A Cypherpunk’s Manifesto

Eric Hughes addressed the Cypherpunks community, whose early discussions survive in mailing-list archives. The essay bears the precise date 9 March 1993. Internet services have changed since then, but the basic question remains: must a provider know everything we do through its service? This is a different text from The Crypto Anarchist Manifesto, which Timothy C. May wrote in 1988. May placed greater emphasis on predicting changes in political power. Hughes focused on privacy and the work needed to protect it. The two essays share an intellectual setting, not an author or identical content. [1] [2] [3] [5]

You might tell a friend about a family worry, while simply telling a colleague that you need to leave early. Both receive truthful information, but different portions of it. This helps explain Hughes’s distinction between privacy and complete secrecy. You need not hide your entire life to have a reason to protect a particular conversation. Privacy concerns boundaries between situations and people. When messages, purchases and contacts are automatically combined into a single profile, those boundaries disappear. Information shared for one purpose can then reach someone for whom it was never intended. [1]

Imagine a shop selling a downloadable file. To complete the purchase, it needs to identify the order, verify payment and enable the download. Further information might be used to issue a receipt, for example. Even then, it makes sense to distinguish that purpose from what is technically necessary to deliver the file. A customer can ask: why should I hand over my phone’s entire address book for this one service? Information a service never collects cannot later leak from its database. Limiting collection is a different and important form of protection from merely promising not to misuse information already gathered. [1] [10]

Return to the personal message. Encryption can hide its contents while the service still records the sending time, a network address or the accounts involved. These accompanying details are called metadata. Their extent depends on the system; not every service collects the same things. A record of regular contact can be sensitive without a single sentence being read. So the privacy questions go beyond whether an app encrypts: who holds the keys, what is stored, and what can be connected? Here, keys are long numerical values that the app uses to encrypt a message or restore its readable form. Protecting content and limiting traces of communication are related but distinct tasks. [8] [4]

Encryption turns a readable message into data that only an authorised recipient with the necessary key should be able to recover. Properly implemented end-to-end encryption protects the contents even from the server carrying them. A digital signature tackles a different problem: it lets someone verify that signed data have not changed and that the signature corresponds to a particular key. It does not by itself conceal the message or establish the writer’s legal name or honesty. A nickname can remain associated with a key over time. Separating these jobs helps explain why the manifesto discusses a collection of tools, rather than one magical anonymity button. [4] [9]

Two services might display the same sentence: we do not read your messages. One stores readable copies and restricts staff access; the other, using end-to-end encryption correctly, receives encrypted data without the keys to their contents. These involve different assumptions about trust, despite identical advertising. The second service can still have weaknesses in its app, backups or checks of the person at the other end. Hughes’s reluctance to rely solely on large organisations’ goodwill leads to a useful technical question: what exactly could the provider learn if it changed its mind or someone compromised its server? [4] [8]

The short phrase turns attention from wishing to making tools. Usable software can change everyday experience: a person sends a protected message without doing the cryptographic calculations by hand. Source code is the text of a program that developers write and edit. Publishing it lets others study the design and look for mistakes. Availability alone, however, does not prove that anyone has reviewed it thoroughly or that a downloaded application matches the published version. Evaluating software therefore also means considering distribution, updates and how discovered flaws are handled. The manifesto calls for creation and sharing; the quality of each result still needs its own assessment. [1] [11]

Your friend needs to be able to receive the encrypted message. If the tool is too difficult, she may move the conversation to a less protected channel. Once she reads it, she can repeat it or take a photograph. Protection in transit cannot itself prevent that. Hughes therefore ends by stressing cooperation: distributing software alone is not enough. His optimism about code spreading worldwide is a belief about the future, not a guarantee that every service will always remain available. Alongside algorithms, privacy has a human side: whom we talk to, how we use tools and how we handle other people’s information. [1] [8] [10]

A message can travel safely across a network and still reach a stranger who reads an unlocked phone or takes control of the device with malicious software. A conversation’s backup may have different protection from the transmission itself. Even a correctly verified signature might belong to a fraudster who faithfully signed a false claim. Such cases do not make encryption worthless; they show which part of the problem it solves. A useful approach starts by asking what we are protecting, from whom and how they could obtain it. Only then do we choose tools. The manifesto encourages that thinking but does not replace a concrete security assessment. [8] [9] [10]

Bitcoin allows value transfer and shared verification of rules without a central keeper of the payment ledger. Its transaction history is public, however. When a merchant connects a payment with an order and a customer’s name, using Bitcoin does not erase that connection. Satoshi Nakamoto’s whitepaper discusses privacy and the risks of linking transactions; a Bitcoin address is not an automatic cloak of invisibility. The connection to Hughes lies in the questions and the approach to building tools, not in his manifesto having technically designed Bitcoin. Reading both works, it pays to examine concrete properties: who sees what, whom we must trust and what we can verify ourselves. [6] [7]