Note. The original report was published on Medium in English. It also contains quotes from respondents and links to participants. A shortened version in the form of a tweet storm is available .
What the research is about
The term DWeb (Decentralized Web, Dweb) or Web 3.0 is most often collectively for a number of new technologies that will revolutionize the web in the next few years. We spoke with 631 respondents who are currently working with distributed technologies and building a decentralized web.
In this study, we have collected topics of current progress and the main obstacles that developers are facing in the new web. As with all new technologies, there are many difficulties in developing decentralized solutions, but the overall picture is promising: the decentralized web offers a lot of hope and opportunities.
The web was originally conceived by Tim Berners-Lee as an open, decentralized network for communication. Over time, the technology giants of the FAANG five began to create user-friendly interfaces and took the lead in gaining critical mass.
It is convenient for people to use fast and free services, communicate with friends, acquaintances and audience. However, this convenience of social interaction has a downside. More and more cases of surveillance of users, censorship, breaches of privacy and various political consequences are opening up. All of this is the product of centralized data control.
Now more and more projects are creating independent infrastructure and trying to get rid of intermediaries in the form of FAANG.
In the early 2000s, big indie projects - Napster, Tor, and BitTorrent - heralded a return to decentralization. They were later overshadowed by their centralized competitors.
Interest in decentralization subsided, and revived with the appearance of scientific work on a new decentralized currency - Bitcoin, authored by Satoshi Nakamoto.
From now on, new DWeb protocols such as IPFS are paving the way for fundamental changes to the web. And the surviving projects of the early 2000s, such as Tor, I2P, and even Mixnets, are entering a new round of development. Now, a generation of projects and developers are striving towards the original vision of the decentralized web, conceived by Tim Berners-Lee in 1990 at CERN.
There was a lot of disagreement in the community as to what the new web is. Our research reveals the general principles shared by developers in this area.
The study begins by examining the most significant challenges with the current web and ends with how DWeb can overcome the challenges it faces.
Main conclusions
- Most of the projects are less than two years old, which suggests that DWeb is still emerging and is still an immature technology.
- Three-quarters of respondents believe DWeb is driven mainly by ideology and enthusiasm, which is not yet understood by ordinary users.
- Data privacy and control, as well as technology resilience to disruptions, are the most anticipated features of DWeb.
- DWeb peer-to-peer .
- DNS, SMTP, XMPP ., HTTP.
- - DWeb ; .
- IPFS Ethereum , DWeb .
- DWeb , : , , DWeb .
- , , COVID-19 - , .
Differences between Web 3.0 and DWeb
Research Participants
Current Web
3.1 Problems of the Current Web
3.2 DWeb Web Protocols 4.1 The Concept of Decentralization 4.2 Values and Mission 4.3 Technical Challenges 4.4 DWeb Applications in the Future Dweb Implementation 5.1 Main Constraints 5.2 Barriers to Mass Use 5.3 The Role of Blockchain DWeb Projects 6.1 Project Types 6.2 Motivation 6.3 Project and Team Status 6.4 Technical Specifications 6.5
-
Web 3.0 DWeb
In our DWeb research, we focused on several differences in the perception of distributed web technologies compared to Web 3.0. In particular, how developers and community supporters define the perspectives of two rather vague terms.
The survey responses show that the common goals and vision of DWeb and Web 3.0 overlap significantly.
Web 3.0, mainly promoted by the blockchain community, focuses on commercial development - finance, e-commerce, AI and big data for companies. DWeb proponents (for example, IPFS and Internet Archive), by contrast, are more focused on the ideology of decentralization: data sovereignty, security, privacy and censorship resistance. DWeb projects cover a wider range of technological innovations than Web 3.0.
In general, the two perceptions of the next iteration of the network do not contradict and can actually complement each other.
From a research navigation perspective, it is best to focus on the views of DWeb proponents and how these developments (e.g. P2P, decentralized storage, data privacy) will shape the infrastructure of the future network.
Study participants
The study consisted of a survey that was completed by 631 respondents, of whom 231 are actively working on DWeb-related projects.
1. What is your background?
The poll consisted of 38 questions. The percentage distribution in the answers is based on an unlimited choice of answers by the respondents - in most cases the total percentage of answers will be over 100 percent.
The study sample focused primarily on developers and engineers working on DWeb-related projects. We did not specifically target blockchain developers, so they represent a small percentage of all respondents.
For those who want to see the raw data, we have published anonymous raw results.
Current Web
The web as we know it has evolved over the past two decades. Information is available instantly and free of charge. Powerful applications are built on top of existing infrastructure. An entire service-oriented cloud computing industry is flourishing. The whole world is connected through instant communications.
However, the current web has come up with some behind-the-scenes compromises. The Internet is evolving every second, absorbing more and more data, increasing and combining capacities. As a result, users become a resource and their privacy is relegated to the background, especially when it comes to generating ad revenue.
In this section, we examine the ideological and technical considerations of research participants about the structure of the current web.
The most vulnerable spots of the current web
The general opinion about the state of the current network is largely based on the exposed vulnerabilities. First of all, they stem from a common problem - centralized data storage. The result is untoward side effects ranging from major data breaches to censorship leverage by FAANG and governments.
2. What are the main problems in the current Web
At first glance, many of the most significant issues may appear to be ideologically driven and limited by privacy advocates. Nevertheless, the younger generation - the main audience of netizens - are increasingly asking questions. They are tired of intrusive ads, data leaks, and a general lack of control over data or lack of privacy.
It is interesting to note that the range of opinions shows a strong dislike for the current web paradigm, especially when it comes to how the web is currently monetized.
Whether the long-term consequences of ad monetization are unfavorable (such as centralized data control and invasion of privacy) is irrelevant - respondents are unhappy with the outcome.
In addition, respondents expressed antipathy towards closed systems. In particular, the inconvenience is caused by the closure of products or the lack of control by users over their data. Users have little influence on what content they see in feeds, data, or navigation through closed systems. More accessible and user-friendly standards need to be found.
3. What should be fixed in the current web first?
The responses echoed somewhat the comments about the most vulnerable points.
- Data sovereignty was the clear leader. Moreover, 75.5% of respondents indicated that the return to the user of control over the data is paramount.
- Data confidentiality - 59%
- Technological resistance to destructive events or accidents (for example, in the case of Cloudflare) - 56%
- Security, in particular the widespread use of cryptographic signatures in applications - 51%
- Network anonymity - 42%
There is clearly a growing dissatisfaction with centralized data warehouses and the power of FAANG companies. The rapid evolution of tools such as cryptography offers hope of breaking the data monopoly and, as a consequence, the abuse of privacy. Therefore, respondents prefer to move away from the third party trust model.
Web protocols
4. What needs to be added or changed in existing protocols?
The answers to this question varied greatly.
- Embedded personal data layer - 44%
- Built-in user authentication - 42%
- Offline operation by default - 42%
- Built-in peer-to-peer layer - 37%
- Some of the responses, such as platform independent identification and user authentication - 37% - can be combined under a wider layer of personal data.
In additional comments, respondents identified the lack of standards and complexity of composition as the main problems of limitations of existing protocols. In addition, some developers have also pointed to the lack of user incentive models built into protocols. Exactly how to motivate people to use DWeb services can be critical in attracting them to open network protocols.
5. Which existing internet protocols need to be redesigned?
Diving into more technical details, the participants agreed on specific protocols to be redesigned. For example, this:
- Resource Addressing Layer Protocols (DNS) - 52%
- Communication protocols (SMTP, XMPP, IRC) - 38%
- HTTP - 29%
One of the most notable findings was the need for a more secure transport layer, namely equipping it with data security, digital rights management, and even incorporating Tor into the transport layer.
However, some participants are skeptical about the decentralized approach. The reason is the need for additional development of improved hardware for decentralized protocols. In their opinion, it is better to simply supplement existing protocols than to completely change them.
DWeb
Decentralization concept
6. What does “D” mean in the concept of Dweb?
The letter "D" in DWeb stands for decentralized, that is, a kind of distributed or decentralized system. There is no clear definition of such a system, but, in practice, it can be a dynamic movement from a centralized model of the current network to a decentralized one. However, such a movement is non-linear and faces certain difficulties.
This section of the study reveals the tasks and prospects for the implementation of the DWeb concept.
According to the respondents, the movement towards DWeb is ideologically oriented.
- Most understand DWeb as an architecturally decentralized network, where there is no single point of failure or data accumulation - 82%,
- 64% of participants see Dweb as a politically uncontrolled network,
- 39% say that the logic of the network should be decentralized,
- 37% of respondents indicated that the network should be “distributed” or “decentralized” according to the principle of “do not trust, check”, where everything is verifiable.
The respondents have high hopes for DWeb as an ideological construct. It should be more than just a new technical network. It should be a tool to help create a collaborative environment on the Internet. The massive use of open source can lead to better scaling and the development of more powerful custom applications. As a result, companies and ordinary web users can use a huge amount of resources previously isolated by corporations.
DWeb values and mission
As we noted earlier, DWeb's focuses, according to respondents, are mainly related to data sovereignty, censorship / redundancy resistance, and privacy. The rest of the answers operate as additions to the main focuses in one form or another.
7. What are the most significant changes that DWeb can bring, in your opinion?
- Taking back control of personal data - 75%
- Failure to fake or censor content - 55%
- No user tracking / tracking or supervision - 50%
The opinions of the respondents are undoubtedly ambitious. But this is what the new DWeb infrastructure requires, and as we will see, there are a number of technological changes that are supporting this movement.
8. What is cool about DWeb technology compared to the traditional web?
The answers to this question relied heavily on "values and mission", once again reflecting the ideologically motivated nature of DWeb.
- Security - 43%
- Community and support - 31%
- Compatibility - 31%
- Scalability - 30%
Offline / local application development, shorter latency and high resiliency were cited in the comments as the main technical advantages of DWeb.
Technical problems
9. What technologies can contribute to the massive use of DWeb?
The responses to the survey in this section revealed the views of the participants about the technologies that will help bring about the new web.
- P2p communication protocols - 55%
- Targeted storage - 54.5%
- P2P file sharing - 51%
- Decentralized DNS - 47%
- Privacy-focused networks - 46%
10. Have you tried making applications with DWeb technologies? Which ones exactly?
- IPFS - 36%
- Ethereum - 25%
- Dat - 14%
- Libp2p -12%
IPFS and Ethereum, in particular, are among the fastest growing open source projects of all DWeb applications and protocols.
The developers also mentioned a number of other projects, including WebTorrent, Freenet, Textile, Holochain, 3Box, Embark, Radicle, Matrix, Urbit, Tor, BitTorrent, Statebus / Braid, Peerlinks, BitMessage, Yjs, WebRTC, Hyperledger Fabric and many more.
11. What disappoints you the most about DWeb technology?
Similar to our research on DApps and blockchain developers last year , many of the frustrations listed were due to a lack of documentation. We see the same with DWeb technologies.
- In particular, the main disappointment is the lack of documentation, tutorials, videos and other educational resources for developers - 44%
- There is also a problem with understanding where and how to apply Dweb technologies in practice - 42%
- The complexity of integrating technologies with each other - 40%
- Scaling problems for distributed technologies - 21%
That many of these limitations mirrored last year's results for blockchain applications can be largely attributed to the lack of readiness of new technologies.
Lack of services, incompatibility of services, fragmentation, lack of documentation, and too much choice of decentralized protocols still in development were also among the most frustrating issues mentioned by respondents.
12. What are the most difficult technical moments in development using P2P
Answers to the question about the difficulties of DWeb focused on specific problems of implementing p2p projects. We again observe the difficulties already mentioned earlier.
- Scaling issues - 34%
- Stability of connection of peers in the network - 31%
- Productivity - 25%
* * *
The next part will be useful for developers interested in specific challenges in the DWeb ecosystem. Some of Dweb's concerns include technical complexity, such as P2P layered architecture.
DWeb clearly shows difficulties in motivating users. Other unresolved issues are related to user registration issues, network latency, peer discovery, network testing costs, and data synchronization issues.
Additionally, there are certain difficulties of program and browser incompatibility, network instability, user identity management and analytics.
Use of DWeb technologies in the future
13. How likely are you to use DWeb technology in your next project?
Respondents already working on DWeb projects expressed a greater desire to use DWeb technologies in their next project. Conversely, developers who were simply interested in DWeb technology indicated a lower preference for using DWeb technologies for their next project.
Perhaps interested developers are just waiting for the technology to mature a little before using it. On the other hand, developers already working with Dweb don't want to waste their time, effort and contribution to the overall ideology, and will continue to work with DWeb for the foreseeable future.
DWeb implementation
14. What are the most difficult obstacles to DWeb?
Despite the technical challenges facing the continued growth of DWeb, they are not the main bottleneck — the user issue.
- Users are not sufficiently aware of what DWeb is and its benefits - 70%
- Unavailability of new technology - 49%
- FAANG Resistance - 42%
- Lack of business models for DWeb projects - 38%
- Lack of integration of decentralized technologies with web browsers - 37%
It seems that centralized data-driven business models and the current network structure will prevail until broader user awareness reaches a tipping point and DWeb projects find viable monetization avenues.
15. What exactly is preventing the massive use of your DWeb application / protocol?
- Project unavailability - 59%
- Difficulty learning / explaining to new users how DWeb works - 35.5%
- Relatively small number of DWeb users - 24%
User awareness of decentralized technologies is needed to distract them from the centralized, traditional paradigm that dominates the web today. Along with the advantages of UX / UI centralized systems, DWeb ideology brings many more positives for users. So far, understanding and especially use is too difficult for an ordinary user without a technical background. Many p2p applications are launched differently from regular applications.
DWeb services are still almost impossible to use from traditional browsers. And DWeb services that can be used on a daily basis are still quite few. All of this is among the barriers faced by new users of the decentralized web.
The role of blockchain
Blockchain technology was at its peak during the massive ICO launch at the end of 2017. Since then, developers and companies have interacted with various blockchain services with varying degrees of success.
The answers were divided between those who support Bitcoin and its accompanying cryptocurrency industry, and those who do not believe that blockchain can be the solution to all problems. Opinions about blockchain vary greatly, especially with regard to its performance and disadvantages compared to centralized systems.
The results point to growing doubts among developers about the advantages and disadvantages of using blockchain. Instead of trying to build everything on blockchain and claiming that it is a panacea for the world's ills, respondents are simply showing interest in its future use.
16. What do you think about the role of blockchain?
- Blockchain is not the solution to all problems - 58%
- Blockchain is convenient for digital currency and payments - 54%
- Blockchain is ideal for decentralized IDs - 36%
- Usefulness of blockchain for a wide range of DWeb tasks - 33%
- Blockchain can be used in digital certification - 31%
- Blockchain technology is a "waste of time" - 14%
DWeb Projects
Project types
The respondents working on various DWeb projects are geographically dispersed all over the world, and work in both unknown and more popular projects in this area. Some of the more famous projects include IPFS, Dat and OrbitDB, and smaller ones like Lokinet, Radicle, Textile, and others.
17. Types of DWeb projects
The types of DWeb projects varied considerably. We have summarized them into groups depending on the goals. Here are the most popular directions that respondents give their ideological preferences:
- Data storage and exchange areas - 27
- Social networks - 17
- Finance - 16
Interestingly, social media censorship and the limited ability to share data without using the FAANG infrastructure were among the most pressing issues with the current web.
In addition, the financial revolution, manifesting in the most practical use-case of DeFi on Ethereum, is the fusion of blockchain technology and P2P DWeb protocols.
DWeb project types accurately reflect the ideological preferences of the research participants. They show that projects are working on actual problems, not theoretical technology platforms.
18. What are you developing - a protocol or an application?
Of all the participants in the study, 231 people indicated that they were working on a project.
- Develop applications for end users - 49%
- Working on infrastructure or protocols for developers - 44%
Motivation
19. Why did you choose P2P over a centralized architecture for your project?
The developers have previously noted the ideological preference for using DWeb and P2P technologies. When asked why they choose peer-to-peer technologies,
- The majority relies on fundamental ideological values - 72%
- Chose DWeb for technical reasons - 58%
Judging from the comments and answers to other questions, the second outcome appears to be related to technological advantages that support Dweb's values. Namely, censorship-resistant P2P network, distributed storage and other P2P technology developments.
Project and team status
20. At what stage is your project?
- Still under development - 51%
- Launched - 29%
- At the idea / concept stage - 15%
- Are at other stages of development - 5%
21. How long are you working on your project?
Relatively speaking, most DWeb projects are new compared to their centralized web counterparts.
- Only work 1 - 2 years - 31.5%
- Existed for more than 3 years - 21%
- Work less than 1 year - 17%
22. How many people work in your project?
Team sizes vary within small ranges.
- Two to five people - 35%
- Working alone - 34%
- More than 10 developers in a team (usually well-known projects such as IPFS) - 21%
- A team of 6 to 10 developers - 10%
Specifications
As for licensing open source DWeb projects, developers choose licenses that are relevant for traditional technologies.
23. What license did you choose for your project?
- MIT - 42%
- AGPL 3.0 - 21%
- Apache 2.0 - 16.5%
- The decision on licensing has not yet been made - 18.5%
- Don't license their code - 10%
24. What is the main stack of your project?
The project stack is a combination of the most commonly used front-end, back-end and DWeb technologies.
The frontend is mainly represented by:
- React - 20
- Typescript - 13
- Angular - 8
- Electron - 6
For the backend, respondents mainly use:
- GO - 25
- Node.js - 33
- Rust - 24
- Python - 18
In general, the selection reflects massive trends in open source development, such as the Github report "State of the Octoverse".
In DWeb technologies, the leaders are:
- IPFS - 32
- Ethereum - 30
- libp2p - 14
- DAT - 10
Business models and investments
25. What is the business model of your project?
Business models in DWeb have been cited as one of the most pressing challenges facing developers. It is difficult to extract value from open protocols that do not adhere to centralized data monetization schemes.
- There is no model for generating income from your project - 30%
- I'll think about it later - 22.5%
- Freemium model - 15%
- Paid DWeb product - 15%
Some of the conceptual ideas of monetization are still half-baked for use in DWeb. For example, SaaS and licensing have been mentioned several times in the comments. Staking and governance in blockchains have also been mentioned in several projects. While they definitely have potential, they are still in their very early stages and not ready for widespread adoption.
Financing
Investment can be critical to turning an idea into a viable project.
26. How was the first investment for your project obtained?
- DWeb project funded by its founder - 53%
- Received investments from venture funds or business angels - 19%
- Received grants - 15%
- The number of token sales and ICOs has been significantly reduced since 2017, and accounts for a small share of all projects - 10%
Participants in the study were not shy about expressing their frustration at the difficulties of obtaining investment for DWeb.
Project audience
27. Monthly audience of your project
The problem of attracting and training users affects the number of users of DWeb projects. The number varies greatly downward compared to centralized applications.
- Have not launched the product yet - 35%
- Less than 100 users per month - 21%
- They are unable to assess their audience - 10.5%
- Don't know the number of users - 10%
- 100 to 1K users - 9%
Conclusion and conclusions
- The notion of “DWeb” among its proponents is largely driven by both semantics and the broader goals of decentralization: data sovereignty, privacy, anti-censorship and related changes. Apparently, all this is the main leitmotif and point of growth of Dweb.
- DWeb. .
- DWeb, DWeb . , . , DWeb .
- DWeb. . , , - , DWeb.
- , , . , . , DWeb . , . , -.
- DWeb . , , . DWeb , -. FAANG . DWeb .
DWeb's vision is to disrupt many of the centralized models, such as the client-server data model and the advertising-driven business model, and recreate the decentralized ones from scratch, which is very ambitious.
DWeb technology is generating deep interest and is growing rapidly. Outstanding projects like Ethereum and IPFS already have a huge following. However, the number of users and the acceptance of small projects are decreasing due to the monopolization of the market by traditional tech giants. For these projects to be further developed, infrastructure is needed. For example, developer tools and supported documentation, as well as leverage to attract the average web user to DWeb applications.
The number of users in crypto, blockchain and DWeb is significantly less compared to conventional applications. However, many developments over the next few years can greatly contribute to the growth of DWeb. This is influenced by such factors:
- A growing awareness of the need for higher levels of privacy following exposure of government oversight, serious breaches and massive consumer data breaches. Users want to control their data. Digital privacy is now in high demand. DWeb will be able to show users practical solutions.
- Uncertain economic and monetary policies during a pandemic could push many to explore crypto technologies, and thus introduce them to part of the DWeb.
- open-source , , .
- -, DWeb (, Opera) (Brave), .
The Internet, despite its humble, decentralized origins, has been centralized for decades.
The resurgence of decentralized technologies and the active grassroots movement supporting them have given us hope to suppress further centralization of the Internet. Getting back to basics would mean a decentralized, open and accessible Internet, free from the control of both governments and tech giants.
This is a vision worth developing and this is the reason why so many engineers are working towards this goal today. The answers in our study have highlighted several significant barriers to implementing a thriving DWeb, but the potential is real.
We concluded that while DWeb is clearly in its early stages, this does not prevent it from meeting, and even fitting perfectly into, the changing preferences of today's Web users.
The list of research participants can be found here . Anonymized raw data is also available . Thank you all for participating!