The full text has a total of 5933 words and the estimated learning time is 17 minutes
Photo: VCG / Visual China Group / Getty Image Company
In his new book "Facebook: Insider", Stephen Levi, a long-time science and technology reporter, described in detail Mark Zuckerberg's journey to transform the boring dormitory social network experiment into the world's largest social network business.
To commemorate the release of this new book, the author would like to share an excerpt from Levi ’s early book "Cryptography", which describes a person who runs counter to Facebook's data utilization and privacy violations. This is the story of Witt Diffie, who changed the way people think about encryption and paved the way for people to enjoy digital security today.
Bailey Whitfield Diffie (Bailey Whitfield Diffie), born on June 5, 1944, has a very independent personality. An old friend described him, "This kid had another lifestyle at the age of five." It was not until 10 years old that Duffy began studying. This is not because he has any illness, but he hopes his parents can read it to him. It seems that his parents did it patiently. Finally, in the fifth grade, Diffie finished reading "Space Cat" himself, and then read "The Wizard of Oz".
Later that same year, his teacher spent an afternoon explaining one thing that would accompany him for a long time: the basics of cryptography. Several decades later, Defie stated on the postscript 178 pages of the book, "Her name is Mary Collins. If she is still alive, I would love to find her."
Diffie found that cryptography is a pleasant secret expression. Its users collaborate to keep secrets while spying on the world. The sender converts the private information into another state, which is encryption, a mysterious language. Once the information is converted into garbled characters, it will frustrate potential eavesdroppers. Only those who have mastered the conversion rules can restore the chaotic information to the harmonious state of the original writing, that is, decryption. Those who do not have knowledge and try to decrypt information without a "key" are practicing the "cryptanalysis" skills.
Although Diffie did well in school, he did not show his true strength. In 1961, he scored high on the standardized test and entered MIT to study mathematics. He also began to learn computer programming, although Diffie now says that his original purpose was to escape conscription. Diffie accepted the job of Mitt Corporation, which as a defense contractor can shelter young employees from military service.
Diffie ’s team does n’t have to work at Mitt, but in 1966, they became regulars at Marvin Minsky ’s MIT Artificial Intelligence Laboratory. In the artificial intelligence laboratory, information enjoys the same status as air. The MIT Wizards do not have a software lock on the operating system.
Source: Pexels
However, unlike its peers, Diffie believes that technology should provide a sense of privacy. Diffie often discusses safety issues with his boss mathematician Roland Silver. Therefore, cryptography inevitably entered the discussion category.
Silver has a certain understanding of this field and explains that there is a lot to do behind it, especially behind the steep barriers established and maintained by government intelligence agencies. Diffie was very dissatisfied with this and rebuked Silver. He believes that cryptography is essential to human privacy! He suggested that perhaps those enthusiastic researchers in the public sector should try to open up the subject. He told Silver, "If we do it with our heart, we can rediscover many cryptographic materials."
Silver doubted this. "Many very smart people work at NSA," he said. NSA (National Security Agency) refers to the National Security Agency, the cryptographic bastion of the US government. In accordance with President Truman ’s top secret order in the fall of 1952, the National Security Agency was established, costing billions of dollars and operating entirely in the government ’s “black” area. Only those who can prove “need to know” are entitled to obtain knowledge. In the early 1970s, these were not discussed publicly. In Washington, insiders jokingly called the abbreviation NSA a NoSuchAgency "no such organization."
In 1969, as the funds ran out, Diffie finally left Mitt. He and his girlfriend moved to the western region, and Diffie went to work at John McCarthy's Stanford Artificial Intelligence Laboratory, where he began to think more deeply about privacy. At the same time, he was recommended to an assistant professor of electrical engineering named Martin Herman.
Herman was born and raised in New York and received his Ph.D. from Stanford University in 1969. His first job was at IBM research, where he developed a strong interest in cryptography. After leaving IBM in 1970, he accepted the position of assistant professor at the Massachusetts Institute of Technology, where cryptography was his research focus, and then went to Stanford. Herman resisted the temptation of most scientists in his field: working under the restrictions of the National Security Agency. After his first paper in the field of cryptography was published, his follow-up work has not seen any progress. Diffie joined. "This is the intersection of ideas," Herman said. Both Diffie and Herman firmly believe that the emergence of digital communications makes commercial cryptography an absolute necessity. Herman hired Diffie as a part-time researcher.
In March 1975, a dull government document impacted the two Stanford researchers. A Federal Register published by the National Bureau of Standards (NBS) proposed something rarely mentioned in public literature: a new encryption algorithm, this algorithm comes from the cooperation between IBM and the government, called the Data Encryption Standard (DES) .
Although Diffie and Herman regarded the data encryption standard as a taint of IBM and the US government, and even considered it a fraudulent method, its advent became a gift from God. Through sorting out the existing technical data and speculating on unpublished standards, Diffie and Herman determined their own efforts. Since Diffie heard the first report on government standards at Louie's, a Chinese restaurant where Stanford computer wizards gathered in 1974, he has been thinking about the possibility of NSA having some kind of "trap door". This gave him a deeper consideration of the concept of "trap door": Can the entire encryption scheme be built around a trap door?
Designing such a system poses considerable challenges because it must resolve a basic contradiction. Trapdoors provide those who have the appropriate knowledge with a way to bypass security measures and quickly obtain encrypted information, which seems to be very effective. But the idea of using trapdoors in security systems seems to be too high risk, because a cunning intruder may find a way to use it. This is the same problem faced by trapdoors in reality: if the enemy cannot find it, it can be hidden with it. But if found, the enemy will know where to find you.
This contradiction makes the prospect of design traps daunting. After all, the strongest encryption system has been fine-tuned in all aspects to prevent content leakage. Tampering with its internal structure and inserting backdoors (ie holes) can easily create many unexpected weaknesses. When Diffie explained this to Herman, they both came to the same conclusion: such a system might be impractical. But Diffie still thinks this is interesting and adds this question to the list called "Ambitious Cryptography Questions."
One day, Diffie and Herman brought a Berkeley computer scientist named Peter Blatman to join their informal seminar on encryption held at the school. Later, Blatterman mentioned that his friend was working on an interesting question: when two people in a conversation never had contact, how to conduct a safe conversation through an unsafe line? Obviously, if the two did not know before, there would be no chance to exchange keys before the private conversation.
This is actually another way to express Diffie ’s major problems for many years: is it possible to use encryption technology to protect a huge network from eavesdroppers and evict eavesdroppers from the network?
How to create a system that allows people who have never met to talk to each other safely? In the system, all conversations can be conducted with high-tech efficiency, but they must be protected by cryptography. In the system, when receiving an electronic message from someone, how to ensure that the message is the person whose reply address appears.
During the exploration process, Diffie worked hard to collect information in an environment where almost all information was classified, and eventually got something beyond anyone's expectations: one-way functions, password protection, recognition of friends or enemies, trapdoors. All of these are answers to privacy questions. Diffie knew that coordinating the different protections provided by these different systems was crucial to his exploration.
One afternoon, Diffie suddenly realized one thing: designing a system that would not only provide all the functions of the one-way authentication scheme he recently envisioned, but also provide encryption and decryption in a novel way. The system will solve the problem of untrusted administrators and even other problems.
Diffie will separate the keys.
Source: Pexels
In the historical background of cryptography, Diffie's breakthrough itself involved an absolute heresy: the public key. Prior to this, there was a set of seemingly inviolable rules in encryption. This virtual dogma made people ignore the abyss after the secret was revealed. One of them is that the same key can both disturb information and become a tool for decrypting messages. This is why the key is considered symmetrical.
This is why it is so difficult to keep these keys secret: the tools obsessed by eavesdroppers (decryption keys) must be passed from one person to another and then stored in two places, which greatly increases the chance of leakage. However, Diffie's brain is full of information that has been carefully collected and considered in the past five years. He envisioned a new possibility that a key pair could be used instead of a separate key. This tried and tested symmetric key will be replaced by a dynamic key. Although people can steal plain text information (perform tasks that cannot be read by outsiders), secret traps are built into the information. The other part of the key pair is like a latch, which can open the trap door and let the holder read the message. This is the beauty of the plan: yes, the second key, the one that opened the trap door, must of course be protected from being stolen by potential eavesdroppers. But its counterpart, the key that actually performs encryption, is not secret at all. In fact, people would not want it to be secret, but would be happy to see it widely distributed.
Now, the idea of using fully publicly exchanged keys to ensure privacy is completely unintuitive, which may seem strange on the surface. But mathematical methods that use one-way functions may work. Defie realized this, and in a flash, he figured out how to use a one-way function to do this.
This is the answer. Since then, everything in the field of cryptography has changed.
First, by proposing an alternative to using a single symmetric key system, Diffie solves the problem that has actually troubled cryptographic systems for a long time, and almost no one has successfully solved it: assigning the secret key to future secret message reception Without being leaked or eavesdropped. If you are working in a military organization, you may be able to protect the distribution center that handles symmetric keys after solving this problem, although God knows that mistakes can occur even in the most important operations. However, if such a center enters the private sector when the public needs to use it, not only will inevitably bureaucratic robbery occur, but it will also bring continuous threats and risks. Imagine that you need to crack encrypted messages, so would n’t the place where all the keys are stored give the thieves the opportunity to steal, bribe or other coercion?
But with the public key system, everyone can generate a unique key pair by themselves. It consists of a public key and a private key, and no outsider can access the key part. This allows for private communication.
The principle is here: Suppose Alice wants to communicate with Bob. Using Diffie's philosophy, she only needs Bob's public key. Alice can ask Bob for the public key or get it from a phone book-style public key index. But it must be Bob's personal public key, a long bit string that only Bob in the world can generate. Then, through a one-way function, Alice scrambles the message using the public key, so that only the private key (the other half of the unique key pair) performs decryption calculations. (Therefore, the secret key is the "trap door" in the one-way trap door function considered by Diffie.)
Therefore, when Alice sends encrypted information, only one person in the world has the necessary information to reverse the calculation and decrypt it: Bob, the holder of the private key. For example, someone wanted to know what Alice said to Bob, and intercepted the disrupted news. But this is no big deal. Unless the snooper can obtain the unique correspondence of Bob's public key, which is the tool Alice uses to obfuscate the information, the snooper will not get more obfuscated information. Without a private key, it is too difficult to reverse the mathematical encryption process. Remember, making an error in a one-way function is like trying to piece together a powdered plate.
Of course, Bob has no problem reading the information designed for him. He has the correct password part of the key, and can use the private key to instantly decrypt the message.
In short, because he is the only person who holds the key pair on both sides, Bob can read the information. When trying to crack the message, those who obtained the public key had no advantage. When it comes to encrypting messages, the only value of possessing Bob ’s public key is actually changing the message to Bob ’s language, which is a language that only Bob can understand (because it owns the half of the key pair secret).
This encryption function is only part of Diffie ’s revolutionary concept and is not necessarily the most important feature. Public key encryption also provides the first effective method to truly verify the identity of the sender of an email. As Diffie envisioned, trapdoors have two directions. If the sender scrambles the message with someone's public key, only the designated receiver can read the message. However, if the process is reversed-someone messes up some text with their own private key, then the resulting ciphertext can only be cracked by using a single public key that matches it. The significance is that if you get a message from someone who claims to be Albert Einstein and want to know if it is really Albert Einstein, then there is now a way to prove it , This is a kind of mathematical touchstone. You can find Einstein's public key and apply it to encrypted ciphertext. If the result is plain text instead of garbled characters, then you will definitely know that this is Einstein ’s information, because he holds the world ’s only private key, and it can create a message that can match the public key to decrypt it.
Source: Pexels
In other words, applying a personal key to a message is equivalent to digitally signing. But this is different from the kind of signatures written on bank checks, divorce certificates and baseballs. No one can forge John Hancock's digital signature. Without a private key, a thief cannot forge a signature.
Forgers will not monitor the phone line and wait for the digital signature of the prey to appear before stealing, so as to use the digital signature to create forged documents or intercept future messages. In fact, the digital signature is not used as an attachment to the attached documents or letters. Instead, it is deeply intertwined with the numbers that make up the actual content of the entire message. Therefore, if a file containing a digital signature is intercepted, the eavesdropper cannot extract the digital signature from it, and therefore cannot decrypt other files encrypted by the digital signature.
This technique can also ensure the authenticity of the entire document. It is impossible for others to change the subtle but important part of the digital signature document (for example, change "I will not pay the debt for my spouse" to "I will pay off the debt for my spouse", and the sender is not aware In the case of love, the signature was retained). If the message is digitally signed with the private key but not encrypted, the thief may intercept it, decrypt it using the sender ’s widely distributed public key, and then change it in clear text. But what about that? In order to resend the text with the appropriate signature, the forger will need the private key to repair the signature on the entire document. Of course, the private key will not be available and will be owned only by the original signer.
It is also easy if the person who sent the signed message wants to keep the rest of the content secret in addition to the signature. If Mark wants to send an order to his banker, Renoir, he first signs the request with the private key, and then encrypts the signed message with Renoir's public key. Linuoer will receive two encrypted messages: one for privacy and the other for identity verification. She will first apply her private key to decrypt messages that only she can see. She will then use Mark ’s public key to unlock a message that she knows only Mark can send.
Digital signatures have another advantage. Since no one except the person with the encrypted private key can produce a digitally signed message, the signer cannot deny its role in generating the document for legitimate reasons. This undeniable function is equivalent to the electronic seal of a notary office.
This is the first time in history that people can conceive of various formal transactions (contracts, receipts, etc.) through a computer network without having to be present.
In short, Diffie not only found a way to protect privacy in the digital communications era, but also implemented a new form of business. This kind of electronic commerce may not only match the current agreement in commercial transactions, but also possibly exceed the current agreement.
What's more impressive is that his breakthrough was entirely outside the scope of the authority of government agencies, and contained the most trivial details of the most obscure encryption system.
Crypto by Stephen Levi
Leave a comment
Let's share the dry goods of AI learning and development together.
Welcome to the full platform AI vertical self-media "reading core technique"
(Add editor WeChat: dxsxbb, join the reader circle and discuss the newest artificial intelligence technology together ~)
