# The data spectrum: portable, usable & interoperable

> Moving beyond just exportable data.

**Published by:** [Colin Armstrong](https://writing.cma.xyz/)
**Published on:** 2024-10-15
**Categories:** web3
**URL:** https://writing.cma.xyz/data-spectrum

## Content

Social media platforms own the relationships you have with your fans, as well as the data you generate interacting with them. Building an audience on Twitter means your interactions with your followers are restricted to Twitter. If you’re banned from Twitter, you can’t simply move to Facebook and begin interacting with the same followers. Similarly, all your tweets, retweets, likes, DMs - pretty much every interaction you have - is stored on Twitter servers and inaccessible to you. Some data is exposed via API, but Twitter has famously discouraged 3rd party development, even after developers built businesses atop these APIs. Once a developer emerges as a competitor, Twitter is incentivized to cut off their data access - of course they’d prefer if all engagement and interaction happens on Twitter, not on this developer. Some data is *exportable* - you can login at any time and export a list of your followers and tweets. Though, if you export a list of Twitter followers, this isn’t very useful - you *still* cannot import them into Facebook. This data is *exportable*, but not *useful.* Other social media platforms have varying levels of exportable data, and varying levels of useful data. Substack, for example, lets you export a list of your email subscribers at any time. These are email addresses you can contact at any time - even off platform. Mailchimp also lets you export subscribers, but earlier this year they banned & locked out several popular cryptocurrency newsletters, preventing them from accessing their subscribers - even though they were usable \ --- MailChimp issued a ban on several popular cryptocurrency newsletters earlier this year. Mailchimp supports exporting your subscriber list at any time, but these bans locked users out of their own subscriber lists. Building an audience on Mailchimp, even via email, was risky. \ \ When you use social media apps today, your data is stored on their servers Data is portable if you can Data portability is good - it’s the minimum of what we should expect - but I’d argue that data *usability* is what we should strive for. mn Using a social media platform today means your data exists Facebook, Twitter, Substack, Instagram, Google, Medium - each has varying levels of data portability. Building your audience on Twitter There’s incremental levels of usability: **Audience built on-platform. Interaction must occur on-platform.** Twitter, Facebook, and Instagram are examples of this. • audience built on platform, interaction must occur on platform (twitter \n • audience built on platform, interaction can occur off platform (Substack audience export) \n • Audience built on or off platform, interaction can occur on or off platform (decentralized social protocols) --- Data exists in silos. All activity on most websites you use today are stored on that website’s servers. Often, You can export your subscribers from Substack. You can download your tweets from Twitter. You can download all your data from Google Takeout. This data existed in these company’s silos, then you download them and it exists on your hard drive. \ *But you can export data out of Substack!* Sure, but why should I be forced to move off-platform if \ Usable data means that anyone can \ Examples of unusable (yet portable) data: * I’d like to add categories my Substack posts. I can export my posts from Substack and build my own solution. * I’d like to reach out to my Twitter followers via email. I can export my Twitter followers, but I’ll struggle to reach them off-platform. Examples of usable data: * I don’t like this particular open source software. I’ll fork it and * I hate the public-facing UI of [Paragraph](https://paragraph.xyz). Let’s create a new frontend that permissionlessly reads the data from Arweave. * [Farcaster](https://farcaster.xyz) lacks a bookmarking service. Let’s build [Perl](https://perl.xyz) on top of it permissionlessly. \

## Publication Information

- [Colin Armstrong](https://writing.cma.xyz/): Publication homepage
- [All Posts](https://writing.cma.xyz/): More posts from this publication
- [RSS Feed](https://api.paragraph.com/blogs/rss/@colins-blog): Subscribe to updates
- [Twitter](https://twitter.com/colinarms): Follow on Twitter
- [Farcaster](https://farcaster.xyz/colina): Follow on Farcaster

## Optional

- [Collect as NFT](https://writing.cma.xyz/data-spectrum): Support the author by collecting this post
- [View Collectors](https://writing.cma.xyz/data-spectrum/collectors): See who has collected this post