Why Nostr matters: Freedom from being ruled by servers

Author: Digi Monkey; Source: Byte Yuan CKB

The following content is shared by Digi Monkey, the author of Nostr client flycat.club, at the 2023 Nostraia HK event. Original link:

https://flycat.club/post/[email protected]/why-nostr-is-important

Thanks to Nostr for giving us such an opportunity to get together and share some ideas about social networks. I am Digi Monkey, the author of Nostr client flycat.club. Flycat is a Nostr client that provides functions such as Relay Group Switching / long text + short text information flow / Community / JoyID login/ MetaMask login.

Today I want to talk about Nostr in my eyes and why it is important for cyberspace.  

Back to 1996

I would like to share a passage first: 

We will create a civilization of the Mind in Cyberspace. May it be more humane and fair than the world your governments have made before. 

Some friends may know the source of this sentence. It is excerpted from the end of the "Declaration of the Independence of Cyberspace" and is an article published on the Internet on February 8, 1996 by a man named John Barlow. In 1996, you can imagine that the Internet was still a very new thing. For reference, Netscape Navigator was released in 1994. So at that time, the concept of the Internet, or cyberspace, was still very vague to people at that time. However, we can see that the declaration just now actually used a very firm tone to depict a very idealistic utopian vision of the Internet community. In the Declaration of the Independence of Cyberspace, it advocates two basic ideas: 1. Cyberspace and the physical body are dualistic and completely independent of each other. There are no borders/no discrimination/no matter in cyberspace. 2. Cyberspace does not need to rely on the governance of the real world, and is not bound by mandatory laws, but forms order and its own social norms through unwritten "codes" (codes/morals/ethics/self-interested ideas/common welfare).  

Today we are talking more about social networks in a broad sense, which refers to such a cyberspace, a new digital continent, the entire online life of the Internet, a large community where people can exchange information and services.  

This is the same thing that is discussed in the manifesto. We can read the original text carefully to see how people expected such a new thing at that time. Here are some excerpts from the original text:  

"The online world consists of information transmission, relationship interaction and thought itself... Our world is both omnipresent and ethereal, but it is by no means a world where entities exist.

We are creating a world where everyone can join, without privilege or prejudice based on race, economic strength, force or place of birth.

We are creating a world where anyone, anywhere, can express their beliefs, no matter how strange, without fear of being forced to remain silent or conform.

Your legal concepts and contexts of property, expression, identity, and migration do not apply to us. All of these concepts are based on material entities, and material entities do not exist here.

You can see that the imagination at that time was very idealistic. Its vision of governance was also vague. Compared with today's online world, which has become our daily, real online world (Internet), it is very different and has forked. There are even many Web3 / Blockchain communities that would say that the Internet has taken a detour and deviated from its original intention.  

Today's Cyber ​​World

The two basic ideas in the "Declaration of the Independence of Cyberspace" mentioned above are violated by the current Internet:

1. The reason why the Internet is so powerful today is that it is stitched together with reality, not a binary relationship that is independent of each other.For example, many o2o services, such as food delivery, taxis, accommodation, etc., have completely surrounded people's lives from head to toe. The Internet has invaded reality and even guided real life in turn. In the early days, the Internet was more like a new continent of pure thought. Everyone was optimistic that a free home could be created for human thought alone.  

2. This home is not subject to the management of real laws and the state, and is purely autonomous. Today, this point has also been overturned. Today, all countries are actively managing the Internet and issuing various telecommunications laws, mainly following the principle of territorial management: for an Internet service, the subject to which it belongs, the location of the server, and the location of the data storage, where these places are, the government and local laws and regulations there will correspondingly regulate the Internet services hosted on these lands. Here we will not discuss whether the overthrow of these two basic concepts is right or wrong. We will not discuss whether what the "Declaration of the Independence of Cyberspace" advocates is necessarily right. We only say that there is a big gap between the ideal and the actual reality, and changes have occurred. The question I am interested in is, how did this change happen? 1. From the perspective of the mainstream melody, it is a kind of cyberpunk occurrence. The development of large companies/governments/technology itself is out of people's control and has created a reality of "high tech, low life".  

Example 1, the delivery guy is a bounty hunter in science fiction, hired by a system and living under the management of the algorithm. Many people want to escape, but more often they need to rely on it.

Example 2, life on a purely spiritual level, closer to the part on social networks, and the same is true for people's lives on Twitter today. You don't have the most basic rights on Twitter, that is, the ownership of the account. Users only have temporary use rights, not to mention the right to speak freely, all of which are in the hands of Musk and Twitter.  

2. From another perspective, we look at it from the perspective of machines and humans. I think this change is also a power war between humans and machines.  

a civilization war between machines and humans 

The machines here refer more to servers. A service on the Internet, especially a platform-based product, usually consists of two parts: a server and a client. The server is controlled by private enterprises or developers and can only be accessed through authorized APIs. The client is the software used by users locally, whether it is accessing the web through a browser or an App on a mobile phone. Users use the client to access the server through authorization, read and write some data, perform certain calculations, and consume a service.  

The server holds the greatest power in this process: 

• ID (name/identity/account, "who am I, how do I prove I am me", and a byproduct: relationship chain)

Identity ID is what you request to register with the server, asking it to give you a name. If it is willing to give it to you, it will give it to you. If it is not willing to give it to you, it can also give your identity to another person. For example, Weibo user names cannot be repeated, so who should be given a good name? The server has the final say. At the same time, because the server can decide "who you are" and who your friends are, the so-called social relationship chain is often a byproduct of identity ID and is also controlled by the server. The most typical example is WeChat, and the address book is WeChat's most valuable asset.  

• Data (means of production — where is the data stored/who has the right to take over/whether the data can be forged/who owns the ownership/right to use)

Although most of the data is produced by users, the data is stored on the server, and who produces the data/whether the data can be forged is also endorsed by the server. In most cases, you only have the right to use the data you produce, and the ownership belongs to the platform.  

• Transaction (production relationship — who can provide services/consume services, what trading rules to abide by/whether commission is required)

The rules are written in the server, and the server can change the rules at any time because the server also owns the data, and the modification of the rules often involves the use/acquisition of data, which needs to be adjusted, and it can change it as it wants.  

We can see that the process of the server providing services to users is very similar to the management of people by a country.  

Registering an account on Twitter is equivalent to registering your household registration at the Public Security Bureau. The Public Security Bureau will issue you a string of ID numbers, your ID, and then the ID card is your pass to prove yourself. Only with an identity can you rent a house, produce, and consume in this country. The whole process will also be accompanied by social rules such as laws and morals to constrain your behavior.  

There are also similar things like "User Terms of Use" on Twitter, which are equivalent to laws in the real world. If your account has certain behaviors that are not allowed, Twitter will revoke your account. But this set of online things is much worse than offline laws. Citizens of a country have the right to participate in the formulation of laws, and there are also judicial/lawyer systems that give citizens space for debate and ensure procedural justice, but this is completely absent online. Twitter has the final say. So we need to realize that in cyberspace, people's basic rights, the so-called "human rights", are actually more lacking than in the real world. If people in the real world are already modern civilizations and have entered the stage of sovereign states, then in cyberspace, we are probably still in the era of slavery or feudalism. The server is the king, monarch and emperor there. What role does Nostr play in this change? Nostr has a very special positioning. I like to use this diagram to look at its positioning. From left to right, there are the two extremes of "centralization" and "decentralization". Nostr is probably in the middle and a little to the right: Note that in this diagram, we are actually describing and talking about different types of services provided in cyberspace. It is a way of doing things, or a choice of technical architecture, and has nothing to do with offline political identity. Left and right are just a metaphor. Right Wings (centralization) Most of the centralized extremes use the model of maximizing server power that we mentioned above. We can label them as right-wingers and conservatives. They have the following components: Commercial companies (mainstream forces, large companies/startups) Individuals/small groups (geek self-hosting/indie developers) Governments (government portals/services provided by social organizations/online government affairs) Right Wings builds a lot of servers, and lots of machines, and we live under the rule of those machines. It is worth noting that there are also some good independent developers on the right wing, who often provide some unique software and services in the form of handicrafts. However, the model of maximizing server power is still adopted, that is, the developer has the final say.  

Take another example, https://tilde.town/, this is a community built on a Linux server. The initiator provides the server resources. You can apply to the community to join, get the ssh credentials to access this public server, and then access the server to write, draw, and make things on it. The whole community can share the creations of residents with each other.  

This community is a typical rule of man. The application for entry is manually confirmed by the initiator. If you behave badly, he can also kick you out directly. The population of this community is now close to 1,000 people. It is a very small and beautiful community, but it is still the model we mentioned above.  

So many times, the right-wing approach is also reasonable. As long as the residents of the community agree to the rule of the server and the community development is not hindered, it is not necessary to replace this centralized model with a decentralized one. Centralized services have their reasonable existence, which is why I disagree with the view that "blockchain eats everything".

Left Wings (decentralization)

Left wing refers to the extreme approach of decentralization. They include the following forms:

• p2p network (voluntary nodes, Bittorrent/Tor/SSB)

• blockchain (code is mandatory guidelines, requires incentives & consensus)

• free software (no services provided, open source/donation) 

Left Wings actually wants everybody to run their little homebrew machines, and unit all the devices from people under one global cyber law. That is the code of the law, the so-called consensus layer in the blockchain network. 

Left-wing radicals are also a relatively important force at present. Taking blockchain as an example, how does the left do it? Basically, they hope to create a fair system in which everyone can participate. The rules of this system are very strict. For example, the total number of bitcoins is 21 million. Each one needs to be mined and calculated, and any node needs to abide by the same set of code rules. How to modify and upgrade the rules of this set of code, and there are rules about modifying this set of rules. For example, some blockchains use DAO to vote to decide whether a proposal can be included in the consensus and become a new rule. Of course, Bitcoin may use more off-chain governance, and each system upgrade may mainly depend on the adoption and support of miner nodes. So the blockchain is actually creating a whole set of very idealized systems, which is like the very detailed system of ancient Greece where all people participated in political life. In contrast, Nostr does not have such ambitions. It is more of a loose form, with only a few core rules.You can do whatever you want with the rest. It may be closer to modern liberalism. There is a clear line between personal life and political life. Outside this line, the law does not need to interfere at all, and has no right to interfere.

We can take a closer look at the difference of Nostr.

Nostr: A more pragmatic golden mean

Nostr's protocol is minimalist and its positioning is special. It is neither left nor right, but chooses a more moderate position. We also compare from three aspects: ID (identity), data (means of production), and transactions (production relations).  

• ID (identity): It is just a pair of public and private keys, which are controlled by the user. The server cannot deprive the user of his identity, it cannot make "I am not me", only "I" can prove "I am me".

• Data (means of production): Nostr defines the format of data (a very simple JSON) and propagation (communication standard between server and client)

• Format of data: The same standard needs to be followed, and the most important provision is that any message sent must be signed.

• Propagation of data: hardcode uses websocket for server/client communication and defines the basic communication format. Why hardcode? More conducive to boostrap, pragmatic approach.

• Transaction (production relationship): It can be embedded in the Bitcoin Lightning Network to provide the payment capability of native digital currency, but it is only an option (very important). There are no other regulations except this. It can be done as long as it is not prohibited by law.  

The server is called Relay in the Nostr ecosystem. The data produced by the user is only temporarily stored on the Relay. Because each data has a signature, the server cannot forge user data or deprive the user of the relationship of producing certain data.  

The dissemination of data depends on initiating a request to the public Relay to obtain the data of a certain user. A piece of data can be stored on multiple Relays, and a user can also request the same piece of data from multiple Relays. Such a mechanism makes the server (Relay) a service provider that users can switch and choose freely. It has lost the power to define identity, control data ownership, and formulate transaction rules. It only has the function of temporarily backing up and storing data. It is more like an API-connected hard disk, and can only charge fees by providing this part of the service. The server has become a simpler role. Conclusion: The power of the server (Relay) has been weakened, and the power of the client has been amplified. Some people may worry that the power of the client will be too great. For example, a certain client will dominate the market, locking the user and making it impossible to exit? The answer is no. Because the power of the server has been deconstructed, the cost of users switching clients has also been reduced a lot, so there is no need to worry about this part. Of course, there is a situation where the client may introduce too many client-customized services that are beyond the scope of the Nostr protocol in the name of "caching/optimization". This practice still needs to be warned. Because once there are too many of these so-called optimizations, we will easily be restricted by the dominance of a particular client's services, rather than following the standards of the Nostr protocol, resulting in the inability to switch to other clients. This is the case where the ecosystem is kidnapped by a certain product. But for now, because there is almost no cost to switch clients, the bigger problem is how client developers can provide differentiated (but still compatible with the Nostr protocol) products, and even how clients should have a business model, which seems to be more difficult to explore than Relay's business model. But this part belongs to other content, so I won't discuss it for now. On the other hand, I think Nostr's relay-client architecture is closer to the mode of operation of human society in reality. 1. Relay as a free-cache or paid-long-term-storage Relay can be regarded as a free cache/paid long-term storage service (free to cache, pay to save) in the Nostr ecosystem. Many users who have just joined Nostr will ask a question, what if Relay goes offline, will my data be lost? The answer is yes, your data will be lost. But this may not be a problem. Nostr's social network is like going to a coffee shop offline to talk to people. The free public Relay only helps you cache the data you send for free. It is a medium of communication, just like the conversation in the coffee shop, the sound is transmitted through the air, Relay helps you spread the sound so that other people participating in the conversation can hear your message. Then after the conversation is over, everyone goes home, and the information of the conversation disappears into the air.  

Of course, if you think what you said is particularly valuable, you can also run a Relay to store these messages permanently, just like some people will write a diary when they go home to record what they said in this conversation today, but most people may not keep a diary. If you pay special attention to your own data, you can also use a paid Relay and let the paid server save your data for you. This is just like in life, you have too many things at home, so you go out and rent a warehouse to store your things.  

2. Relay as a localized autonomous community 

Today's social networks are globalized, and globalization will have globalization problems. The human brain is actually unable to process globalized information, because we have lived in a small village for a long time, and our neighbors may be only a few hundred people. When the network gives you globalized information, your brain can easily not handle it, so you always feel anxious when you are browsing Twitter. You should pay attention to the news of war, and you should also worry about the trade war and technology bottlenecks over there. The Relay model has the opportunity to return to the model of small communities. Each Relay is a community that promotes local autonomy, and these communities can be switched freely by switching Relays.

This is also why I feel that Nostr is pragmatic. We are actually taking a step back, giving up the pursuit of P2P, and turning to look for a "decentralized, small-scale community-autonomous" social network. Why give up P2P? Because P2P won't work (this is faitjaf's original words, and I agree). P2P networks have been around for so long, and we have no way to make it a social networking service available to everyone, and it is indeed not easy to use. So we turned to pursue a model of multiple Relays and just freely switching Relays. Obviously, it is more practical and at least it can be run.  

A misunderstanding of many people is that Nostr is decentralized and has no censorship. In fact, each Relay is a center. Each Relay chooses what kind of data to store, which is a kind of censorship of the data. But I think this censorship is acceptable, because Relay needs to bear the corresponding legal risks in reality. There is a specific person behind the Relay who is operating it. He is paying his own costs to provide services. He has the right to choose what kind of users to serve and what kind of data to accept. What we don't want is that everyone is forced to obey the rules of a unified server. Relays can be switched freely. If you don't like the rules of a certain Relay, you can completely cut off this Relay and use other Relays.  

The difference between Nostr ecology and blockchain 

Blockchain is a very expensive and more stringent system. Its data needs to reach a consensus through nodes to get eventual consistency. Blockchain can provide trust. For example, if a developer writes a smart contract and puts it on the chain, he can trust that the result of the contract is the same as what is written in the code. For users, it means that in a system like Bitcoin, I can trust the Bitcoin chain, which records how many Bitcoins I have, and the entire ledger is credible. But this kind of trust is very expensive and requires a lot of effort. Every Bitcoin mining machine is paying this price when doing PoW mining.  

On the contrary, Nostr is very cheap because its system is very loose. On Nostr, we only guarantee that every message sent out has been signed by the account. All you can get is a very thin layer of trust, "whether the received message is really from a certain public key." In addition to this trust, Nostr does not guarantee you anything else, including the availability of Relay data and the consistency of message order.  

But the corresponding benefit is that Nostr is very loose and very flexible. Many protocols do not make provisions, which is a space for the ecosystem to grow and develop freely. Such freedom and flexibility mean that Nostr is easy to connect to other systems and become the standard of the DID layer because it is thin and simple enough. And it is not like other blockchain systems that have a bias of a certain system (for example, the BTC community will not accept the ETH community, and the ETH community will not accept the BTC community).  

In terms of the way of doing things, Nostr and blockchain are fundamentally different. Blockchain systems often require a more meticulous architecture design in advance, specify the edges and corners of the protocol, imagine the behavioral logic of each ecological role in the protocol, design a good economic incentive mechanism to coordinate these different ecological roles, and also need to design a protocol upgrade mechanism. Nostr's approach is to only formulate a few of the most important rules, ignore the others, and let the ecosystem develop on its own. For example, how does Relay make money? This is not a problem that the Nostr core protocol should care about. Let the people in the ecosystem figure it out. Only the Relay that can actually provide value can survive and find a truly viable business model.  

In general, blockchain is great. Blockchain may solve 5% of the problems of the entire human society, 5% of very precious and critical problems, such as currency and finance. But blockchain is also very expensive, and you can hardly expect it to continue to solve the other 95% of problems. And not all problems require the heavyweight trust of blockchain. In contrast, Nostr may be able to solve the other 80% of problems, and these 80% of problems may only require lightweight trust, while the remaining 15% can continue to maintain the original centralized, right-wing approach.  

For me, what is more interesting about Nostr is the part other than the protocol, that is, the "ecosystem". No matter how well designed a protocol is, if there is no group of people working around it, then the protocol will never develop and evolve. Nostr’s greatest wealth is not the protocol itself, but the group of people gathered around the protocol. Many of them are BTC maxi, but there are also many people who are not interested in blockchain. If you look deeply into the development of the Nostr ecosystem, you will be amazed at the vitality and vigor it demonstrates. In my opinion, this vitality resonates cleverly with the principles followed by the TCP/IP group (IETF) when they formulated the standards: “We reject: kings, presidents and voting. We believe in: rough consensus and running code. That is, our credo is that we don't let a single individual dictate decisions (a king or president), nor should decisions be made by a vote, nor do we want decisions to be made in a vacuum without practical experience. Instead, we strive to make our decisions by the consent of all participants, though allowing for some dissent (rough consensus), and to have the actual products of engineering (running code) trump theoretical designs.

That is, our creed is that we do not let one person (king or president) dominate decision-making, nor should decisions be made by voting, nor do we want decisions to be made in a vacuum without actual experience. Instead, we strive to make decisions with the consent of all participants, although some dissent is allowed (rough consensus), and let the actual products of engineering (running code) trump theoretical designs. ”

In the Nostr ecosystem, there are some similar characteristics.  

• Organization structure: no official organization, light funding by people like Jack 

• NIPs: loosely join, everything is optional, openly debated, the founder is not afraid of being subjective 

• Developers ship fast: Everything happens publicly on the Nostr protocol(launch/discussion/feedback/debate)  

Most new things added in the Nostr protocol are done this way: firstly some clients/relays software developers introduce a new feature and then push the feature to public users, users give some feedback, and other developers notice that and start the discussion, some people writing NIPs, have debate on the NIP drafts, and then finally merge the NIPs ——this is true “rough consensus and running code” looks like.  

Summary

The original imagination and reality of cyberspace have diverged. The development of technology, whether it is P2P or the left-wing forces of blockchain, has made people try to re-examine the original imagination of cyberspace. At this time, Nostr, with a pragmatic and moderate positioning, has given a different solution. In essence, Nostr is a second-best option, because P2P/blockchain is too cumbersome for many scenarios that only require lightweight trust, both in terms of ease of use and technical cost. Nostr has a very unique ecosystem, pursuing rough consensus and executable code. The simplicity of the protocol itself leaves more room for the free development of the ecosystem, making it possible to become an origin that connects everything. 

This article is scattered and talks about a lot of fragmented opinions. Due to limited time, many topics cannot be expanded. I look forward to having the opportunity to talk in depth about the challenges and opportunities facing Nostr in the future.