Sparks Interview — Dimension: Data Belongs to the People
X-Order is an investment and research organization dedicated to the study of value capture in open finance. We strive to be a bridge between new finance and interdisciplinary fields with relation to science and research. It is founded by Tony Tao, who is also a partner at NGC Ventures.
We interviewed Suji Yan, founder of Dimension, and CTO Yisi Liu. We were deeply impressed by their understanding of technology and perseverance to their dreams.
Short Introduction of Dimension
Dimension, founded in 2018, is committed to building a new user data system, paving the way for the next generation of Internet applications.
Maskbook is a free and open-source extension developed by Dimension for user privacy protection. It is suitable for Chrome, Edge, Brave (all Chromium-based browsers), Firefox. iOS and Android versions have also begun public testing.
Maskbook allows users to post encrypted posts on Facebook. Only users and their friends can decrypt the content. In short, only users with your permission can view the content you publish, even the developers of Facebook and Maskbook cannot view them.
Can you introduce what Dimension does? How is the project coming along?
Dimension’s vision is that in the future when the architecture of the privacy Internet has gradually taken shape, we can play a significant role in it. At the same time, in the scenario of us having enough users, we can help them better use their data under the privacy of Internet architecture.
Since I personally advocate technology, I believe that in our daily life, the benefits that technology grants us is not limited to money. What I hope to achieve is to improve our service by better using our data.
Many companies now use our data well, but they don’t think from our perspective. They are simply commercial — monetizing our data.
However, I believe that some data processing from the users’ perspective can bring us more power. If we can process data with more customization, or even with the assistance of an intelligent robot, it can help us in all aspects of life.
This is a long-term vision. But before that, we need to help users build the underlying infrastructure.
“We should at least help them take back the ownership of the data first — regaining the privatization of data.”
We want to bridge the gap between Web 2.0 and Web 3.0. It is not a complete departure from Web 2.0, but a better bridging of value from both sides. I don’t think Web 2.0 will die in the long run, but there will be countless problems. We hope to solve these problems without users leaving the Internet giants’ platforms.
Which part of the project do you think is the most challenging, and how do you resolve it?
From a technical point of view, Dimension is an applied cryptography project, and the biggest challenge is also from Applied Cryptography: basically, 99% of proposed methods on research papers cannot be applied.
We need to integrate many traditional cryptography methods. Due to technology and efficiency factors, many algorithms designed by cryptography are unlikely to be run by computers. It may not be very difficult to actualise some algorithms into codes, but it is very difficult to implement them without loopholes. This requires a lot of people to study and implement it.
Another challenge is that we are developing projects in a decentralized system. If it is a centralized system, it is sufficient to just deploy all services on a few servers. However, if we need to run our whole system in a decentralized network, it will generate many new and unimaginable problems.
Decentralized identity is one of the products we want to develop. We hope that when users use some of our services, they don’t need to deal with a central server or a central service provider.
“It is the foundation of freedom that gives users identity and privacy.”
There is a difficulty in realizing this function under the decentralized network, that is, to achieve offline first. We need to make sure that your network or your service can operate without an Internet connection. We are now tackling this problem with a decentralized database structure, GUN DB. However, we still encountered a lot of problems during this process. For example, we have no way to track some of the data operating and transmitting in the decentralized network, which leads to a limited understanding of our users.
In addition, in the decentralized network, there is also a difficulty in multi-point synchronization. For example, if you have two mobile phones and three computers, in an extreme network situation, it is very difficult for you to synchronize the status of a device that is not connected to the Internet, with other devices.
I think the biggest difficulty besides technology is the change of social perspectives. The transformation of perspectives is not very intuitive, but if viewed from the aspect of social science, it is very intuitive.
What we are doing is akin to the traditional trade union in social sciences. In the Internet world, Facebook is like a huge factory with 2.2 billion people working in it. However, this factory is different from the factories in the 18th and 19th centuries; it has become a virtual thing. There is no supervisor whipping you at the back, and you don’t know what you are producing. Yet, this productivity relationship is the same as that of the factories in the 18th and 19th centuries, it is just more elusive.
Today, when we communicate with people on the Internet, we need to find a service on the Internet open market. This is because most people will not build a server at home or make a chatting software just to communicate with other people through the Internet.
“This is akin to having your “land” on the Internet taken by capitalists, thereby being forced to become a “proletariat” on the Internet.”
What does each of us produce in this virtual factory? Very simple, it’s data. Everyone is trapped in a trend by their own social relations, entering social media such as Facebook and Twitter. It is just like how the hooligans in the 18th century moved from the countryside to the city, only to be trapped in factories.
Now, we produce data components and parts, which are very clear in the political economy. In the past, strikes and the proletarian revolution were well understood. However, it is unusual if you say that social media software such as Facebook is exploiting you.
“In my opinion, this is the value transfer of the Internet.”
We are surrounded by a bunch of Internet companies in a virtual factory who is making people work, yet no means to organize workers’ movement to fight for everyone’s rights and interests.
Therefore, we want to do something that can connect all the data of workers and users, so that they not only know that they are exploited but also participate in moderate strikes and earn money.
One of our products, Maskbook, encourages users to obtain the ownership of their data on Facebook before forming a free private data market.
To achieve this goal, the first step is to encrypt users’ data. In addition to encryption, we can even issue stable cryptocurrencies, cryptokitties, etc. on the original platform. Finally, program our own data similar to contracts. These plans are being pushed forward step by step.
Why do you want to initiate the Dimension project?
I used to do NLP related research, and the main data sources are from social media. I needed to crawl a lot of unstructured data, before either directly processing them through a model or through information extraction — structuring the raw data before processing them.
These works are very complicated and I have been working on them for a long time. However, I later realized that in this long period, we never questioned whether we are doing was right. This is actually a philosophical question.
“I believed I was doing good because the aim of the project was to help people, but it was a double-edged sword because the very same action could also infringe on people’s privacy.”
All the analysis we have done was actually one-sided because we had too little data. The big problem is that we only knew the user’s behavior on a certain platform, therefore the understanding of this person is not holistic. We have no way of knowing what he really is like.
Thus, in the beginning, we would like to start from the users — to get more holistic and comprehensive data, some of which they may or may not know, and then make personalized recommendations based on these data to bring them some benefits.
However, this is not safe, as it violates the rights and interests of users. Therefore, we want to help users solve the problem of data ownership first, and then help them do what I mentioned before.
There have been great changes in the market since 2017, do you have any new perspectives on the industry?
The most immediate feeling is that financing has become more difficult. Since the beginning of 2018, there were many ICOs. The bad cryptocurrencies in the market were driving out the good ones, which eventually led to the lack of money and poor operation of many venture funds.
Even if there are still funds with capital, the period of withdrawal after an investment has become very short, with some even requesting for withdrawal after a year. Under normal circumstances, the withdrawal period of Internet funds are 3+1, or 3+2, which is not very long, but at least it gives start-ups more time.
With our country supporting blockchain, it led to some funds that were previously afraid to touch blockchain projects coming to us for a chat.
What are your views on Consortium Blockchain and Public Blockchain?
In my opinion, it is a bit like Silicon Valley before Google came out. At that time, the hot innovation project was an industry search engine. Ma Huateng also wrote an article in the era of CFido, discussing him working on an industry search engine outsourced project. It was a very profitable business at the time.
Postal, telecommunications and water conservancy bureau, libraries, banks and insurance companies in the U.S. all needed industry search engines. Just like some search engines in law schools now, it is able to search papers, government reports, etc., which was a hot business at that time. Many Silicon Valley investors have invested in B2B and B2G companies. However, after Google came out, the search engine companies in these industries all folded.
In the past, the ability to search was a privilege — it had to be done by the librarian. Such privilege would be eroded away if everyone can use search engines. This vertical segmentation of the industry search engine can only survive in a small industry.
I think the same goes for the blockchain industry. The problem now is that B2C’s solutions are too difficult, coupled with a lot of hype bubbles, many projects have collapsed. Therefore, many blockchain companies will embrace the government or enterprises, which at least allows them to earn some money to support themselves.
However, it’s not good to develop a pure B2B or B2G application.
“As long as one application can integrate all these functions, then the applications in these subdivisions will lose their competitiveness.”
In our opinion, in the short term, we can do consortium blockchain, but we still need to strive to become a part that integrates all functions. This will allow us to survive in the future.
From your WeChat moments, we understand that one of the products of Dimension — Maskbook, is banned by Facebook. How do you plan to deal with this?
To clarify, it is information that includes links to Maskbook.com or Maskbook.io on Facebook that is blocked. The first version (launched in July 2019) uses the Maskbook plug-in to encrypt information, producing an output “decrypt this with Maskbook.com”. It is the content after this statement that requires encryption.
This is a common means of “increasing the number of hackers” and “virus transmission”. The early core users of Maskbook had many means to recommend their friends or posting encrypted posts. Facebook does not want individuals or groups of people to send many links frequently as that would violate the so-called Community Standards of Facebook. Therefore, in August 2019, it blocked these links. After that, the blocked users can appeal, and we have filled it too. However, the “Community Standards” of Facebook is actually very much double standards. Although it operates on the Internet publicly, they as private companies are not obliged to give you any replies (at least the American court will deal with you).
In addition, the ban was actually for all of Facebook services, in which Facebook, Messenger, and Instagram was blocked altogether. WhatsApp is unable to be blocked due to the encryption, however, Facebook wanted to remove the encryption much earlier. This was also the reason why the founder of WhatsApp left and objected to Facebook at that time. These Apps occupies most of the European and American Mobile social media traffic entry. It is shocking that we got this treatment once we went online.
However, this kind of ban can only block centralized items such as the link, but not the ciphertext itself (because encryption of ciphertext can even be made into emoji, stickers, pictures, etc.). It is also unable to block other communication methods, such as the posting of Chrome Web Store links, official community links (such as Twitter). We are also integrating IPFs to host content. Eventually, unless Facebook seals off all IPFs nodes, there is no way they can truly block us.
Recently, during Ethereum’s “Devcon” itself, MetaMask announced its new plugin system. Its founder, Dan is also very supportive of us. He opened an issue on MetaMask’s Github to discuss the technical issues of integrating Maskbook. If the integration is completed, Facebook will have to kill Ethereum before banning us.
P.s: Besides Maskbook, Dimension also launched another product, Tessercube, which allows the users to distribute decentralized “red packets” on various platforms such as WeChat.
In our next article, we will be discussing our observation on token derivatives and an analysis of the FTX double options — MOVE contract.
Originally published at https://www.datadriveninvestor.com on January 10, 2020.
Connect with us on Linkedin!
Translated by (via our WeChat Account): Xin Yue
Editor: Daphne Tan
Sparks Interview — Dimension: Data Belongs to the People was originally published in Data Driven Investor on Medium, where people are continuing the conversation by highlighting and responding to this story.