This post might not resonate with you if:
- You don’t have any digital information that you want to preserve long-term.
- You’ve never experienced a major data loss incident, therefore, it might be hard to realize the importance of data backup.
- You’ve never encountered vendor lock-in or platform-specific data loss (e.g. with SaaS or “the cloud”), they reveal the risks of relying on non-portable, proprietary data formats and services for data storage.
In the world of software, change is constant. Businesses come and go, and the same is true for software applications. But when it comes to ensuring data longevity and future-proofing, the only data format that I believe is: Plain text.
Unlike other formats, it remains reliable and accessible over time1.
Avoiding complex formats is crucial for data preservation.
Even venerable3 software like Microsoft Office can’t guarantee seamless compatibility over time, old docx files often fail to be recognized by modern Office versions, and platform differences further complicate matters.
Embracing plain text and simple, widely adopted4 formats is something everyone can do to make their best effort against the volatility of software and ensure the accessibility of their data well into the future.
This topic was further discussed in The Zettelkasten Manual.
You can find a (plain) text editor on any operating system made for practical use. Devices from decades ago can read and write plain text. ↩
The format’s specification should be easy to understand, making the implementation of reader and writer applications relatively straightforward. This, in turn, increases the likelihood of successful data retrieval in the future. ↩
Modern Windows 10 can run binaries that were compiled 20 years ago for Windows 95 or NT4. How has Windows maintained binary compatibility for over 20 years? ↩
To consider more aspects: publicly standardized, unrestricted, well-documented, and self-documented. ↩