@atoponce the poll is rigged
@atoponce many of us archivists are not on fedi. it's likely you will see n(x) % of results being linux/bsd users with tarball fetish.
i like tarball myself, but i also have a different approach to archiving
@ax3 I'm a CPIO fan personally.
Mostly because I learned that there were different flavors of tar(1) that were not compatible between Unix and Linux in a heterogeneous environment (mostly HP-UX, Solaris, and RHEL).
But then I learned the advantages that find(1) and cpio(1) can do when combined together and it's just so much more powerful.
That's a pretty unique edge-case where your combination prioritizes error-handling over ubiquity.
Personally, I've been good so far with tar and the various flavors of external compression (gz, bz, xz), and further optionally combined with gpg for encryption.
Question: do you choose cpio for it's better integration with find?
@UncleIroh @ax3 I've gotten good enough with it that if I need to do any exception handling for file matching that just isn't "everything in a directory", I reach for cpio(1).
EG, archive files order that 30 days, or archive all files owned by a specific user, or only if not empty. That sort of stuff comes up a lot for me.