The Abolition of the Free Copy
On true digital scarcity, the trusted enclave, and property whose laws are chosen by the one who makes it — beginning with the oldest law of all: that to give a thing is to no longer have it
Keywords: digital scarcity; the free copy; rivalry and excludability; trusted execution environment; secure enclave; remote attestation; non-fungible tokens; programmable property; resale royalty; code as law; the non-rivalry of information; first-sale doctrine.
Abstract. The defining fact of the information age has been that a copy costs nothing. From this single fact flowed the abundance, the lawlessness, the unpayable creator, and the curious condition of a civilisation whose most valuable goods could be owned by no one because they could be had by everyone at once. This essay asks what changes when that fact is repealed — when a digital good can be made to behave like a physical one, scarce in possession and transferable such that the giver is left without it. I trace the problem from the non-rivalrous nature of information, through the solution of double-spending for money, to the general apparatus by which arbitrary files may be made expensively and accountably scarce: cryptographic access bound to a key, a public ledger of title, and a hardware root of trust — the trusted enclave — that can hold a secret its own owner cannot extract and prove remotely what it is doing. I am candid about the limits of that apparatus, which has been broken and will be again, and I argue that the right word is never “uncopyable” but “costly and accountable to copy,” which is all scarcity has ever meant, even for gold. I then describe the deeper consequence: that the maker of a digital good may now compose it to obey whatever law he chooses — to be open, or owned, or to return him a fraction of every resale — and may keep, delegate, or abolish the authority that enforces it. The conclusion is that this revises the foundation of information technology itself, for information may at last be not merely shared but given, not merely accessed but owned; and a world in which scarcity is chosen rather than suffered is a world that must, for the first time, argue about what it chooses.
I. The original sin of the copy
There is a sentence that has governed the digital world more completely than any law passed by any parliament, and it is this: a copy costs nothing. To duplicate a file is not to labour but merely to permit; the second copy is as perfect as the first, the millionth as perfect as the second, and the whole apparatus of reproduction asks of its operator nothing but consent. We have been taught to regard this as a triumph, and in a narrow sense it is — the cost of distributing human knowledge has fallen to a rounding error, and a child with a telephone now commands a library that would have humbled an emperor. But every liberation redistributes power, and it is worth asking, with the cold attention such questions deserve, what exactly was liberated, and from whom.
The non-rivalrous character of information is not a discovery of the computer age; it was stated with uncommon grace two centuries ago, by a man considering whether an idea could be owned. He who receives an idea from me, he wrote, receives instruction himself without lessening mine; as he who lights his taper at mine, receives light without darkening me [1]. The image is exact and beautiful, and for ideas — for the pure abstraction, the theorem, the melody considered apart from any recording of it — it is simply true, and no amount of engineering will make it false. The economist of a later century gave the property its formal name, defining the public good as one whose enjoyment by one man does not diminish its availability to another, the rigorous statement of the taper’s light [2]. A thing that can be infinitely shared without loss is a glory when the thing is knowledge and the aim is enlightenment.
But not everything we wish to treat as property is a pure idea, and the digital file is a peculiar hybrid: it is an abstraction wearing the costume of a thing. We speak of owning a film, a song, a book, a game, a work of art, a key, a credential, a coin — and we mean by it something more possessive than mere acquaintance with an idea. We mean the thing should be ours, in the strong sense that to give it to another is to part with it, that to sell it is to no longer have it, that there is, somewhere, an original of which we are the keeper. And here the free copy, which was a glory for knowledge, becomes a quiet catastrophe for property. For if every file is a taper that lights infinite tapers without dimming, then nothing made of bits can be owned in the strong sense at all. It can only be accessed, licensed, permitted, surveilled — held at the sufferance of whoever controls the copying. The free copy did not merely make information abundant. It abolished the possibility of possessing it, and we mistook the abolition for a gift.
II. Why a thing is a thing
To see what was lost, one must say plainly what it means for a thing to be a thing, and the saying turns out to be less obvious than the having. A physical object has two properties that the philosophers of economics have named without quite capturing their strangeness. It is rivalrous: if I am holding the apple, you are not, and my eating it forecloses your eating it. And it is, to a degree set by locks and law, excludable: I can keep you from it. From these two homely facts the entire institution of property descends. Ownership is meaningful precisely because the owned thing is rivalrous — because to have it is to deprive others of it, so that the question whose is it? has a single answer and a real stake. A good that everyone can have at once needs no owner, in the way that air needs no title deed; it is only the scarce thing, the thing that one hand’s holding denies to another, that property was ever invented to govern.
The rationalist’s favourite axiom does honest work here, as it did when we considered the honest game. A thing is itself; it is one, it is here and not there, it is had by this owner and not that one. Identity, for a physical object, is enforced by the brute geometry of the world: the apple cannot be in two mouths, the painting cannot hang in two houses, the coin cannot rest in two pockets. This is not a limitation to be engineered away; it is the very thing that makes the coin a coin and the painting a possession. When I hand you the apple, the transaction has a moral and economic completeness that we feel without articulating: I have given, which is to say I have parted with, and you have received, which is to say you now have what I no longer do. To give is to be without. That is what giving means. It is the small daily sacrifice that distinguishes a gift from a boast.
The digital file broke this, and we called it progress without counting the cost. In the world of the free copy there is no parting and no being-without; there is only proliferation. When I “give” you a file I have not parted with anything — I have merely caused a second instance to exist, and I retain the first, undiminished, as does everyone I have ever sent it to. The word give has been hollowed out and refilled with the word copy, and the two are not the same, however casually we have allowed them to merge. A thing one can give without losing was never one’s own; it was only near one. And so the digital age, for all its abundance, became a place curiously incapable of three of the oldest human acts: it could not truly own, because nothing was scarce; it could not truly give, because nothing was parted with; and it could not reliably pay the maker, because the thing he made escaped his keeping the instant it left his hand. We built a civilisation of infinite reproduction and were surprised to find it a civilisation in which the creator starved, the owner owned nothing, and the gift had become a mere notification.
III. The double-spend, and its cure
The first domain in which this poverty became intolerable was money, because money is the one digital good whose entire function is to be scarce. A coin that can be spent twice is not a coin; it is a counterfeit engine, and a currency in which every unit can be perfectly duplicated by its holder is not a currency at all but a cruel joke played on whoever accepts it last. The problem even acquired a name — the double-spend — and it is nothing more than the original sin of the copy applied to money: if a digital coin is a file, and files copy freely, then the holder of a coin can pay it to a thousand merchants at once, and the medium, which knows only how to duplicate, offers no defence.
For a long while the only cure was to reinstate the very figure the digital world had seemed to make unnecessary: a trusted third party. The pioneering schemes of digital cash placed a bank at the centre, issuing coins by a cryptographic signature and standing ready to detect, by clever means, any coin presented twice [6]. The cryptography was elegant — it could even preserve the payer’s privacy while exposing the cheat — but the architecture was the old architecture: a mint that must be trusted not to inflate, not to censor, not to fail, the dealer returned to the table under a new name. Scarcity was achieved, but only by appointing a sovereign of scarcity and praying he was honest.
The decisive move was to achieve the scarcity without the sovereign. A public, append-only ledger, maintained by no central keeper and agreed upon by its participants, can record who owns which coin and refuse, by the plain logic of the record, to let the same coin be spent twice [9]. The double-spend is defeated not by a trusted mint but by a shared and unforgeable history of title, so that the question whose coin is it now? has a single answer that anyone may check and no one may quietly rewrite. This was a genuine achievement, and it taught a lesson larger than money: that uniqueness of title — the fact that exactly one party owns a given digital token, and that ownership transfers cleanly from one to the next — can be established for a pure number, in the open, without a master. If a coin, which is only a number, can be made scarce and transferable in this way, the obvious question is why a song, a book, a key, or a work of art cannot.
The answer reveals precisely where the difficulty lies, and it is worth stating with care, because here is where enthusiasm usually outruns honesty. A ledger can make title unique. It can establish, beyond dispute, that this token is owned by this person and was transferred to that one. But title is not the same as possession of the bits. To own the deed to a house is not to occupy it; to own a token that points at a film is not to control the film, which may sit on a thousand hard drives regardless of who holds the token. The ledger solves the uniqueness of ownership and leaves untouched the uniqueness of the thing owned. To make the file itself behave like a scarce object — so that the owner of the title is genuinely the only one with access, and a transfer of title is genuinely a transfer of the thing — requires something the ledger alone cannot supply. It requires a place where a secret can be kept against the will of the very person holding the machine.
IV. The enclave: a room no one can enter
Consider what would be needed to make a file scarce in the strong sense — to ensure that access to its contents could be granted, transferred, and revoked, and that when Alice passed it to Bob she truly lost the ability to open it. The contents would have to be sealed under a cryptographic key, for encryption is the only lock the digital world possesses [4, 5]; and access would then reduce to control of the key, transfer to the handing over of the key, revocation to the changing of the lock. But this merely relocates the problem, for a key is itself a file, and files copy freely. Whoever once holds the key can keep a copy of it, and with it a copy of the contents, forever. The whole edifice of digital scarcity therefore rests on a single, almost paradoxical requirement: a place where a key can be held and used without its holder being able to extract it. A vault that the owner of the house cannot open.
This is the strange and beautiful idea that hardware has lately made real, and it goes by the name of the trusted execution environment, or the secure enclave. The notion, refined over the past decade in commodity processors, is to carve out a region of a machine’s execution — a sealed room inside the silicon — whose contents are protected even from the operating system, even from a user with full administrative power over the device, even from one who physically owns the hardware [10, 12]. Code and data placed inside the enclave are encrypted and integrity-protected by the processor itself; the owner of the computer may run the enclave but may not look inside it, may invoke it but may not pry it open. Into such a room one may place a content key, and a small program that will use that key to render a film or unlock a book only under the conditions its author specified — and the user, for all his control of the surrounding machine, cannot reach in and steal the key out.
Two further capacities turn this sealed room from a curiosity into the foundation of digital property. The first is sealing: the enclave can encrypt data to its own identity, so that secrets it stores can be recovered only by the same enclave on the same machine, and by nothing and no one else [11]. The second, and the more remarkable, is remote attestation: the enclave can produce a cryptographic proof, verifiable by a distant party, of exactly what code it is running and that it is a genuine enclave on genuine hardware [11, 12]. This is the mechanism by which a creator, or a ledger, or a counterparty, may decline to deliver a secret to a device until that device has proven it will keep it — until the online hardware in question has attested its own integrity and the rules it will enforce. The “thing in the middle,” the stored secure element that holds the key and adjudicates access, need not be trusted on faith; it can be made to demonstrate, before it is entrusted with anything, that it is what it claims and will do what it promises. Here, at last, is the apparatus by which bits may be given a body: a hardware room that keeps a key against its owner’s will, that proves remotely what it is doing, and that can therefore be made the sole gate through which a file is opened — so that to control the gate is to possess the file, and to transfer control is to transfer the thing.
V. The honest limits
It would be a fraud, and exactly the kind of fraud this lineage of work exists to make impossible, to claim that the sealed room is impregnable. It is not, and intellectual honesty requires that the limits be stated as plainly as the powers. The enclaves of real processors have been broken — not in theory but in practice, by researchers who extracted the very keys the hardware was built to protect, using subtle flaws in the way modern processors execute instructions [14]. The same technology that some film distributors adopted to guard high-resolution video was shown to leak its secrets to a sufficiently ingenious attacker [14]. A determined adversary with physical possession of a device, or a newly discovered microarchitectural weakness, can sometimes pry the room open; and no enclave yet built should be regarded as a permanent guarantee against a well-resourced attacker who holds the hardware in his hands.
There is a second limit, more fundamental still, and it is one that no advance in hardware will ever repeal. A copy made before the lock is applied is beyond all recall. If the contents were ever in the clear before the maker imposed his scarcity — if the film was filmed, the song recorded, the manuscript typed in the ordinary world before being sealed — then a copy taken at that moment exists outside the system entirely, and no enclave, no ledger, no revocation can reach it. Scarcity engineered after the fact can govern only the instances that pass through the gate; it cannot un-ring a bell already rung. The honest claim is therefore never that a file has been made uncopyable, which is a word for marketing and not for mathematics. The honest claim is narrower and far more defensible: that a file can be made expensive and accountable to copy — that copying it can be made to require defeating a hardware root of trust, an act available only to the skilled, the patient, and the well-equipped, and not to the ordinary holder at the press of a button. This is, when one thinks about it, all that scarcity has ever meant, even for the most physical of goods. Gold is not uncopyable; it is merely expensive to counterfeit and difficult to forge undetected, and on that “merely” the whole institution of money rested for millennia. To make a digital good as scarce as gold is not to make it magical. It is to move it from the category of the costlessly duplicable into the category of the expensively forgeable — and that single migration, modest as it sounds, is enough to make bits behave as property for every practical and economic purpose, provided one names, honestly, the parties one is trusting: the maker of the hardware, the integrity of the enclave, the soundness of the attestation. State the trust surface, and the achievement stands. Conceal it, and one is selling a perpetual-motion machine.
VI. The property that chooses its own laws
Grant the apparatus, with its powers and its stated limits, and something appears that is more interesting than mere copy-protection, and that the language of “digital rights” has always been too crabbed to express. The maker of a digital good may now compose it to obey whatever law he chooses, and may write that law into the thing itself, so that it is enforced not by a court he must petition after the fact but by the construction of the object, automatically and in advance. The notion that a relationship or a rule might be embedded in a self-executing mechanism, rather than entrusted to the goodwill of the parties and the slow recourse of the law, was set down with prescience before the means existed to realise it generally [7]; and it was understood, even then, that such mechanisms reduce the need to trust and to litigate, that the humble vending machine — which dispenses on payment, with no clerk and no contract signed — was the primitive ancestor of a whole coming order [7]. The complementary insight, stated in the same era, was that in the digital world the architecture is the regulation — that code, no less than law, determines what one may and may not do, and that whoever writes the code writes the rules of the space [8]. These two ideas, joined to the apparatus of scarcity, yield the central possibility of this essay: that property, in the digital world, need no longer take its rules from the accidents of the medium, but may take them from the deliberate choice of its author.
Consider the range of that choice. A maker may decide that his work shall be open — freely copyable, given without reserve to the commons, a taper offered to light every other. This is a noble choice and a common one, and the point worth insisting upon is that it becomes, for the first time, a choice rather than a fate. In the world of the free copy, openness was not chosen; it was imposed by the medium upon every maker alike, the generous and the grasping, the one who wished to give and the one who wished to be paid, with perfect indifference. To make a thing scarce by construction is to restore to its author the dignity of deciding to make it free — for a gift is only a gift when one might have withheld it, and generosity imposed by the impossibility of property is not generosity at all. Alternatively, the maker may decide that his work shall be owned — a single, scarce, transferable object, of which there is an original and a keeper, which may be sold and thereby parted with, lent and thereby returned, inherited and thereby possessed across generations. The token standards of the public ledgers gave the first crude grammar for this — a way to represent a unique, transferable title, a deed to a digital thing, distinguishable from every other and assignable from one owner to the next [13] — and joined to the enclave that makes the underlying content genuinely scarce, the deed becomes more than a pointer; it becomes the title to a thing one truly holds.
And between the wholly open and the wholly owned lies a continuum the older medium could not represent at all, in which the maker attaches conditions to the thing he sends into the world. He may decree that a fraction of every future resale shall return to him — the resale royalty, which the law of some nations grants to artists under the name of a continuing right in their work, and which the construction can now enforce without any law at all, so that the painter shares in the rising fortune of his painting through every hand it passes [3, 7]. He may decree that the work may be used but not modified, lent but not duplicated, enjoyed for a season or in perpetuity. He may, in short, design the bundle of rights that travels with the thing, and the bundle will be honoured by the mechanism rather than merely promised by the parties. The economist who studied the emergence of property observed that rights to a thing arise when a change in technology makes it economical to define and enforce them — that property is not a fixed datum but a response to what the times make worth owning and possible to hold [3]. The times have now made it possible to hold a digital thing; and so the rights, which the free copy had rendered unenforceable and therefore moot, become once more economical, and therefore real.
There remains the question of who enforces, and here the present essay rejoins its predecessor in a pleasing symmetry. The earlier argument concerned the abolition of the trusted third party — the demonstration that a fair game needs no dealer, that distrustful parties can transact with no master between them. The apparatus of digital property does not contradict that abolition; it generalises it into a choice. The maker of a thing may decide that no third party shall stand over it at all — that it shall be governed entirely by its own construction and the public ledger, with no one able to alter, revoke, or interfere. Or he may decide to reinstate a third party deliberately and on terms — appointing himself, or a chosen group, or a quorum of trustees, as the authority that may, under conditions he specifies, revoke access, resolve disputes, or update the rules. The third party, banished as a necessity, returns as an option: present or absent, singular or distributed, by the maker’s design and the participants’ consent rather than by the helpless default of a medium that offered no alternative. This is the deeper meaning of property that chooses its own laws. It is not merely that the rules of a digital good may be authored; it is that the very constitution of the good — including whether it has a sovereign, and who that sovereign is — passes from accident into deliberate, contestable choice.
VII. True scarcity, and the return of the gift
It is worth dwelling on the most human of the consequences, the one that has nothing to do with markets and everything to do with the texture of ordinary life. When Alice gives the file to Bob and loses all access — when the transfer is a true transfer and not a duplication, so that what Bob now has, Alice no longer does — the digital gift becomes, for the first time, a gift. This is a small sentence with a large content. It means that the oldest and gentlest of human transactions, the parting-with that distinguishes giving from showing, is restored to the digital world, which had forgotten it. It means that one may give a digital book as one gives a physical one — by handing it over and being without it — rather than by transmitting a copy and retaining the original, which is not giving but lending one’s own abundance. It means that the thing borrowed may be genuinely returned, the thing sold genuinely relinquished, the thing inherited genuinely possessed by the heir because it was genuinely possessed by the deceased.
The list of what this restores is long and touches the foundations of cultural and economic life. It restores the possibility of a genuine market in used digital goods — the sold game one no longer owns, the resold book that leaves one’s shelf, the secondary market that the doctrine of first sale once guaranteed for physical objects and that the free copy had quietly annihilated. It restores the artist’s stake in the rising value of his work, paid automatically as it changes hands. It restores the meaningful distinction between the scarce original and its infinite reproductions — the difference, in the digital realm, between the signed first edition and the library copy, between the master and the print, a distinction the medium had flattened to nothing. It restores, in a word, the possibility that a digital thing might be one’s own in the full and serious sense, with all the dignity and consequence that ownership of real things has always carried.
But the honest essay does not sell a paradise, and there is a shadow here that must be acknowledged with the same plainness as the light. The very mechanism that lets a creator be paid and an owner truly own is also a mechanism of enclosure, and a world of engineered scarcity can become a world of new fences. The thinker who first taught that code is law saw the danger clearly: that if every use of a work can be metered and licensed by a trusted system built into the hardware, then the breathing room that older law left around property — the right to quote, to lend, to resell, to make fair use, to do with one’s own book the ordinary things a reader does — may be foreclosed not by any statute but by the silent architecture of the thing, which simply will not permit what it was not designed to permit [8]. Scarcity, made a tool, serves whatever hand wields it; the same construction that lets the painter share in his painting’s fortune lets a rentier lock a door and charge for every glimpse through it. The reader who once owned his book outright may find he now merely rents access to it on terms he did not write and cannot change. This is a real danger and not a rhetorical one, and the answer to it is not to pretend the mechanism does not exist, nor to wish it away, but to insist that because scarcity has become a choice, the choosing must be done in the open and argued over honestly — that the maker’s freedom to compose the rules of his property is matched by the buyer’s freedom to refuse rules he finds extortionate, and by a public that understands, at last, that the architecture of its goods is a political and moral question and not merely a technical one. An accident cannot be argued with. A choice can. That the matter has become arguable is itself the advance.
VIII. Why this changes the nature of information technology
Stand back from the particulars and the magnitude of the shift comes into view. For the whole of its history, information technology rested on one foundational fact, so pervasive that it was rarely stated: that to compute is to copy, and to transmit is to duplicate. Every consequence flowed from it — the abundance and the piracy, the collapse of distribution costs and the collapse of the creator’s bargaining power, the law’s long and losing war against the perfect copy, the culture’s giddy sense that information was free because it could not be made otherwise. The free copy was not one feature of the digital world among many; it was the nature of the material, the grain of the wood, the thing every other thing was built upon.
To give the bit a body is to revise that foundation, and a revised foundation changes everything built upon it. Information may now be not merely shared but given — parted with, so that the giver is without it. It may be not merely accessed but owned — held in the strong sense, scarce and singular and transferable. Its rules may be not merely imposed by the medium but authored by its maker — composed deliberately, enforced by construction, chosen rather than suffered. This is not an incremental improvement in digital rights management, a phrase that has always smelt of the lock and the lawyer; it is a change in the metaphysics of the digital object. The thing that could only ever be a taper, lighting infinite tapers without dimming, can now, at its maker’s election, be made a candle that is either given away or handed over but not both — a thing with the rivalry, the excludability, the singular identity that we thought belonged to matter alone.
The two revolutions, taken together, describe the shape of a more serious digital world. The first taught us to deal honestly with no dealer — to transact, to play, to agree, among parties who need not trust one another and need appoint no master between them. The second teaches us to own honestly with no copy — to possess, to give, to sell, to make scarce by choice the things that the medium had made promiscuously abundant. Between them they dissolve the two great disabilities of the early digital age: that nothing could be trusted without a trusted party, and that nothing could be owned because everything could be copied. What remains, when both are abolished, is a digital realm that is no longer the lawless and weightless place of its infancy — the realm of the free, the forgeable, and the unownable — but a place where things can be made as definite, as scarce, and as genuinely one’s own as anything in the world of matter, and where the maker of a thing decides, deliberately and in the open, what kind of thing it shall be. That is not a new application. It is a new nature for the material, and a civilisation will build differently upon it than it built upon the old.
IX. Coda: the wealth of the scarce thing
It will be said that there is something cold in all this — that to make the infinitely shareable scarce again is a kind of vandalism against abundance, a fencing of the commons, a retreat from the generosity of the free copy. The objection mistakes the nature of what was abundant. The free copy gave us an abundance of instances and a poverty of everything that matters about a possession: nothing could be owned, because nothing was scarce; nothing could be given, because nothing was parted with; the maker could not be paid, because his work escaped his keeping the moment it was made; and the very word “mine,” applied to a digital thing, was a polite fiction that everyone agreed not to examine too closely. That is not wealth. It is a crowded room mistaken for a full one. To engineer scarcity, deliberately and with one’s limits honestly stated, is not to impoverish the digital world but to give it, for the first time, the possibility of riches — of things owned, things given, things made and sold and inherited and treasured, things that carry the weight of consequence because they can be lost.
There is a beauty in a thing that obeys the law its maker wrote into it, a beauty of exactness and intention, of an object that is precisely what it was meant to be and does precisely what it was meant to do — neither more, claiming no power it lacks, nor less, betraying no promise it made. And there is a morality in it too, the same morality that ran through the honest game: that value should pass by consent and on stated terms, that the maker should be paid for what he makes and the owner secure in what he owns, that the rules should be visible and chosen rather than hidden and suffered. These are, once again, the two faces of a single achievement, and it is the achievement of the reasoning mind doing the one thing it does better than any sermon — not exhorting the world to behave as if digital things could be owned, which would change nothing, but building the structure in which they actually can be. The free copy was the condition of the digital world’s infancy: a glorious, lawless, weightless abundance in which nothing could be held and nothing given. To give the bit a body — to make it scarce by choice, transferable by design, governed by the law its author chose — is to let the digital world grow up. A thing you can give without losing was never yours. We have learned, at last, how to make a digital thing one can give and be without — which is to say, how to make a digital thing one can truly own. That is the whole of this second revolution, and it is enough.
References
[1] T. Jefferson. Letter to Isaac McPherson, 13 August 1813. (On the non-rivalrous nature of ideas: “He who receives an idea from me, receives instruction himself without lessening mine; as he who lights his taper at mine, receives light without darkening me.”)
[2] P. A. Samuelson. “The Pure Theory of Public Expenditure.” Review of Economics and Statistics, 36(4):387–389, 1954.
[3] H. Demsetz. “Toward a Theory of Property Rights.” American Economic Review, 57(2):347–359, 1967.
[4] W. Diffie and M. E. Hellman. “New Directions in Cryptography.” IEEE Transactions on Information Theory, IT-22(6):644–654, 1976.
[5] R. L. Rivest, A. Shamir, and L. M. Adleman. “A Method for Obtaining Digital Signatures and Public-Key Cryptosystems.” Communications of the ACM, 21(2):120–126, 1978.
[6] D. Chaum. “Blind Signatures for Untraceable Payments.” In Advances in Cryptology — CRYPTO ‘82, pp. 199–203, Plenum Press, 1983.
[7] N. Szabo. “Formalizing and Securing Relationships on Public Networks.” First Monday, 2(9), 1997.
[8] L. Lessig. Code and Other Laws of Cyberspace. Basic Books, 1999.
[9] S. Nakamoto. “Bitcoin: A Peer-to-Peer Electronic Cash System.” Self-published white paper, 2008.
[10] F. McKeen, I. Alexandrovich, A. Berenzon, C. V. Rozas, H. Shafi, V. Shanbhogue, and U. R. Savagaonkar. “Innovative Instructions and Software Model for Isolated Execution.” In Proceedings of the 2nd International Workshop on Hardware and Architectural Support for Security and Privacy (HASP), 2013.
[11] I. Anati, S. Gueron, S. Johnson, and V. Scarlata. “Innovative Technology for CPU Based Attestation and Sealing.” In Proceedings of the 2nd International Workshop on Hardware and Architectural Support for Security and Privacy (HASP), 2013.
[12] V. Costan and S. Devadas. “Intel SGX Explained.” IACR Cryptology ePrint Archive, Report 2016/086, 2016.
[13] W. Entriken, D. Shirley, J. Evans, and N. Sachs. “ERC-721: Non-Fungible Token Standard.” Ethereum Improvement Proposal 721 (Final), 2018.
[14] J. Van Bulck, M. Minkin, O. Weisse, D. Genkin, B. Kasikci, F. Piessens, M. Silberstein, T. F. Wenisch, Y. Yarom, and R. Strackx. “Foreshadow: Extracting the Keys to the Intel SGX Kingdom with Transient Out-of-Order Execution.” In 27th USENIX Security Symposium (USENIX Security 18), pp. 991–1008, 2018.