Proving the Creative Contribution: How can you show what it means to contribute sufficient creativity for a work to be protected by intellectual property? Panel featuring Douglas Cawley (McKool Smith), Stephen Coats (Amazon) and Pamela Samuselson (Berkeley), moderated by Jeanne Fromer (NYU)
Full video is available on YouTube.
- Good morning everyone and welcome to the first day of Proving IP. I am Michael Weinberg. I am the executive director here at the Engelberg Center. For those of you who are not as familiar with the Engelberg Center, we are the center here at NYU that focuses on law and policy and technology and innovation. Both here at NYU and within New York City and then more broadly beyond that. And so, we're really excited today to have this event's opportunity to bring so many different people together to talk about the practical sense of Proving IP. One of the things that we do here at the center is that when we're engaging with questions around innovation and law and policy, we do it with events, we do it with convenings, we do it with publications and all of them are designed to make sure that we have a number of different parties who can come together and explain what they're seeing and what challenges they're having. So when we're doing that research and we're trying to find solutions, those solutions are informed by the on-the-ground reality. And that's very much what we're seeing here today in proving IP. This is an opportunity today and tomorrow to bring together academia and practitioners, judges and say we all work on different aspects of IP, but today we wanna focus on the questions of what does it mean in practice to prove different parts of IP and how do you think about what it means to litigate one of these cases? And what do you actually need to do to either bring these cases against someone or defend yourself against one of these claims. And so, I'm very excited to see so many familiar faces here. Welcome back, I'm glad to see you. If this is your first event at the Engelberg Center or your second event, welcome and hope this is the first of many. If you are curious about the agenda today, there's agendas in the back. There's also agendas online. If you are on Twitter today we're using the proving IP hashtag. So I invite you to check it out there. And then see what that second conversation looks like. And now I want to invite up the first panel. The first panel the topic today is proving the creative contribution. And so, once again this is an example of the kinds of questions we're gonna be working through today and tomorrow is thinking about what it means to really prove that there are creative contributions happening across a broad spectrum of types of IP copyright, patent and trademark. So I'd like to invite that panel up and please join me in welcoming the first panel today.
- Welcome everyone. My name is Jeannie Fromer, I am a professor of law here at NYU Law School and one of the faculty co-directors of the Engelberg Center. I don't want to take time away from the substance of our panel, so I'm going to just briefly introduce everyone and give you a sense of what they're speaking about and then we'll dive right into it and hopefully have plenty of time as well for discussions. So to my right is Doug Holley, who is principal at McKool Smith. Doug is going to be talking about patentable subject matter and in particular about some of the percolating issues about proving the creative contribution in the patent law context. Next is Steve Coates, senior counsel trademarks and domain names at Amazon. And Steve is going to be providing an in-house perspective on issues that the company is dealing with with regard to counterfeit social media domain names and the like. Next is Pam Samuelson. And Pam is the Richard M Sherman, distinguished professor of law and information professor at UC Berkeley School of Law and co-director of the Berkeley Center of law and technology. So it's great to have a wonderful counterpart from the West Coast here as well. And Pam is going to be speaking about the copyright side about a number of things including distinguishing between expressive creativity and functional creativity and the burdens of proof in copyright infringement cases. So we're gonna have Doug first, then Pam and then Steve.
- Okay, so thank you Jeannie.
- Your slides are about to come up.
- Well, I was I was wondering about that. I've been doing this a long time, this meaning giving presentations like this in various contexts and it's always a challenge about how you go about beginning and creating and impression of incompetence. It used to be some of you are old enough to remember this, it used to be that you only had recourse to this. Is this thing on? But now we have an infinite number of ways in which we can demonstrate our incompetence before we get started. So having gotten that important issue out of the way, I'm here today to talk about some procedural features of recent developments in the law on patentability which is sometimes called 101 and is sometimes called Alice and sometimes it goes by unmentionable names around a lot of my offices. And I was interested in this because in the meetings that I've attended of the center, there are a lot of really heavy-duty academics involved in this program and that's always very interesting. But I think that a lot of times the procedural framework in which the courts are faced with resolving these legal issues are kind of neglected. And I thought it would be interesting to spend a few minutes talking about how the law has developed in the last five years after Alice and what kind of procedural devices the courts have resorted to to make patentability to decision. So we don't have time to get into a lot of heavy-duty substantive discussions of the law of patentability. You probably remember if you're in this field at all this started with Bilski a Supreme Court decision dealing with the patentability of software then along Came Mayo another Supreme Court decision that that was dealing with diagnostic procedures and then Alice came down that at least was billed as a case that was going to definitively answer some lingering issues about patentability and particularly patentability in the software space. So let's talk about Alice at least for a minute. Alice involves some patents on third party settlement of exchange payments and probably as close as we are to Wall Street, there's might actually be somebody here who knows what all that means, but anyway that's what it did. The defendants CLS Bank filed a declaratory judgment action of invalidity. The district court granted summary judgment to the defendant finding that the patent described an abstract idea that was not patent eligible. The case went to the Federal Circuit. In the Federal Circuit, the original panel voted two to one to reverse the district court and to find that what was described in the patent was patent eligible, but then the case went on bank and the on bank Federal Circuit proceeded to after substantial delay, unleash a series of decisions that as far as I'm concerned are virtually unprecedented. You had a fractured panel of 10 judges of the Federal Circuit, that 10 issued seven different opinions and no single opinion was supported by a majority of the court on all points. If that wasn't enough, Chief Judge Rader at that time the chief judge of the court, also filed something he called reflections. And judge Rader's reflections in explaining the concept of natural laws being unpatentable opined that natural laws are restricted to universal constants, created only by God, Vishnu or Allah. I was bitterly disappointed that he left out Shiva, destroyer of worlds. Well needless to say the Supreme Court took this case and eventually out came the Supreme Court's Alice decision. And that court gave us two steps that the district court and all the courts are going to follow in determining whether an invention described in a patent is truly patent eligible. The first question is, does the patent describe an abstract idea. Now that step bugs me. And it bugs me because I am in the business of figuring out how to describe patents as abstract ideas. When you get ready to explain a patent to a jury or really even say at Markman claim construction to a judge you start off at a level that people can understand before then you have to drill down into the minutiae. And we spend, we meaning people in my firm and other people who do work like this on both sides of the docket, spend an enormous amount of time and energy testings to ourselves sentence by sentence, okay is this understandable but it is also does it fairly capture what the invention is about. And I'm convinced that for any patent that I've ever seen, it can be described as an abstract idea. It can be described as an incandescent bulb that will last for a 1000 hours. It can be described as a flying machine. So what I have come to think of in terms of the first step of Alice is, it basically is the dorm room test. If the patent sounds like something that a couple of guys or girls late at night in a dorm room could say, whoa you know what would be cool, would be computer implemented third party transactions surveillance. If that's all you've got, that's an abstract idea and you're likely to have an unpatentable subject matter. What you have to have to get over that, is the second step of Alice. You have to have an inventive concept and I guess that's what we're here today is talk about inventive concepts and how you prove that. So let's look at some of the ways and I particularly want to look at not so much the substance of how you give this subject matter or that subject matter sufficient weight to be an inventive concept, but what's the procedural vehicle that you're going to ask the court or your adversary is gonna ask the court. And the court district court may choose to decide these questions. The one we have seen most typically is a motion under 12B 6 to dismiss. I would say and I'm not an academic so I haven't counted up the cases I'm sure somebody has or will, but I'm gonna guess that 75% of the cases in district court since Alice have been decided on a motion to dismiss under 12 B 6. Just to take one of them to give you an example. iSentium versus Bloomberg from right down here in the Southern District of New York. iSentium had patents on trading stocks based on sentiments derived from social media. So basically what they have is an algorithm that is going at the speed of light, to query social media and what's going on what's current and to make trading recommendations based on that. It's something you could do maybe if you an army of people who were all in the business of doing that and if they could communicate quickly enough but probably nobody can communicate that quickly enough to make intelligent trading decisions on. The district court in the Southern District of New York found that that invention was an abstract idea. It could be done by humans and the court found that the mere implementation of that human accomplishable idea by a computer is not enough. That's not enough. Everybody knows that computers go fast. So merely by suggesting that a computer get into the act, that's not going to supply you with sufficient inventive concept. Significantly from the procedural perspective which is really what I'm trying to get to here is, the court believed that it was able to decide that on a motion to dismiss simply by reading the patent. It didn't need to hear evidence, it didn't need to delve into the state of the art, it needed to read the patent and find out is this an abstract idea and if it is, is there some disclosure of an inventive concept that still saves it from unpatentability. A closely related, sorry, I got a little behind in my slides. A closely related procedural vehicle is a motion under 12 C for judgment on the pleadings. An example of a case like that is Baggage Airline versus Roadie from the District of Delaware. This was a patent that protects using servers to distribute checked baggage in the court as you would expect followed the Alice structure. It found that the idea of organizing baggage delivery in a particular kind of way was an abstract idea. Could have basically been thought up in the dorm room and step to that the only real inventive feature was using computers to make this process fast enough to be practical. The court said that's not enough, there's no inventive concept there and again what we're interested in isolating here, the court believed that it was capable of making this decision without resort to external evidence of the underlying issues that the court was trying to divine from the patent. So that program that existed for the first four years after Alice, presented us with certain benefits and costs. One of the benefits was that the Supreme Court told us an Alice that this is all a legal issue not a fact issue for the jury. And we have legal issues being decided by judges. Now that may sound kind of tautological but what I'm getting at there is, at the end of the case it's not unusual for example and in this context, what if you have a submission to the jury of some factual underpinnings of a 103 claim of obviousness and that the jury decides that there is not obvious, that there is something in it that is not present in the prior art or wasn't suggested by the prior art and now the judge is thinking, you know I really think that this was not sufficiently inventive. And so, when you get all the way to the end of the process, you're at risk of having a mixture of issues that are pretty hard to sort out in terms of what the judge decides and the jury decides. So the decisions under Rule 12 are useful because they're clean, the judge decides them, he does so or she does so without any procedural distraction The second and major benefit of the Rule 12 dispositions is that it comes early. Typically it's the first thing that happens in the case before any discovery, before massive production of documents by both sides, before substantial expense by either side and I can tell you that both plaintiffs and defendants are interested in having this issue decided early from the plaintiffs perspective. We have a saying around my firm, if we're gonna fall off the mountain we'd rather fall off the bottom than fall off the top. So if that's gonna be the result at the end of the day if the district court is gonna tell you your patent is unpatentable, you'd like to know that as soon as you can so that you can make an intelligent decision either to abandon ship or to appeal to the Federal Circuit and that's a big advantage. But there are some costs. There's little opportunity to articulate or discover facts. We've already talked about that but, it's inevitably it would seem that the inquiry into whether there is a disclosure in the patent of an inventive concept, requires some factual inquiry into what was the state of the art at the time. And that limited opportunity to articulate and discover facts, I think is going to lead to a poorly developed record on appeal for the same reason the Federal Circuit can't just launch off and make up facts, it needs to derive those from the proceeding below and if it's not there, I think that that's a big hole in the case. So that was the state of the procedural law that we were dealing with in this inquiry, into how do you prove enough to achieve patentability, until last year we got the decision in Berkheimer from the Federal Circuit. Berkheimer was not a Rule 12 motion. It was a motion for summary judgment under Rule 56. Berkheimer had patents protecting a certain method of archiving data, the district court found all of the claims to be patent ineligible, they were abstract ideas and the court found that they did not contribute sufficiently to the art to be an inventive concept. So that district court granted summary judgment against the patent holder. The Federal Circuit reversed in part will the part that got affirmed, we won't take time to discuss, but the part we're interested in is, the Federal Circuit reversed the district courts finding on summary judgments. Yet found under step one that the district court was correct Yes, that our archiving data in this way was an abstract idea, could have come up with it in the dorm room, but under step two that summary judgment was error because there was a fact issue about innovation. And specifically, the Federal Circuit told us in Berkheimer, that summary judgment as in most cases is inappropriate where there's a fact issue and in this context there's a fact issue whether something was new or and these are what have become at least over the last year or so, the magic words from the Federal Circuit. Whether something is well understood, routine and conventional. So the battleground here is going to be over at least for, now these words. Was it routine, well understood and conventional or was it something more than that? And the court the Federal Circuit found you can't decide that on summary judgment, it presents a fact issue, so you're gonna have to have a trial, you're gonna have to have some way to resolve that fact issue and it's not going to be on a summary judgment. The court didn't opine that the patent was or was not patentable, it's simply on those claims reversed and remanded for determination of those fact issues. So what I think we're now in the realm of seeing in a typical case is, in a typical case the defendant is going to advocate immediately in the case that it's an abstract idea and not inventive. The plaintiff either in response to a Rule 12 motion to dismiss or a motion for summary judgment is going to offer evidence that the patent teaches something that is not routine. The defendant of course is not gonna take that line down and they're going to want to introduce evidence that it was routine and conventional. Now both sides are gonna say well, wait a minute. If you're filing a bunch of expert reports on this issue, we wanna do that too and while we're at it, we wanna get discovery to find out ahead of time what they're gonna say, so we're now we're basically off to the races and all of the benefits that we experienced at least for a while under the Rule 12 determinations are gonna get swallowed up in the search for fact determinations. Now Berkheimer, is currently before the Supreme Court on a petition for cert. Let's see. There we go. Just give me this last slide. HP petition for cert. Perfect. But it's odd because HP which was the losing defendant in that appeal, pitched this to the Supreme Court in terms of whether this issue is a legal issue for the court or a fact issue for the jury which it's really after Alice no I'm not aware that anybody disputed. It certainly wasn't disputed in the Federal Circuit and this to me really is a question on search that is not the question that the court struggled with at all. But the Supreme Court has as of January requested that the Solicitor General weigh in on whether it should grant cert or not. And the Solicitor General apparently is thinking about it real hard because they still haven't made a recommendation. So what we're left with currently is some unanswered questions. What's the appropriate vehicle for deciding 101 issues, would a motion to dismiss on the pleadings without a fact-finding ever be sufficient after Burkheimer. Now of courts do have even in the face of fact issue what ordinarily we would think of as fact issues on summary judgment. They do have a couple of tools that they can use for example, summary judgment or for that matter a motion on the pleadings, can be defeated only by showing that there is a quote genuine issue of material fact. The court frequently in the face of what they consider to be a conclusory affidavit, will say well that's not a genuine issue and it doesn't bar summary determination. Likewise the rule says that you have to have an issue of material fact. Well, if the Court finds well there's a the proper claim construction undercuts this argument so the facts that you're talking about aren't material, okay that still may be a quick determination. But it seems as though for the vast number of cases unless the Supreme Court says something different, we're going to be dealing with in inquiry into the facts. And to me there's a couple of issues there. First of all, can you allow limited discovery into those facts? Well, you know we already have limited discovery in a typical case on claim construction. Now we're gonna have more limited discovery on inquiry into the facts and how far does that go. I mean why doesn't that involve almost all the whole case? So I'm not sure that that's much good news. Second is a little more theoretical. For decades, the Federal Circuit has battled to keep intact the all elements rule and you would see many cases saying, wait a minute you can't prove infringement by saying well these are the really important claims and they're there so don't worry about these elements of the claim that are not know. You've got to have all the elements. Likewise if you're using a reference to prove invalidity, we don't want to here or oh these are the important ones. You know, you don't have to worry about the fact that our reference doesn't talk about these unimportant ones Nope, the Supreme Court or the Federal Circuit has said, all elements rule, every element is in that sense equally important with every other element. Well, now the path we're going down is, let's look at the inventive features of the claim. Those are the ones the ones that are the gravamen of the invention that's what all this is about those now it seems to me we're getting off into a world that's different than the one that we've been living in for the past several decades. So that's all I got.
- Great, thank you. Pam?
- So, thank you for inviting me to participate in this. It turns out that this happens to be an opportunity for me to talk about some of my pet peeves. So stick with me. So one of the things that I'll be talking about today is what is an affirmative defense and what does it mean to say something is an affirmative defense. So one thing that is clear that if a defendant in a copyright case, wants to say that there's merger or there is fair defenses or fair use, it's up to the defendant to raise those issues and certainly the Federal Rules of Civil Procedure, say that you're supposed to give notice, adequate notice to the plaintiff about sort of the ground on which you will rely in denying liability. And at least as my civil procedure friends tell me, you can either do that by mentioning it in the body of your answer or to have it as listed as an affirmative defense in the list that kind of follows the main body of it. But it's really important to understand that a defense can be affirmative in more than one sense. So one way that a defense is affirmative is if the defendant has to raise it. Second is that a defendant may have to produce evidence in support of certain kinds of defense's that's a burden of production. And a third is a burden of persuasion. And it's important to sort of understand that the fact that the defendant must raise an issue doesn't automatically mean that the defendant bears the burden of persuasion and some of the software cases recently have failed to recognize this. So Cisco versus Arista was a case in which Cisco was claiming that about 500 of the command line interface terms that engineers used to configure Cisco software that were used by Arista was a basis for copyright infringement. And the way the jury was instructed was first to answer the question whether Cisco had proven that Arista infringed its user interface copyright. And then secondly, if they answered yes to that, did Arista proved A, fair use, B, SAF &fair or third a merger defense. And the jury check yes as to infringement no as to merger and fair use, but checked yes as to an affair. So Arista actually won trial. But notice that the way that this is structured, basically puts the affirmative defense as though Arista bore the ultimate burden of persuasion and this seems to me to be wrong as a matter of law. So one of the things we can sort of say, merger and SAF &fair are definitely defenses in that the defendant must raise them but they're not they're not affirmative defenses in the burden of persuasion, since both of these defenses attack the plaintiffs prima facie case. It's up to the plaintiff to prove that original expression was appropriated by the defendant and if there's merger of idea and expression, those are elements that don't get protection from copyright. Similarly, if the similarities between two works are based on their being sent off air elements, common elements for works of that kind, then that undercuts the plaintiffs case altogether and both the Goldstein and the Patrie treatises basically say the courts that say that merger and an affair are affirmative defenses in the burden of persuasion sense are just wrong. And there are some cases also that recognize that. Although unfortunately not all of them. I think it's very prejudicial to the defendants to go into the sort of that part of the case by saying, did the defendant infringe and then is there some excuse that the defendant has and it seems to me that the jury instructions in that case are are wrong. So as you know Cisco versus Arista case, the jury in the Oracle v Google case was instructed to consider first whether Oracle had proven that Google infringed the copyright in the Java API commands that were at issue. And second whether Google had proven its fair use defense. Now the question about whether fair use is a defense to infringement in the raise it sense or is an affirmative defense in the burden of persuasion sense is something about which there are, shall I say differences of opinion. Lydia Loren has written a paper which I cite here in the Washington Law Review, which I think makes a very persuasive case that actually fair use is a raised defense but not an affirmative defense in the burden of persuasion. A sense of that word. The statutory statement is that fair use is not an infringement, not as an excuse for an imprint. But it is not an infringement at all. Section 108 F basically says that fair use is a right under copyright law . And the lands versus universal case I decided by the Ninth Circuit relatively recently, basically said that universal couldn't rely on it's notion that there was prima facie reproduction as a basis for sending the take down notice. They actually had an obligation, because fair use is not an infringement to take fair use into account. And there is a lovely part of Lydia's article where she basically goes through the legislative history of the 76 act and shows that congress considered the possibility of making it an affirmative defense in the burden of persuasion and decided not to. And so the Campbell versus Acuff Rose decision in so far as it basically, says that it is an affirmative defense and that the defendants bear the burden of persuasion, that's just wrong. And it's the it's a shocking thing that it hasn't been corrected. So is that a holding of the case is it dicta? I would like to think it's dicta but Lydia's article basically shows that it matters. So if Lydia's right and I think she is, then the the burden that the district court judges put on the defendants to prove their event those defenses was just error. And in the interest of time I'm not going to talk too much about why the burdens matter but there is actually a lovely part of Lydia's article where she talks about that. And I think it's also important to sort of recognize that the way that this can happen on appeal, is that if if an appellate court sees that the defendant was said to be an infringer by the jury, and then the question is well did the defender overcome that through the defense that puts things I think in a really prejudicial way. And I think it also contributed to the pressure on Arista to settle, while it's peel the appeal actually Cisco's appeal was pending before the Federal Circuit cause Cisco had the oral arguments in oh this case is totally decided by and governed by the Oracle the Google case. So another consideration it seems to me in kind of trying to figure out where the burden of proof in the persuasion sense should lie. So I think the case that Lydia makes that fair use is a defense but not an affirmative defense, but you kinda think about who is in a better position to know what kinds of information. And I think plaintiffs are generally in the best position to prove the nature of the protected work, the amount and substantiality of the appropriation and whether there has been harm to actual or potential markets. Now defendants are probably better situated to say what my purpose was in making the use although plaintiff certainly have something to say about that but it seems to me that if you consider, who has what information that actually is a pretty good idea to put the burden on the plaintiff at least for most of those factors. Now here's another one of my really really think at this and that is the way that courts have been allowing plaintiffs in software copyright cases to basically say that somebody use the structure sequence and organization of my program and therefore they're infringer and certainly the Whalen versus Joselow case from 1986 endorsed that. And there have been SSO claims in Lotus people and Cisco versus Arista and Oracle V Google. And plaintiffs basically say that SSO is protectable by copyright law as long as there is a modicum of creativity to it and there's more than one way to structure the software. And my thesis is the court should require plaintiffs to prove that any non-literal elements claimed as a basis for infringement is actually expressive. One of the things that we wanna know is that functional creativity is creativity to and functional creativity is not supposed to be what copyrights protecting rather what we're supposed to protect it is expressive creativity. Most functional designs are ineligible for copyright protection. And I list here some things that are clearly true. Now Congress created an exception to the non-functionality rule for software and architecture, but it is still the case that functionality limits the scope of protection that's available from copyright law And the question that hasn't really been addressed in the case law is really about what kind of originality. So if you have something that's a functional design that can be creative but it isn't necessarily creative in the sense that copyright there was really looking for. So if you kind of think about what is copyright protection and what does it not protect in software and other things. So clearly copyright protects the code and audio-visual elements of video games, some case law accepts that copyright can protect some non-literal aspects of computer software but there are many things in software that are not protectable. The computer associates the Altai decision gives us several examples, efficient design elements, elements that are constrained by external factors such as interfaces needed for interoperability, public domain, elements these must be filtered out before you get to the comparison stage to make the final determination about whether someone has infringed or not. And then the cases following all tie courts also have been directed to filter out procedures, processes, systems and methods of operation exclusions that copyright added to the 1976 Act to ensure that the scope of copyright would not be too broad. So how do you distinguish SSO from these unprotectable elements. I think we start with the premise that anytime we're talking about a procedure. So an algorithm for example is a procedure that's by definition and computer program algorithm is a procedure. Procedures, processes, systems and methods of operation are all SSO. Okay, they're all SSO. That's what it is. A structure that constitutes a process or a system and at least some of that program SSO is eligible for patent protection and that's the kind of SSO that Congress meant to filter out through section 102b terminology. Also some program SSO is after LSB CLS Bank. Not even protectable by patent law. And it doesn't make any difference how original it is and it doesn't make any difference how many different ways there might be to perform a particular function, it's just not protectable at all. So it's an abstract idea for the purpose of the patent system. Why wouldn't it be an abstract idea for the purposes of copyright law? It seems to me perfectly logical. So I think that there should be much more attention in these software copyright cases to unpacking what kind of SSO are we really talking about and to the extent that you show that SSO is the product of functional creativity rather than expressive creativity. It seems to me that courts really ought to be paying much more attention. So I won't repeat my conclusions but here they are. Thank you for your attention.
- I feel like I'm gonna be much lower energy. I don't have slides. I didn't even fully conceptualize what I was gonna talk about. But can I ask some questions to the audience. How many people in the room practiced trademark law as part of their practice? Ah, so I've got experts here, that's cool. We speak the same language, I appreciate that. How many people in here have worked in house? Oh man, my people are here. This is great. Maybe I will be a little more lively. So I worked 14 years at law firms before I went in house and it was an experience. I went to my first naming meeting with Dave Limp who runs our digital products and services. And it was the naming meeting that became what is now known as the Fire Tablet. And I sit down and his first question to me, brand new baby in-house lawyer is what do you think of this name. And I would, I said arbitrary. I did not have the language of communicating in-house, trademark or marketing speak. And I had to learn that. I spoke a week ago we could go in London at a MarkMonitor Conference and I started out by showing a Malcolm Gladwell quote, in his book tipping point where he says or he argues that it takes 10,000 hours to be an expert. Which you know, for that particular audience was mostly for domain names. And domain names management is like the redheaded stepchild in every organization. It's given to someone who is not an expert. And it's such an important thing. Becoming an expert at a law firm takes about five years if you're an associate, but if you're doing something part-time it takes you know 20 hours a week for 10 years. So becoming an expert is really hard. And I work on things like a few weeks ago we launched Project Kuiper which is going to be a low orbit satellite constellation around the world which will provide Internet service to everyone. I don't know a lot about that. So I'm not an expert about Project Piper. When I sit down with a client I ask lots of questions. They are seemingly stupid. And I got used to asking really silly stupid questions because my clients and I refer to them as my internal clients, they ask me stupid questions all the time too. And there is this this happy medium and I think you find happy accidents when you've asked lots of questions. So I work on technology but I'm not sure what it's going to be. When we launch Kindle 11 years ago, we thought Kindle might be the name that extends to the entire ecosystem for all our devices, but now we know Kindle is largely for e-readers and services around reading. We know Fire is a tablet. It's something that you can hold in your hand and Fire TVs in your living room, but when we launched these names, I have no idea what they're going to be. So I have to have conversations with the clients about what are your plans and what do you think this could be in 10 years. And that's a really hard thing to do we don't we don't necessarily know what things are gonna be in 10 years. When I talk to my clients I do give them an education. Maybe once a quarter I'll go to a client, it could be our mass advertising business, it could be prime video, it could be our device team, last week it was the Alexa team and I give them the distinctiveness spectrum and we talk about it. It's I think the the closest thing that you can use to proving creativity for ownership or register ability or protect ability of trademarks and they're starting to speak my language now. So we go through it and I test them, fanciful mark you know that's a totally coin word, right? Kodak is a great one, arbitrary we talk about how Apple is arbitrary. It's a defined term. Most of our brands are defined terms and that can be arbitrary. Amazon is arbitrary, suggestive suggests the quality or the the essence of the service. So Greyhound might be arguably suggestive of the speed with the bus system. And then descriptive which Weedmaps which is very popular in Seattle right now. Weed maps is completely descriptive of the app and service that it seeks to function. And then Generic is the product itself. And they don't always get it perfect but we talk about that in meetings. The naming meetings the actual creativity itself can be, there's not a lot of magic in it. You're working in a law firm and you get these these marks that you go ahead and search and clear and provide reports on and you wonder what the the machine is behind that, is there some sort of market testing and things of that nature. I was in a naming meeting with with Jeff and Dave who runs our device business. And Jeff asks, you know what were these naming meanings like it at Apple what Dave used to work. And he said, it's usually just a bunch of dudes around a table, and it's not always dudes but it was that day. But that's generally what happens. So over the, it was last fall we had a naming meeting with Jeff and some of our SVPs on what is now called Scout. And Scout is a sidewalk delivery robot that looks like it looks like an ice cooler that just goes down your street and it's beta-testing in in the northwest right now. And it's parked at Amazon lockers which are these lockers that UPS delivery people can just stick your stuff in a locker and they're very popular in urban areas. And the the robots there too and Scout will go self-drive itself down the last mile to your home and then send you a notification when it's on your front porch, and then you go out and you tell Scout this word this keyword that you get in your notification. It'll pop open and deliver your product. So it's small. And we're around the table, there's usually a host of suggested names for the product and Jeff didn't like any of them. So he said I really I really think we should think about dog names. Can anyone everyone just start brainstorming dog names. And the SVP whose product this was, he said well I just adopted a dog four months ago and we called it Scout. And he's like Scout, perfect, done! Of course there was some more work that went into that. But he thought that that was the perfect story for this particular robot, because it's more personal for the SVP who created it. And that's generally like that's the creativity that goes in there. So we talked about how important those names are gonna be, right? So Alexa for example, I argue is Amazon's most important trademark because it's the name of the service, the AI system. Does anybody even know what Alexa is here? Okay, you know I speak in different countries. So not always I always get the same audience. So Alexa is the voice assistant. It's something that people are personally attached to. I am, she doesn't make me feel lonely when I go home tonight or wake up really early in the morning. It's also used by so many third parties. So many third parties put Alexa into their products and services. Alexa is the name that allows so many devices not just ours but others function. So that's a really important mark that if we were enjoined, the impact would be significant. If we lost the Amazon trademark, we could just rename it and people would find us. If we had to rename Alexa, things would break and people would be personally, I think have a sense of loss for losing that term. And so, I argue that Alexa is the most important term and when we go into meetings and we talk about the future of these devices, I need to ask a lot of questions about what this device is going to do, where is it going, is it going to be just in or is it going to be an operating system that other people will use. And we're gonna see more more things like Alexa, Siri and Cortana where those things are important. Alexa also is a 20 year old brand. Well, Alexa the voice assistant is not 20-years old but the name originated from an SEO company than Amazon owns called Alexa, alexa.com. And in that space it's very important. Who knew that Alexa 20 years later would be what it is today. So there's some of the things that we talk about in these meetings in terms of creativity and how important that is. And there are some devices where the attorneys in the room have named them. So Echo shows my name. I named that one. So the creativity is a two-way street and it's collaborative process. What I do give my trainings, I throw up the distinctiveness spectrum and then I flash Amazon's brands after. And I quiz people on they what they think it is on the spectrum. And I play tricks of course, but the one I like to throw up there is Prime. Prime is a term that is considered laudatory in many countries for a host of goods and services. So it may arguably be unprotectable. Prime is Amazon's second most important mark and survey additive evidence has suggested that broad numbers of people not just in U.S. but in Europe and elsewhere, hugely high numbers associate Prime with video services that Amazon provides, with delivery services that Amazon provides, music services that Amazon provides, with with association rates of 85% or higher. That suggests to me that that is a protectable mark even though trademark registries may disagree with that. The fact that consumers associate those things with us and that's something that I think it's very apparent when you're in-house and you're talking about the protectability of that brand. Well in the world of online services, take down notices are incredibly important. And so, when we're having these conversations even though we might have a brand that will have a huge consumer knowledge or association, it may not be protectable and therefore I may be flat-footed later, if we have to make sure that people are using it for phishing or try and get user login credentials or other nefarious things. I'm gonna be a little flat-footed when I have to go deal with those things or there's other things when we talk about protectability it's more on the trade dress front and less on the trademark side of let's say voice assistants Voice assistants are increasing, they're going to be used for nefarious purposes in the very near future. It's not hard to record Alexa's voice and then use that for counterfeiting, getting people to buy counterfeits or other other things, getting bank account information. So we're gonna see that and there is no great mechanism to protect voice assistants through prima facie rights. So of course you can go to court and you can argue that we own intellectual property rights in the voice assistant, but I don't have something that I can go to a hosting service or a registrar or another ISP that says actually these are rights that someone else is stamped and said there are prima facie and I should take this down for removal. So there are other areas where I do flag to the business that hey we have this great creative thing, there's not this mechanism or echo devices the Pringles cans have a light ring around them. That light ring has certain activity that means certain things. And it's changed over time. We do have a motion mark here in the U.S. registered motion mark on that light ring activity but it's now different colors. And so, our ability to protect that light ring is not as quick as the innovation around that light ring but people associate that light ring with Amazon and Alexa. The light ring has also changed shape. So it's a square now, it's an oval. We license it to other parties to include in their technology and people do that is there's a strong association, super hard to protect. And so, those are those are some of the things.
- Well, thank you. I want to make sure there's time for questions. I think this is really great to put the talks back-to-back because I think it really highlights the different emphasis on the types of creativity in patent, copyright and trademark. I have tons of questions to ask but I think I'm gonna be bothering the panelists afterwards. I'll let you ask your questions now. So there is a microphone over there and over there. So if you have any questions if you could come up to the microphone and we could take them. I'll ask in the meantime while people.
- [Man] Well, I may start off by breaking the rule because I think I have a comment rather than a question. But if I could follow up on something that Pam Samuelson said about the Lydia Loren article about fair use being the burden of proof being on the plaintiff. I think that is clear from the statute that establishes a boundary between the rights of a copyright holder and fair use. And I don't like using real property and tangible property analogies generally in this arena but I think it's useful here, because in most states a coastal land owner, owns up to the mean high tide line of the coast. And if that owner wants to sue somebody for trespass for walking up the beach, it is not the walkers duty to prove where the mean high tide line is. It is the property owners duty to prove the boundary and to prove that the trespasser was on it. So if we apply that analogy to fair use, it is the copyright holders duty to prove where that boundary is between the rights and fair use. If I could follow on one additional area and I think there is a lack of attention sometimes to where fair use fits into secondary liability cases. And actually the Supreme Court spoke and this often gets neglected. In Napster the defendant was arguing a defense of fair use in a secondary liability case and judge Patel didn't agree and said it's the defendants burden. But in a contributory infringement case, it is the plaintiffs duty to prove the underlying infringement. And the defense there is not the defendants defense. It's the absent party's defense. The plaintiff still has to prove the burden and the Supreme Court ratified that view in the Sony Betamax case. Because the Supreme Court said that the respondents that's the movie studios had not carried their burden on establishing one of the fair use elements. So I want to make sure that we tease out at least under current law. The Supreme Court has spoken in secondary liability cases that fair use is not the defendants burden. Thank you.
- Can I just add that the Campbell versus Acuff Rose cites to Harper & Row for the affirmative defense. Harper & Row didn't deal with the burden of persuasion issue cites to legislative history. Harper & Row cites the legislative history which basically they mischaracterized. So the actual house report supports that it's a defense not an affirmative defense and yet Harper & Row says it's affirmative defense and then the Supreme Court and Campbell follows up citing to Harper & row that cites to this legislative history that's error. And that's where we are.
- [Man] So sorry two questions in a row for Pam. But it struck me when you said this in SSO cases there is a difference between expressive and functional creativity. And I wanna get your sense of how we could explain that difference. So it's not just cases about SSO, it's cases for example I'm thinking of Delta Dental which was about a taxonomy for kind of naming dental procedures. And the judge in that case Judge Easterbrook said, well the the American Dental Association is making choices and constructing this taxonomy. For example they have five digits in the code instead of four. And you know there's a reason for that. They want to make the code capacious enough so that as new dental procedures are invented, they can code them. So my students always ask well, how is that expressive? And I'm not actually sure what to tell them. So is there a way that you think of the dividing line between what you call expressive creativity and what you call functional creativity.
- So I wish I had a perfect answer to that question but I actually one of the things that I've done in the last few years is I wrote an article that was published in the Houston Law Review about functional compilations. So one of the things that the American Dental Association case sort of says, oh well as creativity in deciding the name and this number to associate with it. And so it's protectable as a taxonomy. It's really a coding system and it shouldn't be protected by copyright at all. I've argued in another article, but I took the opportunity to read a lot and follow up on dozens of cases involving functional compilations where there was some creativity, but it was functional creativity. And so, in that article I began the development a taxonomy of what does it mean for something to be functional. And I think a lot of the cases that I cite to in that particular article are good examples of something where you can sort of say, this is functional creativity, right? In the ATC versus whatever it takes case where the plaintiff was relying on the American Dental Association case saying, oh this is just a taxonomy. Well what it was is the name of parts for transmissions and numbers associated with those transmissions. And the court says that's functional. One of the things that I didn't get a chance to really talk about, but there was a nugget of it in one of my slides and that is that when it comes to pictures of functional things, we can easily see here's the drawing that gets protected as a drawing but the useful article that is depicted therein is not within the scope of protection. Similarly when Bikram was claiming that the yoga sequences was expressive, the court was able to say no you basically gave all these reasons why it's good for health and it's good for safety and it's good for these kind of functional things. And so that's not protectable by copyright law. It's a little harder for us to be able to distinguish between what's the product of functional creativity and expressive creativity when it's in words. And so, the functional compilations paper gives a whole bunch of examples of where courts basically said this is functional creativity that's not expressive creativity. So I think we need to work more on developing that taxonomy so to speak, but it does seem to me that it's critical and software is really where it has been playing itself out as a really difficult problem for the last several decades.
- [Man] There is as you know in the in trademark law a general rule that frames are not trademarks. Have you confronted that in terms of the square or the circle in the Echo cases? Or if you contemplated that as a defense how do you get over that defense?
- Well, thankfully I haven't had to do that yet. I think so much of enforcement efforts, fraction of 1% is actual litigation. So much of it happens behind the scenes.
- Let me just ask a quick questions. Well, Doug do think that, so Markman you know rocked all these changes to patent. It's been hard for me to see that they were fully contemplated by the court. When it made them. Do you think that Alice had this built-in sense of wanting to resolve patent cases quickly, built into it or it was a byproduct of how the courts interpreted it after the fact.
- Probably more the latter than the former. I don't recall offhand anything in Alice either at the Federal Circuit or the or the Supreme Court that addressed that desirability. In Berkheimer though the Federal Circuit did even though it was dealing with the case on summary judgment did sort of reach out and gratuitously say oh and by the way we're not suggesting that the court can't use the motion to dismiss and motion on the pleadings because we've affirmed those in the past. But then they cited two cases in which both of those cases the patent holder had conceded that there was no fact issue. So really I thought that was kind of it wasn't it was an overt effort to preserve the ability to do something early but I thought that it was kind of half-hearted.
- Okay one last question I'll take us a little bit over time because I'm dying to ask Steve, a little bit about the extensions or a lack of extensions of brand names into other spaces. I mean you mentioned how Alexa it started out as a SEO name and it's expanded in other ways. You could have chosen another name but you didn't. You talked about how Kindle was intended perhaps to be used more broadly but then it ended up being used for e-readers and not for all of the potential like tablets sort of space. Is there a sense of how this happens? Is it different in each of these contexts. Is there some rhyme or reason to this? It's really interesting to peek inside the naming--
- I don't know that there's a rhyme or reason for it, but I think about learning lessons. And I think I learn a lot from other people's poor judgment. And I used to work at Twitter and managed escalations. And it there'd be a couple times a month when I'd have to hop on the phone with an angry CEO or general counsel from a company because they were upset about something on the on the platform. And you learn a lot when you see other people's mistakes and poor judgment. And I talked to, I have a team of 12 and lots of new people coming on the larger, our trademark team is 35 and they've grown significantly. And so, training lots of new people. And I and the heart concept is to kind of contemplate that you don't know where this brand is gonna go. And you know certain things. You know like Amazon's moved into the healthcare space now. I probably wouldn't have seen that three years ago. We have low orbit satellites, we have sidewalk robots, we have a host of things that I can't talk about in this room that are in in the works that you know, the Echo brand may or may not extend to. And so, you have to contemplate that all the time that expansion and the flexibility that the brand might have, with also knowing that brand may have you know places or scopes that it may not be able to expand to. And keep that in mind and have that constant conversation with the business. So no rhyme or reason necessarily, but it's a it's a process that we talked of Scouts actually a repurpose brand, but I talked about four, we have two other Scout products. Things get resurfaced all the time. Please join me. Oh well, maybe we'll ask it afterwards. I think we're running a little bit over time. Please join me in thanking the panel.