Decentralized Web: Research Findings

What are CKS Workshops? Upcoming Workshops Notes from Past Events

Who: Anna Lekanova

Date: July 15th

Time: 3:00pm EST / 7:00pm UTC+0

Language: English

What: The Decentralized Web or DWeb is a decentralized or distributed web that is not controlled by a single entity or being logically decentralized. Projects considered dweb are Tor Project, BitTorrent, Napster, or Bitcoin. Dweb projects are picking up steam - some focus on decentralizing identity and social networking, others are building distributed application protocols or platforms that run the web stack (HTML, JavaScript and CSS) on something other than HTTP. There are also block-chain platforms that run anything as long as it can be compiled into Web Assembly.

Anna Lekanova, from the Fluence Project, conducted a survey with more than 650 Decentralized Web Developers to find out their needs, the current challenges and their role inside the internet environment. Anna will share their highlights and conclusions of the survey, navigating through these key points:

• Main web pain points defined by the developers, like the complexity of the systems, lack of data ownership, current web business models
• Worries and Opportunities of the Decentralized Web as its privacy; surveillance and censorship; tech resilience, and the worrying current web business models.
• The hurdles of the DWeb adoption by web developers, like the lack of understanding by the general user or the immaturity of the new tech.
• The analysis of the technical motivations, ethics and fund model that run behind projects being built in the ecosystem.

Anna Lekanova is a decentralized web advocate. She leads community building at Fluence, a peer-to-peer computing protocol & a software licensing system.

>> Check notes out notes from other sessions here

Notes

The Fluence Project constructed the DWeb report by navigating several nuances in the perception of distributed web technologies vs. the Web 3.0. These are the findings on how developers and community supporters differentiate the prospects of the two rather nebulous terms. You can check out all the details of the survey on the Decentralized Web Developer Report 2020 Article

Big picture

Survey

• 1329 views
• 650 respondents
• 417 tech background
• 231 project builder
• 66 minutes to complete

DWeb Panorama

• No privacy, surveillance, censorship
• Tech resilience, security
• Current web business models are worrying users

Motivations & Business

1. DWeb is ideologically driven

• Privacy, security
• Data ownership
• Censorship resistance

2. Governments and big tech are resistant to decentralization

• Cloud lock-in
• Current web business models

3. Monetization and funding are critical

• Critical to growth
• No sustainable business models in DWeb so far
• Need support by Tech Giants and government

4. Users: Poor UX and technical complexity 5. Developers: Lack of resources such as documentation & education materials

Demographics

• Full-stack engineer: 53%
• Backend engineer: 41%
• Research engineer: 26%
• Dev0ps engineer: 25%
• Fronted engineer: 21%
• Founder: 18%
• Student: 14%
• Scientist: 13%
• Business/operations: 11%
• Mobile engineer: 9%
• Investor: 4%
• Other: 12%

What are the main pain point of the current Web

• Massive personal data leaks, exemplified by the likes of Marriott and Equifax, were the leading concern — according to 68.5% of respondents.
• Censorship and access restriction, instituted by both tech giants and governments, came in a close second and third position, both cited as the leading concern by 66% and 65% of respondents, respectively.
• Ads based on personal data (61%) and user data held by applications (53%) rounded out the top five.

What should be fixed in the first place?

• Data sovereignty was the clear leader as the initial fix cited by respondents, with 75.5% of respondents indicating the reversion of data control to the user as paramount.
• Data privacy (59%) and tech resistance (56%) to disruptive events or shutdowns (e.g., the Cloudflare case) followed at second and third, respectively.
• Security, specifically the ubiquitous use of cryptographic signatures across applications came in fourth at 51%, followed by network anonymity (42%) — private access to resources.

The meaning of Decentralization

• As articulated by the respondents, the movement towards DWeb is ideologically driven and is characterized by a bottom-up grassroots ethos. Respondents generally favor the DWeb defined as an architecturally decentralized web (82%), where there is no single point of failure or data accumulation.
• Naturally, these relate to downstream preferences cited by respondents that they prefer a web not politically divided (64%) nor subject to the problems of a single software monolith structure (39%).
• Roughly 37% of respondents detailed the meaning for distributed/decentralized as “Don’t trust, verify” where everything is verifiable.

What is awesome in the DWeb technologies compared to traditional Web?

• The response to this question leaned heavily towards “values and mission,” encompassing 77% of respondents and, once again, reflecting the ideologically driven grassroots character of the DWeb.
• Security (43%)
• Community & support (31%), interoperability (31%)
• Scalability (30%) completed the top five responses.

What are the most significant changes could be brought by DWeb?

Respondents echo the sentiment with their projections of the most significant changes that can be brought about by the DWeb. For example, 75% of respondents cited taking back control of personal data as the biggest change. Roughly 55% subsequently detailed an inability to forge or censor content while another 50% cited no tracking or surveillance as one of the most significant changes — mirroring the primary pillars.

DWeb would be achieved by the following technologies:

• P2P communications protocols (55%)
• Content-addressable storage (54,5%)
• P2P file-sharing (51%), decentralized DNS (47%)
• Privacy-focused networking (46%).

Applications tested on DWeb technolgies:

• Top includes: IPFS (36%), Ethereum (25%), Dat (14%), and Libp2p (12%).
• Other: WebTorrent, Freenet, Textile, Holochain, 3Box, Embark, Radicle, Matrix, Urbit, Tor, BitTorrent, Statebus/Braid, Perlinks, BitMessage, Yjs, WebRTC, Hyperledger Fabric, and more.

Main frustrations about the DWeb technology:

• A dearth of documentation, tutorials, videos, and other educational resources (44%)
• A difficult time understanding how to practically apply the technology (42%)
• Difficulty integrating the tech (40%)
• The challenges of scaling distributed tech (21%).
• A lack of tools, incompatibility of tools, fragmentation, lack of documentation, and developers feeling overwhelmed by choice of different decentralized protocols were among the most frustrating points cited by respondents.

Biggest obstacles towards DWeb:

• Roughly 70% of respondents cited the lack of user awareness about what the DWeb, and its concomitant technologies, actually comprise as the most significant obstacle.
• The immaturity of the technology followed in second (49%)
• The pushback from back tech companies reveals an intriguing third-largest obstacle at 42%.
• A furtive business model for DWeb projects followed suit, respectively, at 38%
• A general lack of integration with web browsers at 37%.

Main reasons to choose p2p over cloud-centric centralized architecture.

• Respondents indicated a strong ideological preference for building on the DWeb with P2P technologies. Considering the multiple tech hurdles and frustrations expressed by respondents, the strong ideological preference (72%) reveals a robust foundation for the broader DWeb movement.
• Similarly, the “technical reasons” option was selected by 58% of respondents, which, judging by the comments and responses to other sections, seems to also hinge on the technological benefits that bolster the ideological preferences. Namely, censorship-resistant P2P communication, distributed storage, and other P2P developments.

Funding DWeb projects:

• More than half (53%) of DWeb projects respondents are self-funded. Roughly 19% are VC/Angel funded, and 15% are based on grants. Token sales continue to count for a diminishing amount of funding, only contributing to the funding model of 10% of respondents — a far cry from the ICO days of 2017.

DWeb project's business models:

• 30% of respondents indicated no way for them to extract money from their projects.
• 22.5% decided on kicking the can down the road on monetization.
• 15% cited a “Freemium” model.
• Only 15% of the respondents had a paid product for their DWeb technology project.

Resources

---

• Decentralized Web Developer Report 2020: https://medium.com/fluence-network/decentralized-web-developer-report-2020-5b41a8d86789
• Raw data of the DWeb research: https://docs.google.com/spreadsheets/u/3/d/1mrZ8tZSh8Q1vonYXM4pgkKf-n4C9YR3bZVAeixcm1e0/edit?usp=sharing
• Decentralized Web Summit videos: https://www.decentralizedweb.net/videos/
• Redecentralized digest https://redecentralize.org/redigest/
• IPFS https://docs.ipfs.io/community/
• #DWeb meetups at meetup.com: Distributed camp https://twitter.com/DistributedCamp
• DApp developer research 2019: https://medium.com/fluence-network/dapp-survey-results-2019-a04373db6452