> Yes, that's absolutely what it is. As a society, we seem to believe that information can be owned

Two wrongs (information cannot be "owned" in the same sense of a physical thing, regardless of how much one wants to be believe it; and that DRM can be effective) do not make one right. Note that I'm not discussing values here - just pure technical issues.

Here's the two problems of DRM in a nutshell, that make it into snake oil.

1. The "owner" of the content wants to make said content available to person X, but not to person X's recording device, which is indistinguishable from X.

2. It is enough for said content to be freed from DRM once, to become universally free of DRM.

No matter how smart your protocols, cryptography etc is, because (1) if you can display it on a screen, and take a picture of said screen, with a high resolution capture device, you've defeated DRM. And then because of (2) the DRM scheme, regardless of its other merits, becomes ineffective.

That's why DRM cannot be done right, even if you assume it's the right thing to do. (which I don't)

There is a very important difference between "owning" information - which is impossible and any attempt to do so is morally reprehensible - and the "creator's right" of being named as said creator - which is perfectly fine. All your examples are more or less instances of the latter - it's not okay, for example, to plagiarize because you are violating everyone's right to truth.

"We have always felt entitled to do what we please with the fruit of our labors, and we feel no differently when that fruit is information."

Except that fruit of your labor is not "information", it's a "copy of information". And of course you are entitled to do whatever you want with it. As am I, if I have gained access to it by any means that did not violate your privacy (eg., breaking into your house or hacking into your computer) or other fundamental rights (copy"right" is not a right, it's a privilege - very, very important difference, as rights cannot be granted by law). In other words, I cannot force you to share something with me. But neither can you forbid me from sharing things I have - including pieces of information - unless they violate one of your fundamental rights (for example, your dignity).

This was really well articulated; thank you for writing it. I find piracy/DRM/copyright theft/etc difficult to wrap my head around, in the sense that I have trouble deciding where I stand on many of those issues. The "information deserves to be free" thing has always rung a bit hollow to me, but

  "Its not just easier. A copy has zero marginal cost. That makes it special and
  different than anything that came before. DRM seems to just be a monkey-patch
  to try to get that marginal cost to be non-zero. I'm not saying that it
  deserves to be free. I'm not even saying that I think it should be. I'm saying
  that it IS free. We don't want it to be, because our economic models don't
  support it (yet), but that's its natural state."

makes way more sense.

> I haven't yet met a DRM scheme that doesn't achieve this purpose backwards

Because most people think that this time around, DRM is going to work perfectly, when actually, you have as much expectation it will work perfectly as that cheap Master Lock from the hardware store.

> DRM seems to just be a monkey-patch to try to get that marginal cost to be non-zero.

This is indeed the right way to think about it.

> If that money were spent providing a better, easier product, they'd make more than they do now with DRM.

Or there's the example with Steam, where you have better and easier combined with just a little DRM applied intelligently.

I wouldn't go tooting Steams horn too loud. Their system is not nearly as wonderful as many consumers assume it is. Steam does not place ANY limitations on what DRM a publisher can use. They can use StarForce, they can come up with their own magic elixir, they can require you to send naked photos of yourself to the publisher for 'safekeeping' and Valve will permit it. Steam has 1 guiding principle: The Publisher is ALWAYS right. The publisher must be able to do anything they wish, without any restrictions whatsoever.

Contrast this to a system like Apple's iTunes which set up a market with strict rules the publishers had to follow to join. They HAD to sell songs individually. They HAD to sell them for 99 cents a track. They HAD to permit the user to burn them to CD. Etc, etc, etc. Steam does not have even one single "have to" when it comes to games publishers. It is a publishers wonderland.

If you have lightweight DRM on Steam games, you have the games publishers to thank for that, not Valve. Valve would not protect you from even the most extreme forms of DRM if a publisher wished to do it. Valve even goes so far as to spend hundreds of thousands of dollars (possibly millions but I doubt it) to have their developers create features which enable publishers to make it look like they're embracing digital distro when they're not. Case in point: the ability to download a game when its done... and then sit there and wait with it encrypted and complete on your hard drive, paid for... until physical retailers can get the game on their shelves. And users think this is a feature! It is a crippling of digital distro, which operates at light speed as opposed to the slow-as-molasses speed of physical distribution. Valve didn't have to do this. They could have said 'when you upload it to our servers, the customers who preordered it get the content available immediately', but they didn't, they followed the Golden Rule of Steam - The Publisher is ALWAYS right.

In other words, Steam lets the market decide on the appropriate level of DRM, and otherwise nothing you wrote is materially different from what I said.

I'm saying that it [copied information] IS free.

When you say it like that, a light really went on for me. In an undergrad economics class (for some reason I can still remember the exact phrasing the professor used, though it's been almost 30 years): "a good is 'free' if, at a price of zero, supply is adequate to meet demand."

This completely explains the unique situation with regards to digital information. The (marginal) cost to provide a supply adequate to meet demand is zero.... meaning the market will drive the price to zero... digital information is free. As a content publisher, you can want it to not be that way, but that is not the reality.

But that definition isn't true of most digital content. Commercial music, movies and software are effectively subject to price discrimination: pirates pay zero, but legal customers pay nonzero and subsidize the pirates. If the price dropped to zero for everyone then supply would plummet as well.

> If the price dropped to zero for everyone then supply would plummet as well.

That's a myth, because people were writing novels and singing songs and doing theatrical plays long before media distribution and consumption. Even now many musicians earn much more money from live concerts.

Also, I pirate stuff, mostly because where I live I can't get that content at all when I want it. However I also go to the movie theaters nearby with my wife and pay something like $30 for 2 tickets and some popcorn, per movie.

I don't mind paying that price because I'll never have the same experience at home, no matter how awesome my equipment can get, because (1) the size of my monitor is upper-bounded by the size of my living-room walls and (2) watching movies at home is boring.

I don't think you have the evidence to back up that assertion. Certainly the world is changing, but people simply like making and sharing stuff. How much money have you paid for the entertainment you get from Hacker News?

Marginal cost = price only in perfectly competitive markets. For commodity content(e.g., daily news), then yes the expected long term price is zero. For content that does not have (perfect) substitutes (e.g., the latest Batman movie), the expected market price is not zero.

> A copy has zero marginal cost. That makes it special and different than anything that came before.

I disagree that this is news. Thought experiment: You steal a car from the local dealer, but you leave enough money behind to pay for all the materials, transportation and manpower that went into building this one car (the marginal cost). Would this be morally okay? Why, why not? If everyone does this, who will pay for R&D?

Exactly the same is happening with digital copies. You are taking something with a marginal cost of 0, but the producer has no way to pay for one-time costs. Distributing them onto the unit price is not a new monkey-patch at all.

This will be an interesting question as 3D printing advances.

> but that's its natural state.

"Natural" is always a great word to turn an intuition into a fact. ;) There are certainly many products that are sold at arbitrary prices that have little to do with the marginal cost, it didn't take computers to get there.

First of all everyone is not going to do this - they haven't in the past. A few points...

If you like the car and you tell others they will most likely buy it at full cost or a least with a profit still on it.

Servicing the car will most likely depend on the manufacturer's own part on which they make a profit.

In terms of sales there is one more on the road, which will contribute to the quarterly reports for the company causing the share price to rise.

If the dealer has had it on the forecourt for several months they actually want to get rid of it now and they will be bring the price down towards, or even below, what they paid for it. You may even end up paying more than they were going to sell it for.

This thought experiment is flawed. When you steal a car from the local dealer, he doesn't have the car any more.

The correct thought experiment is: having bought a car from Ford, you examine it carefully then purchase all the raw materials yourself and assemble an identical duplicate for your wife.

In doing so you save whatever markup Ford places above and beyond their marginal cost.

Asking whether this is morally okay is the true issue.

I agree that the local dealer should be kept out of it. I am not sure if it makes a difference that Ford wouldn't have the car anymore, they have the marginal cost. Asking for more than the marginal cost seems to be immoral to some.

But I agree that your experiment boils down to the problem I am pointing to, and it is better because it is a very real problem that we see every day (e.g. in China).

> I am not sure if it makes a difference that Ford wouldn't have the car anymore, they have the marginal cost. Asking for more than the marginal cost seems to be immoral to some.

It does, and the cases are not comparable. Leaving an empty space (plus marginal cost) where the car used to be requires the original owner to expend time and effort to replace it, and they have opportunity cost as well. None of that is true of the digital example. To be a fair comparison you'd have to leave an atom-for-atom identical replacement for the car (or more accurately, take an identical copy and leave the original) and I doubt as many people would judge that unethical.

When I use "natural", I don't mean it to express an opinion that I believe that is moral to copy that Disney movie just because I can(1). Natural in this case refers to the economic description of 'free'. An unfettered digital file has unlimited supply at no cost. Therefore supply will be sufficient to meet demand, even at a price point of zero. Things can be sold at arbitrary prices, even given away, but with physical things, the supply will not always be sufficient to meet the demand. A unit with a marginal cost that is given away is not free. This is why using cars or other physical goods produces leaky analogies; there can never be an unlimited supply of any physical good.

You are taking something with a marginal cost of 0, but the producer has no way to pay for one-time costs. Distributing them onto the unit price is not a new monkey-patch at all.

This is actually a very good observation. We've reached the point where the entire cost of the good is the one time production cost and we've discovered that we've got no good way to collect it. DRM is the best we've got right now, and its awful. You are more than correct in pointing out that 3D printing (and localized digital micro-manufacture in general) is about to make this problem acute.

As we stumble into our Star Trek future, we should be expending as much innovative energy as we can into finding a way to solve this issue.

(1) At this point in history, with the operation of the market as it currently stands, I lean towards no, it is not moral to bootleg that Disney flick.

> we've got no good way to collect it. DRM is the best we've got right now

While DRM is awful, I think we should keep in mind what advantages we (arguably?) enjoy from this monkey-patch (distributing one-time costs onto an arbitrary unit price). With Kickstarter and other one-time funding, there is no incentive to absolutely excel and make a huge profit from unit prices. There is also little risk in delivering a terrible product because one-time costs have already been covered.

> As we stumble into our Star Trek future, we should be expending as much innovative energy as we can into finding a way to solve this issue.

If this Star Trek future is anything like an utopia, then I don't see anything wrong with agreeing to keep the monkey-patch in place, even if it is not natural in the economic sense, or even enforcable. (Very much like privacy - it's not really enforcable, but I really hope that society starts to respect it anyway.)

"A copy has zero marginal cost. That makes it special and different than anything that came before"

I love the rest of your post, but this stood out. It's special and different from anything that came before hundreds of years ago. But copyright wasn't invent yesterday, and patents weren't invented yesterday either. They've been around for hundreds of years.

Sure they have, but the cost to make each copy of a book hasn't been zero until very, very recently. That's the special part.

True. But this area, and the arguments that go with is, isn't just about the cost to copy a book (copyright). It's also about patent law - ideas were just as easy to copy back then. It's about copying plays, where the cost of copying the "book" is trivial, compared to the cost of the production itself.

I'm no historical expert, but I would wager that looking through history, you'll find that this situation isn't as unique as most people think. People were debating copyright in the English Parliament during the 1800's, including debates on piracy which practically predict the current situation.

>I haven't met a DRM scheme that doesn't achieve this purpose backwards

Steam would be a good example. It is DRM in the sense that it prevents people from simply copying the game folder anywhere, and it is more convenient than pirating (at least most of the time) with the advantages of things like pre-loading most of the files before the release date, and offering fast servers from which to download your game.

Incidentally if you copy the game folder to a friend who has actually bought it, say you've all bought a game for a LAN party, but not all of you have good connections at home, that -does- work, which has proven super convenient for me on more than one occasion.

I did not know this, and it makes me love steam just a little bit more.

Okay, let's say a copy is (essentially) free. So what? There are plenty of things that people buy where marginal cost of production is << than the price.

Concretely, why is selling a DVD with a marginal cost of production of $0.50 for $15.99 ok, but selling a movie with a marginal cost of $0.001 for $15.99 "special and different from anything that came before"?

One could argue that the dvd is an attempt to tie the information to a physical object. The dvd has no value in itself.

Whenever I have bought an optical media, i have thrown it away as soon as I had the data on my harddrive.

Name one real-world DRM system that makes piracy look less attractive for content (movies, music, books).

The only successful shops I can think of, such as Amazon, Steam, etc., aren't successful due to DRM (at least, not from the piracy perspective). It's due to an easier shopping experience, with the DRM getting out of the way. I'm not sure you can even meaningfully call Amazon MP3's approach DRM: it's just watermarking.

Pirating content is trivial and easy. In order to compete, you have to make an as-nice experience. Buying a book on the Kindle store is _easier_ than torrenting the book (if only slightly) - DRM has nothing to do with this aspect. DRM doesn't even remotely stop the spread of content.

The only time I've heard DRM being remotely effective is really complicated systems on games, delaying a crack by a week or so, which apparently increases sales (at the cost of plenty of bad-will from paying customers) - although I'm not aware of any published studies that really investigate that and compare to the negative sides. This is the only time your analogy of medicine delaying death makes sense; one small use case, and only for specific programs, not general content.

But for content? DRM is a non-solution, really.

> Pirating content is trivial and easy. In order to compete, you have to make an as-nice experience.

Yes, but it's a two-part process. Bits are very easy to copy around, and providing an easy way of giving money will only get you so far. What happens when the alternative of pirating is just as easy, but doesn't require money? If you want to make a nicer experience, you will eventually have to make pirating harder.

> DRM doesn't even remotely stop the spread of content.

Of course it does! No, it doesn't stop you. You know what torrents are, you know which torrents are likely to be viruses and which are in formats your computer is capable of reading. Many people out there know none of that, but can easily find the itunes button on their ipads. And all of those people have no way of sharing the things they buy, thanks to DRM.

> Name one real-world DRM system that makes piracy look less attractive for content (movies, music, books).

Here's another: Netflix. All you do is give Netflix a little bit of money each month, and then you don't have to worry about starting the download a couple hours before you want to start watching. You don't have to worry about getting caught under the new six-strikes program. You can't easily copy what Netflix gives you thanks to good old DRM, but that doesn't really hurt you.

You can't easily copy what Netflix gives you thanks to good old DRM, but that doesn't really hurt you.

Unless you happen to run a non-mainstream OS. Or move to Mexico. In which case it hurts you all the way.

Yeah, but for all that running it on Linux is a pain-in-the-bum, Netflix is on iOS, Android, and most set-top TV boxes.

I have 8 devices in my house that can play Netflix and only 1 ubuntu machine that can't -- but only because I don't want to go through the headache of patching WINE.

People have gotten Netflix working on Ubuntu before.

> Many people out there know none of that, but can easily find the itunes button on their ipads.

This doesn't have anything to do with DRM, though. If DRM didn't exist, people would still easily find the iTunes button on their iPad, and would still have a hard time navigating torrents and file format issues. All DRM seems to do is slowly teach those who can't authorize that nth device to learn how to torrent.

Both PS3 and XBox 360 games are much harder the pirate than to buy. Also, depending on how far you stretch DRM, a pirated copy of World of Warcraft is fairly useless.

> People tend to avoid effort, and if you can make pirating content more difficult than obtaining it legitimately, most people will obtain it legitimately.

The problem with DRM is that with information, it doesn't matter if it's more difficult for any given user to pirate. It only takes one person to make a copy of it without DRM, and they can then share it, and now nothing can stop anyone from viewing it. At least, that's the case without horribly repressive measures that don't allow anyone to view their own media.

All the DRM does is make legitimate use harder. If you want to find a free copy, you can; but if you're a legitimate user, and want to make a backup of your DVD? Nope, sorry, people aren't allowed to sell tools to help you do that. Want to capture a few seconds to comment on? Nope, sorry, people can't legitimately distribute tools that allow that. Want to skip the stupid previews? Nope sorry, also illegal.

The problem with DRM is that if it doesn't work perfectly, all of the bad things that it's trying to prevent can still happen, and in the meantime, tons of legitimate uses are banned, and we start to produce technology that is oppressive and works against the interests of its users.

Perfect security is theoretically possible, what gets in the way is human error - AND it really matters what you're trying to protect and from whom.

Encryption deals with sending messages from A to B such that a third-party C can't intercept the message. In a DRM scheme B and C are the same, which is why DRM is flawed by design, because:

(1) no amount of patching can ever fix it

(2) it punishes legitimate customers, as pirated content is far easier to deal with, while not suffering from lock-in effects; the irony might be that DRM is helping the prevalence of piracy

(3) no matter how hard it gets to create the initial DRM-free copy, from then on all other copies are zero-cost - which means it only takes one dedicated individual to create a pirated copy and all people that pirate stuff can enjoy it ;-)

> DRM has a purpose

Of course it does. Its purpose is to save a dying business model that was made obsolete by technology ... 1000 years ago there were businesses selling ice to clients.

The equivalent of DRM would have been to restrict refrigerators from producing ice. Fortunately for us, we realized that selling refrigerators is a lucrative business too.

Perfect security is theoretically possible, what gets in the way is human error

I forget where I read the story (might have been Cryptonomicon), and I don't know if it's true, but I find it quite illustrative of the point.

During WWII the British where finding that some of their messages encrypted with one time pads where being cracked by the Germans. Since they where pretty sure that was mathematically impossible they where quite shocked and immediately launched a full investigation.

Eventually they found that the problem was the team generating the one time pads (basically a room full of people drawing bingo balls with letters on them at random) had started second-guessing the randomness of letters generated. They'd started to subconsciously avoid balls that would lead to what they thought where patterns in an attempt to make the otp more random. Of course this lead to less randomness and broke the security of the otp.

Reply on news.ycombinator.com