Advice for Musk: How to Create a Decentralized Version of Twitter?

Original Author: Preston Bybne, Partner at the renowned law firm Anderson Kill

Original Title: 《How to Build Decentralized Twitter》

Translated by: Hu Tao, Chain Catcher

Musk has won.

On April 25, Twitter reached a final agreement with Musk, whereby an entity wholly owned by Musk will acquire the company for $54.20 per share in cash, valuing the deal at approximately $44 billion. After completion, Twitter will become a privately held company. Of course, while this acquisition has been approved by Twitter's board, it still requires shareholder and applicable regulatory approval, as well as meeting other customary closing conditions.

In short, Twitter has been acquired by Musk, who, according to his previous promises, will implement several reforms including open-sourcing the algorithm, addressing the issue of fraudulent bots, adding an edit feature, and strictly adhering to free speech. One of the most closely watched issues is: how to create a "decentralized social media."

Just as Bitcoin symbolizes freedom, we can actually use similar infrastructure to run social media applications! Technically, at least a proof of concept can be achieved. In fact, as early as 2014, Casey Kuhlman, Tyler Jackson, and I proposed a decentralized autonomous organization (DAO) called "Eris," which can essentially be seen as a "distributed version of Reddit," capable of running on a blockchain backend (specifically Ethereum POC 3), as shown in the image below:

If you look closely, you'll see a "My DAO" button in the upper right corner, and the idea of a DAO was considered crazy by many at the time.

Although the prototype yielded nothing in 2014, the crypto market was still very immature back then, making it nearly impossible to distinguish between smart contracts and "popular pies" (just kidding). People viewed "DAOs" as a topic for Confucian discussion, but the situation is completely different now. Given my experience in the Bitcoin/blockchain space and considering that many venture funds are currently wandering in this field, I really wanted to build a product and secure a $20 million pre-seed round of funding. However, after going through one startup experience, I vowed never to try to develop or sell software again, so I guess I’ll just stay put in my law firm. (Note from Chain Catcher: Preston Bybne is a partner at the renowned law firm Anderson Kill.)

To be honest, designing a prototype is certainly much easier than designing something that people actually want to use. Even in the simpler "Web 2" technology space, there are thousands of social media applications, but only a handful achieve success. Creating a social media application is very simple, but operating a successful social media business is extraordinarily difficult.

Frankly speaking, there have been attempts to build "decentralized social media," but the results seem to be lacking. So far, the most successful attempt might be the free open-source social network software Mastodon, although it is not perfect, with some instances failing to scale well (as Donald Trump found out when trying to create the social media platform Truth Social, which mimicked Mastodon in an attempt to shortcut to social media stardom, only to discover that Mastodon's backend could not handle their traffic).

For the same reason, if you want to store everything in plaintext like Bitclout and dump every communication onto the blockchain, it's not difficult, but the problem is------ scalability will be affected. Facebook does not need to reach a consensus on global status, and they also delete data (by the way, Facebook generates over 4 PB of data daily). Any social media system attempting to mimic Bitcoin (like Bitclout) cannot put data on-chain; otherwise, in the end, that data will still be stored on a few nodes running in data centers (which is quite similar to Ethereum, isn't it?).

So, what other issues need to be considered when building a "decentralized Twitter"?

1. Illegal Content

First, there are legal issues.

It turns out that social media companies are inevitably subject to a series of regulations, such as data privacy laws and other regulations (some are uniform in the U.S., while others vary by country/region), which apply to the destruction and reporting of illegal content within the U.S., copyright issues, data protection, and mandatory disclosure of subscriber records. All of these factors need to be considered in the design of any "decentralized" social media application.

For a long time, lawyers have viewed the issue of illegal content as a major barrier to adopting decentralized storage solutions.

The most prevalent illegal content in the U.S. and around the world is what law enforcement refers to as Child Sexual Abuse Material (CSAM), and intentionally hosting such content can lead to severe penalties, such as hefty fines and long prison sentences. The crypto industry's response to this long-standing internet issue seems to have gone unnoticed, with some even completely ignoring it. In contrast, Web2 applications that host user-generated content, such as Reddit, Twitter, or Facebook, take a very proactive approach to such illegal content.

Whether centralized service providers or blockchain node operators, they are all considered "providers" under U.S. federal law (a term meaning "electronic communication service providers"), and they are required by law to destroy illegal content. Facebook and other companies use various software (including Microsoft's PhotoDNA) to automatically detect, delete, and report illegal content.

But to my knowledge, many blockchain-based services, such as StorJ or Sia, lack such controls (or have very limited controls), allowing encrypted data to be stored without creating user records, and do not require service providers (in this case, node operators) to have the ability to determine the legality of the stored data or assess the legality of the stored data.

It is very likely (in fact, highly likely) that decentralized data storage services are currently being used to host illegal content, and the node operators hosting these services may not even be aware of it. However, for a "decentralized social media" application, this kind of "blindness" is unacceptable, so when designing decentralized social media, it must ensure that law-abiding users can participate in the network and that content does not violate local laws. But so far, it seems that no blockchain solution with a storage component has attempted to address this issue, so if Musk wants to build a decentralized Twitter, he must address this issue in the design; otherwise, no one will run nodes for decentralized social media services.

2. Intellectual Property

Similarly, the intellectual property system is not well-suited for decentralized use.

Social media node operators will act as "content service providers," thus "providing transmission, routing, or connection for digital online communications… of materials chosen by users, without modifying the content of the materials sent or received," especially as specifically pointed out in the Digital Millennium Copyright Act (DMCA), these social media node operators must consider:

(a) The necessity of registering with the copyright office to receive safe harbor protection under the DMCA;

(b) Carefully handling hosted materials that may give rise to copyright infringement claims.

At the very least, addressing this issue may require implementing DMCA notice-and-takedown procedures for any third-party content hosted on the nodes (which would involve the node operators needing to submit their own opinions to the copyright office if they wish to profit).

Worse still, we may see some "copyright trolls" exploit copyright loopholes of social media node operators and then sue them for profit, as once copyright lawsuits are filed, it is very likely to result in financial loss------this is also something that certain unreasonable copyright enforcement law firms often do.

Currently, it is difficult to speculate on what kinds of infringement and enforcement issues may arise in decentralized social media. However, based on what we have seen in Web 2, once "copyright trolls" find it profitable, something similar will definitely emerge in the Web3 space.

3. Data Protection and Disclosure

Another issue arises when participants in a decentralized network may acquire a large amount of user data while operating their nodes------data protection and disclosure.

For example, in a decentralized social media system, the network will allow users to download user profiles and posts. Suppose I follow @A16Z, and @marmotrecovery follows me; @A16Z will be allowed to download and store my information and posts, as well as the information and posts of everyone who follows me, including @marmotrecovery. Given the absolute user count of @A16Z (half a million), it can be confidently said that if A16Z operates a node on this hypothetical network, they may be considered a "service provider" under California's Consumer Privacy Act or other local laws, and may need to implement compliance programs.

For the same reason, node operators may also become "electronic communication service providers" under the U.S. Stored Communications Act (18 USC § 2701 et seq.), and thus may be required to turn over records on their computers to the government without the government first obtaining authorization------at least to the extent that these records relate to third parties that the node operator owns and controls, but users are unlikely to want to operate a network that intrudes into their personal lives. Therefore, decentralized social media needs to be designed to minimize the storage of third-party data on nodes.

Finally------some rough conclusions about the future design of decentralized social media networks

It seems that decentralized Twitter is unlikely to use blockchain infrastructure.

Content removal and moderation will be the most important factor in the design of any decentralized social media system. Ironically, the unfair content moderation in Web 2.0 is also one of the reasons driving the creation of decentralized social media in Web3. At the very least, the centrality of content moderation to the user experience of social media means that simply dumping all content onto the blockchain, as Bitclout has done, is not sufficient.

On the other hand, FTX founder Sam Bankman-Fried (SBF) also believes that censorship is a major issue and has expressed a desire to fix the "broken model" of social media through blockchain technology. He views traditional social media platforms as a very chaotic system, with no interoperability between different platforms, such as not being able to see tweets on Facebook, and WhatsApp not being able to read messages on Facebook, even though they are products of the same company. SBF proposed that different social media platforms could extract information from the same underlying data and conduct independent moderation, thereby helping to address the issues plaguing social media while creating a fair competitive environment.

However, the first truly successful "decentralized social media" system will not attempt to be a world computer that can sing and dance, but will allow participants to replicate the absolute minimum viable information necessary for the network to operate. When using a social network, the only thing the social network may need to focus on is whether specific content is published by a specific person; as for the "blockchain" part, if there is one, it should be reduced to just providing a username and associated public key for registration, and nothing more.

To a large extent, the first successful decentralized social media service is likely to limit the types of data hosted by users to plain text.

First, from the perspectives of criminal, copyright, and data protection law, hosting only text written by users and possibly a select group of followers is a low-liability claim. Of course, hosting text will also be easier on bandwidth and simpler for P2P transmission.

Secondly, for hosting videos and images, it is simply because if there are no other reasons involving large amounts of data, it can actually be handled through "outsourcing." At this stage, although there are many third-party platforms (Bitchute, Cozy, Odysee) with lax content moderation policies for video content, this at least allows for "safe harbor," thus addressing the current market gap served by institutions like YouTube, and relieving node operators of the responsibility of moderating that content (especially copyright issues)------this is particularly useful.

To provide services for content, decentralized systems need to do not block external links pointing to these services (blocking external links is a practice that both Facebook and Twitter engage in), but rather allow users to control what others can see by operating a whitelist/blacklist of third-party content providers. In this case, decentralized social media will become another source of referral traffic for external websites (links).

Of course, the above understanding may also be wrong.

However, for this issue, simpler answers are more likely to be the correct ones, so------the future "decentralized social media" may resemble RSS more than Ethereum.