Tech giants are "monopolizing" the world; how can we avoid becoming "slaves"?
This article is authored by jonstokes and translated by Yin Jiansong.
I am an antitrust advocate, so for the past few years, I have been thinking about how and why we should break up large tech companies through antitrust legislation.
It is precisely because I have been contemplating the specifics of how to dismantle large tech companies and limit their scale that I recently realized the entire XaaS (X as a Service) application technology stack is about to be massively disrupted by blockchain in a very unique and underappreciated way.
From small B2B infrastructure cloud services to consumer-facing SaaS giants (like Facebook, Google, Amazon, etc.), every large-scale software company will face the fate of being cut loose.
What is coming is: public blockchains, equivalent to a unified, super-large user data table for the entire internet. The next wave of distributed applications will be built on top of it. This has various implications for markets and politics.
1. Why User Data Tables Matter
As a society, we recognized long ago that if we allow the banking industry to operate completely unregulated, we would ultimately face these massive, shaky entities that would drag us all down from one crisis to another.
Therefore, when we set out to impose regulatory constraints on banks, we used some simple, hard-to-manipulate numbers as indicators of bank size and systemic risk.
There is still much debate about how much these indicators—such as Tier 1 capital ratio, total assets, total revenue, etc.—can tell us about the overall systemic risk of banks. However, we had some key regulatory numerical targets, which allowed us to limit large financial institutions in the past when we believed it was important to do so.
So what are the analogous metrics for social networks or other XaaS products? I believe the answer is clearly the size of the platform's user data table (user information collection).
Every web application, from the initial MVP to Google, has a user data table. Its function is to store information about user logins and software usage—email addresses or phone numbers and (encrypted) passwords are the basic information you can find in this table, but you will also find a wealth of personal privacy data there and in connected tables.
Your users register through social login buttons—such as those integrated with Google, Facebook, and Amazon—you still have a user data table that records specific details of user interactions, such as transaction history, reputation markers (experience points, badges, and user ratings), social link graphs, and activity information streams, which you can monetize, for example, by selling ads, raising new investments, and so on.
What the user data table does most critically is that it helps you achieve network effects, which provide you with platform lock-in.
The more users your platform has, the greater its value to any individual user (refer to Metcalfe's Law). The greater the value to any individual user, the easier it is to attract new users, and the harder it is for any user to leave (because leaving means giving up all that value).
Thus, the entire online attention economy is built on proprietary user data tables, with different applications carefully protecting these proprietary user data tables and constantly striving to grow them.
In this case, the size of this data table directly measures the size of a tech platform. It is not a proxy measure. It is truly direct. Because it is essentially the same number, regardless of whether it is the internal number used by the platform or the external number used by partners and investors.
If you want to limit the size of a platform, then you need to do something to limit the size of its user data table.
If I were to break Google into two components—search engine and YouTube videos—I would require these two companies to maintain completely independent user data tables. Then, I would prohibit these two entities from accessing each other's user data through non-public interfaces. Therefore, if YouTube wants to serve Google users, it must enable Google integrated login like all other services.
2. What Independent User Data Tables Mean
In the current web system, all these independent user data tables enforce a specific interoperability architecture, even for applications that wish to open their user data fairly. That is, every third-party application that wants to access user data must call an API.
The result is a decentralized network of isolated data tables connected by access control APIs. Depending on the relationship between any two nodes in the network, they can access each other's data in different ways through this temporary, decentralized system.
Let's take LinkedIn as an example to see what this means in practice. If I register on your new website using the LinkedIn integrated login button, your website will not automatically know my user information. To understand my identity, the website must use the API provided by LinkedIn to retrieve this information. (I am not sure if LinkedIn actually exposes this data through an API, but that is not the point).
This arrangement has several key implications:
I can only access the user data that LinkedIn chooses to publish through the API.
LinkedIn can revoke API access to any data at any time, so building a business that relies on accessing LinkedIn data is insecure for me.
For me, if I were to build such a business, it would also be insecure to anger LinkedIn, as I would have a significant dependency on their user data, and they might revoke my access to that data without revoking my competitors' access.
If something bad happens to LinkedIn, such as Google potentially acquiring it and then shutting it down in typical Google fashion, then that API and all the support behind it would disappear.
I will never be able to write my own data into the LinkedIn user data table for others to see unless LinkedIn grants me permission to write data. Again, even then, I could lose that access if it is revoked (for everyone or just for me), or because LinkedIn is no longer available.
The essence of the above points is that LinkedIn's ongoing control over its proprietary user data gives the company leverage over all users of that data and exposes all users of that data to various LinkedIn-specific business risks.
This combination of leverage and risk exposure greatly limits each entrepreneur's ability to build applications relying on data managed by LinkedIn.
3. Blockchain as an Internet User Data Table
Instead of isolated user data tables connected by APIs, blockchain has a unified decentralized user data table accessible through open protocols and a network of storage nodes. Thus, identity-hosting blockchains represent the decentralization of the data storage layer and the re-centralization of the data access layer.
I will illustrate what I mean using the blockchain of BitClout (a crypto social platform), which I have been thinking about for various reasons recently.
With BitClout, I have an identity on the blockchain that anyone running an open-source BitClout node can publicly access.
Continuing with the LinkedIn example, if they were to rebuild themselves on BitClout, they would implement the entire user data table and conduct evaluation operations on the BitClout blockchain. And this data would appear on every BitClout node, available for other BitClout nodes to access or add new records.
If LinkedIn or its internal evaluation of the public recognition of data value on-chain encounters any issues, the public can still access old data permanently and without restrictions.
Moreover, I can easily create a service that allows BitClout users (now also LinkedIn users) to continue recognizing each other's various skills under their original identities (which is a core function of LinkedIn). I could even add new skills and recognition options beyond what LinkedIn already offers.
Of course, even if LinkedIn's user data table is actually placed on the BitClout blockchain, LinkedIn could still require access to its users' evaluation data through a private API.
However, deciding to keep data isolated in this way comes at a cost. My guess is that the temptation to leverage the network effects of a scaled blockchain (Metcalfe's Law) will be so great that companies will default to placing user data on-chain rather than keeping it closed.
Imagine if LinkedIn, Reddit, and GitHub all migrated their user data tables (along with many of their private data, such as evaluations, points, and activity history) to BitClout. What would happen next: every GitHub user would also be a Reddit user, LinkedIn user, and BitClout user.
Similarly, every Reddit user would also be a GitHub user, LinkedIn user, and BitClout user. I could go on, but you get my point.
Every company built on the same virtual user data table would have direct access to the data of other startups on that table, creating network effects. Whenever a chain-based company adds a new user, your service also gains a new user. (In a sense, they may not actively use your service yet, but they effectively already have an account).
In contrast, the network effects based on user data tables are no longer an economic moat, no longer providing anyone with any kind of defense or investment advantage.
The game theory of this arrangement is as follows:
All existing large companies initially want to maintain the proprietary nature of user data tables because that is their moat, providing them with network effects and the lock-in effects valued by markets and investors.
However, to some extent, blockchain user data tables (along with the ecosystems built on them) will grow to a scale where existing enterprises cannot compete alone. When the virtual user data table of BitClout surpasses the proprietary user data table of Twitter in scale, Twitter will either join the blockchain or be left behind. The same situation will occur with Google and Facebook.
3. Ultimately, there will be a blockchain user data table larger than the sum of proprietary user data tables. When this happens, these giant platforms will either abandon their proprietary data tables and reconfigure their businesses around the blockchain or enter a doom loop.
Imagine a world where every startup has as many users as the largest existing enterprises. There is no friction when users onboard or register—at least not friction that is as mandatory as it is now (you can reintroduce it).
If you are standing on a large identity blockchain, then every blockchain wallet is already an (inactive) user of your service, and they have a seamless way to transact with you in money and digital goods.
Businesses need to spend some money to reach their users and convert them from inactive to active users to use the specific services of the business. I believe any chain will have a paid email service because otherwise, spam would render the entire system unusable.
However, companies can query these blockchains to identify a subset of target users and then communicate with them or provide services in any way they can afford.
If you want to build a set of network effects that particularly benefit your company, merely cultivating a large user data table or email list is not enough—you must provide some on-chain services that others are also motivated to use so that the services you provide can spread and become a currency.
For example, if I am the only one providing a specific credential service (such as an on-chain service for user verification through Twitter), the more users use that credential in their transactions, the larger my on-chain network becomes.
How do credential providers gain value from this? I don’t know, but it’s not hard to come up with some random possibilities. For instance, if you accept my credential, then perhaps you have to accept my marketing messages, or for a fee for receiving my information, offer significant price discounts; I could use this discount to communicate directly with you or resell it. This is just a random example I thought of, and I am just an ordinary person—I am sure the market will provide more.
4. Conclusion
Moving identity on-chain, thereby eliminating the possibility of network effects centered around user data tables, completely disrupts the entire landscape of interoperability based on API access control that the current internet is built upon. All the non-technical market and political dynamics surrounding the size of user data tables, leverage, and risk suddenly disappear.
I do not know what the outcome will look like, but I am quite sure we are not clear about it yet. The ongoing dialogue about technological consolidation and regulatory frameworks makes no sense in a world with only one user data table.
In other words, there is now another massive, growing entity, besides the government, that can stand against large tech companies.
Original link: https://mp.weixin.qq.com/s/haCfhPLVtC-mSHHm4n_WWA
