Blockchain Pioneer Decrypted: From the Origin of NFT to the Birth of Bitcoin

Interview: Jason Bailey

Interviewees: Scott Stornetta and Stuart Haber

Original article compiled by: Luffy, Foresight News

When Scott Stornetta and Stuart Haber invented blockchain, they were thinking of something similar to NFTs, not digital currency. Jason Bailey, founder of the art and technology blog Artnome.com, had an in-depth conversation with Scott Stornetta and Stuart Haber about the vision of blockchain in its early days over thirty years ago and its subsequent evolution.

Translator's note: Scott Stornetta and Stuart Haber implemented the blockchain architecture with code in 1991. It was put into commercial use in January 1995 and has been running ever since.

Jason Bailey: I often introduce you two as my good friends who "invented blockchain". Then, I usually see a look of disbelief, "No, it was Satoshi Nakamoto who invented blockchain", even among those who have been studying Bitcoin and cryptocurrencies for years. Can you help people better understand the contributions you have made and how these contributions have become the foundation for Satoshi Nakamoto to build the Bitcoin network?

Stuart Haber: Okay, let me tell this story by reviewing about 30 years of history. From my perspective, this marked the beginning of blockchain. It all happened in 1989, when Scott Stornetta and I were young scientists at Bellcore.

I am a cryptographer. Scott has just joined Bellcore. He is looking for a solution to ensure the integrity of digital records can be proven, guaranteed, and maintained through certain programs or algorithms. Scott strongly suspects that cryptography will play a role in this. Therefore, we wrote several papers with Dave Bayer and created an architecture to solve this problem.

Scott Stornetta: Stuart and I have formed a unique collaborative style characterized by dynamic exchange of ideas. I tend to be fully focused on thinking about such issues. The challenge is that I am not familiar with the underlying mathematics or cryptography that can be used to solve this problem. I have a question, but I don't know how to find a solution. This knowledge gap has always been the driving force behind our productive collaboration.

SH: For those familiar with these concepts, digital signatures and cryptographic hash functions were proposed, implemented, and fully understood in the fall of 1989. These tools presented a relatively simple solution to a problem, requiring a trusted entity (whether an individual, software, or hardware) to ensure the integrity of records within a specific domain. For many, this solution was considered satisfactory. However, Scott and I were not satisfied with this solution because we sought a solution that did not require trust in any party, nor did it require trust in as few individuals, entities, and mathematical assumptions as possible.

We eventually developed a solution, which can be explained using a metaphor: "digital fingerprint". In fact, every blockchain project in the world relies on cryptographic hash functions, mathematical algorithms with input and output processes that can effectively generate digital fingerprints for files. When you apply this process to the same file multiple times, you will always get the same fingerprint output. This is an efficient process, even for large files, especially when we forget to close it and let it run continuously.

Now, another important property of cryptographic hash functions is that when you obtain two different files and calculate their fingerprints, you will get two different results. In fact, even if you make a small change to a file, such as changing 0 to 1, the final fingerprints of these two different files will change significantly and unpredictably. For example, when it comes to financial records, even minor changes can have a profound impact on the meaning of the file. For instance, changing the first digit from 0 to 1 may be more advantageous for one party than the other. Therefore, the metaphor of fingerprint recognition is appropriate here: two different fingers have two different fingerprints.

Another important attribute of fingerprints is that they do not reveal detailed information about me. You cannot discern my height, hair color, or even whether I have more than one finger from my fingerprints. Similarly, the fingerprint in cryptographic hash functions is just a series of numbers and letters that does not reveal any information about the original file. However, if you have the fingerprint and claim to have a file that matches it, you can easily verify its authenticity by "retrieving its fingerprint" again. This ability of digital fingerprint recognition is called cryptographic hash function, and these concepts were even established at that time.

JB: Allow me to summarize to ensure my understanding is correct. Hashing or fingerprinting is a concept that is easy to understand. Your team's goal is to prove the integrity of records and that they cannot be tampered with. Are these hash values or fingerprints similar to blocks in blockchain to some extent? Have you found a way to link these fingerprints into a blockchain?

SS: You're right. As Stuart mentioned, we realized that hash values can not only represent files more concisely and efficiently, but also combine them (and build each block as a Merkle tree), and then link these blocks together, all using the same hash function. With Dave Bayer's involvement, we were able to link record groups in such a way that each participant and their documents became holders of record proofs, acting as early nodes. This means that all records are uniquely connected and widely distributed, containing many of the basic elements that Satoshi Nakamoto later created Bitcoin with. We won't take anything away from Satoshi Nakamoto and his creation, but we see Bitcoin as an application built on top of early blockchain technology. It's worth noting that Satoshi Nakamoto explicitly referenced all publications related to the foundational work we were involved in. Our work is cited three times in the Bitcoin white paper, which cites a total of eight external references, of which we occupy 3/8.

JB: So, how can we help all cryptocurrency enthusiasts who consider Satoshi Nakamoto as the founder of blockchain bridge the gap between your work in the late 1980s and early 1990s and today's Bitcoin?

‍ SH: When you mention your contributions to blockchain to the general audience, they usually immediately associate it with Bitcoin or cryptocurrencies in a broader sense.

Scott and I were not trying to invent electronic currency. In fact, the cryptography community had already been working on creating purely digital currency since the 1980s. Our focus is broader: we are truly concerned with the integrity of all records, including electronic records.

SS: This includes financial records, but our scope extends to every important record ever created, and we believe all of these records can be registered on the blockchain.

SH: Since it is impossible to predict which records will become important in a few years, why not include every record ever created?

JB: Essentially, do you identify more with the concept of NFTs rather than cryptocurrency? When we view NFTs as tools for verifying art, contracts, patents, and various applications, it seems to align with your original goals.

SH: That's right. When discussing algorithmic methods for recording numbers, we used the term "provenance". We focused on various types of records. However, when Satoshi Nakamoto's goal was to establish a digital currency system and needed a way to ensure the integrity of financial transactions within the system, he directly adopted our solution. The data structure of Bitcoin transactions accurately reflects the data structure of our timestamp system, which was implemented in experimental code in October 1991 and put into commercial use in January 1995.

The longest-running blockchain started in 1995 and is still running today. The timestamp is highlighted in red.

SS: I would like to reiterate Stuart's point. Essentially, our vision for blockchain is not the same as Satoshi Nakamoto's. Nakamoto introduced an innovation in the field of currency, but he needed a powerful record-keeping system. He seamlessly integrated this layer and built Bitcoin on top of it.

What I want to emphasize is that we will not ignore the contribution of Satoshi Nakamoto. On the contrary, he built Bitcoin on a blockchain connected by widely distributed Merkle trees, and openly acknowledged that this concept had already been invented. Then, he adopted the same cryptographic hash function or digital fingerprint to directly create the mining mechanism.

One interesting aspect is the recent popularity of ordinal inscription in Bitcoin. If people delve into the footnotes of the Bitcoin whitepaper, they will find that in our third joint paper, we hinted at the concept of using blockchain inscriptions or ordinals to create unique and non-fungible records. This is similar to the concept of NFTs today.

Your observation on cryptocurrency and NFT is spot on. In a way, we see NFT as a more significant long-term achievement of our initial goal.

It indicates that everything important, not limited to various cartoon primate poses, may ultimately need to be uniquely registered as NFT on the blockchain. Perhaps we should delve into our latest NFT series?

JB: I really want to hear more information, especially now that we have clarified your contributions and their continuity with Satoshi Nakamoto's subsequent creation of Bitcoin. It is fascinating that your invention of blockchain is more focused on NFTs rather than cryptocurrencies, which may surprise some people. How did you decide to collaborate with artists and use illustrated news as the art for your NFTs?

SS: The main challenge in the early days was to achieve universal consensus, which became even more difficult due to the lack of the World Wide Web or similar technologies. Our solution was to regularly create a snapshot (a fingerprint) of the blockchain and widely distribute it to prevent manipulation.

In order to achieve this goal, we choose to publish this snapshot every week in the national edition of The New York Times. This version is preserved in libraries and archives around the world. Imagine the tremendous effort required to tamper with a single element in the tamper-proof chain, it's like infiltrating every library in the world and changing its copy of The New York Times.

This is consistent with our approach to NFT collections. We released the initial 12 NFTs, each representing one of our 12 consecutive weeks of publications in The New York Times. We collaborated with an artist to curate weekly events, selecting some imaginative, historical, or noteworthy things and illustrating them.

Welcome to join the official BlockBeats community:

Telegram Subscription Group: https://t.me/theblockbeats

Telegram Discussion Group: https://t.me/BlockBeats_App

Official Twitter Account: https://twitter.com/BlockBeatsAsia