Keeping all your data in version control (preferably checksum-verified, like git, and regularly synched to remote storages) ensures that it’ll likely survive at least as long as you do, while not taking away the freedom to incrementally modify/organize it (as would be the case with large archive backups). This encourages you to actually keep the data organized, and to know which data is unique, to be preserved, making it more useful as a result.
Keeping all your data in version control (preferably checksum-verified, like git, and regularly synched to remote storages) ensures that it’ll likely survive at least as long as you do, while not taking away the freedom to incrementally modify/organize it (as would be the case with large archive backups). This encourages you to actually keep the data organized, and to know which data is unique, to be preserved, making it more useful as a result.