Draft for Verses pluriverse essay
A critical piece of a pluriversal world is one that has interoperability ingrained into the foundations of the system. For us to be truly independent of platforms, the freedom to data portability and ownership must be a given. As we spend more and more of our lives online in different walled gardens, network effects have grown increasingly prevalent to determine how successful an app or platform is. For example, in the productivity space, every company has a slightly different data model for how content is created and how that content relates to each other. While you might work with text blocks in Notion, you have to use database rows in Airtable. While Coda has rich text, schema-powered documents, Obsidian uses a slightly enhanced flavor of Markdown (with wiki links).
While there is an incentive to do data lock for financial reasons, even if you are philosophically aligned and want to be interoperable or compliant with other protocols, there are very hard pragmatic problems. Current data standards and formats are limited in functionality and there’s no practical path to universal agreement around a common format for a structured data file type (the closest we’ve come is the CSV and spreadsheet file types, but even in there you see so many different flavors like SQL and all its variations and the new graph query languages). Each of these shine in specific contexts and feel lacking in others, so there are clear reasons for each of them. Maybe, then, it means that it’s impractical for all data to reside in a common universal data format, and instead, the answer is that we need to have systems that are adaptable enough to extend to new data formats and interchange them easily. Like a currency exchange for data formats.
What does a future look like where the data we create is community-owned and it moves as freely as the air we breathe?
Towards interdependence and interconnectedness
first level is free movement of data, every app has maintained and open api and import/export methods
- what to do about “proprietary” data? How are apps or services compensated for certain data that they produce. Perhaps you need to pay the service for every piece of data you take that they produce for you?
- Could you pay a levy on each photo that runs ML models for you? Right now we get it all for free from these large corporations because they want our data or want us using their platforms.
- What does the incentive model look like if all the data is open? Will companies still want to open this up? Yes, because then users will bring their data to these platforms?
There have been efforts in the space to imagine what it would look like to create interoperable apps in the absence of a universal data format.
- Cambria for making data formats compatible with each other
- reducers from graphQL?
- tanagram?
- using common data format (JSON) and providing parsers
i want it to be commonplace for anyone that is curious about a specific visualization of a set of related data to be able to answer that question without having to integrate with several custom APIs, which requires the site to provide access to the full fidelity of data * can this be like open-source? where “open-data” is as valuable as a stand-alone attribute with nothing else different from existing competitors? * google docs but easy access and manipulation to all data * camera app but easy access to location etc. * messaging app but easy access to content and relationships
Private Data
open questions
-
what about data that necessarily should not be public? or needs to be encrypted by the private key? so you do have access through your credentials?
- can you have anything that isn’t public on the blockchain for state per-user? Like how would you implement google photos on the blockchain. is it just that all data is on the blockchain
-
who builds the API to retrieve this? or is this built into ethereum?
-
how does crypto intersect with 0 data apps?
-
tools like coda and spreadsheets and stuff give us the flexibility of manipulation
- on top of that we just need access to interesting data.
- social graph, decentralized network
- data that we already use and create — our user generated content
- what does this world look like
- spreadsheets let you embed your own logic on your own data
- how do you democratize logic/algorithms?
- how do you democratize access to your data?
Early Draft
- people to ask about opinion
- ask feifan about opinion and get deep-dive on telegram
- ask geoffrey litt about opinion
- michelle lim
- alicia
- linus
In the wake of a lot of corporate distrust and backlash against large tech companies, the past couple years have seen steady momentum around the movement for individuals owning the full rights to their digital devices and digital content. We’ve seen this from Andrew Yang’s campaign to own data, the right to repair campaign, and the crypto movement. There’s an overwhelming energy around seizing back sovereignty in the ephemeral digital sphere. The right to repair is gaining speed because it’s access to manipulating the physical digital devices. The crypto movement is popularizing the ability to own and monetize content and form transparent governance contracts for communities, but I haven’t seen a convincing solution to the problem of people owning the content and data that they generate as well as having the freedom and tooling to manipulate and act upon that data.
I’d argue that the network effect is crucial in a lot of spaces even outside of social media. In the “tools for thought” and productivity space, data and user-generated content is everything. Companies are incentivized to create the data moat and add incompatible logic and serialization forms because not only do they provide a competitive advantage feature-set-wise but they also provide a moat against transferring the content at full-fidelity to a different app. Some apps deliberately invest a lot of time and space into this area (craft being a great example), but it’s time-consuming and doesn’t make good business sense. As a company in this space, it’s an area of moral dilemma between creating the overall best consumer experience and creating the best business, a conflict of interest between shareholder and universal consumer desires. It’s a grey area because there’s a lot of good in providing new functionality to manipulating and displaying data, and we need this innovation in the private space in order to push it forward at the protocol or standard level. Perhaps one day markdown will be supercharged with spreadsheet and database manipulation as well as the standard text displays.
Notes
- my ideal world is one where we have democratized the logic and algorithms that power our apps AND we’ve democratized the access to our created data. We need both the flexibility of manipulation that the primitives of spreadsheets and databases provide us, which will allow anyone to [bring their own client](bring your own client). However, this flexibility and access to new primitives of manipulation are useless when you don’t have data that is interesting
- What are examples where we’ve done one but not the other? Make the argument that they are codependent on each other so it’s meaningless to create one and that’s none of these have been successful so far
- i want it to be commonplace for anyone that is curious about a specific visualization of a set of related data to be able to answer that question without having to integrate with several custom APIs, which requires the site to provide access to the full fidelity of data
- can this be like open-source? where “open-data” is as valuable as a stand-alone attribute with nothing else different from existing competitors?
- google docs but easy access and manipulation to all data
- camera app but easy access to location etc.
- messaging app but easy access to content and relationships
- tools like coda and spreadsheets and stuff give us the flexibility of manipulation
- on top of that we just need access to interesting data.
- social graph, decentralized network
- data that we already use and create — our user generated content
- what does this world look like
- spreadsheets let you embed your own logic on your own data
- how do you democratize logic/algorithms?
- how do you democratize access to your data?
- some prior art in this space
- michelle and other guy talking about decentralizing the data in social media sphere
Areas to explore
crypto
crypto is a big space that touches on this. In theory, everyone owns all their own data and everything is transparent.
open questions
- what about data that necessarily should not be public? or needs to be encrypted by the private key? so you do have access through your credentials?
- can you have anything that isn’t public on the blockchain for state per-user? Like how would you implement google photos on the blockchain. is it just that all data is on the blockchain
- who builds the API to retrieve this? or is this built into ethereum?
- how does crypto intersect with 0 data apps?
who to talk to
- talk to jonah and get who else to talk to about this
zero data apps
there’s a movement of apps that use no persistent data storage unless you provide your own or using just the browser’s storage. In this way, they explicitly do not have the ability to sell your data or otherwise use your data for their own gain. Nothing is sent to the server, only to the servers that you specify or your browser.
related work to explore
- sql in your browser, can you just store everything in the browser? how to make it persistent tho?
export mechanisms
data standardization
Related Work
personal data ownership universally interoperable data surface Reservoir