Please add (I believe this should be trivial) all the tar's metadata-related features (I mean saving links and access rights). There should be only one, all the archive formats different from 7z have to die. TAR and ZIP are both horrible. and should be displaced by 7Z.
7z makes the same mistake as zip: implementing both a filesystem and a compression algorithm. There's no way a single tool can implement all the bells and whistles of the various filesystems in use today. For example, skimming through the 7zip format ( https://py7zr.readthedocs.io/en/latest/archive_format.html ) I can see specific support for file attributes from (current versions of) Windows and Unix, but no AmigaOS protect bits ( http://www.jaruzel.com/amiga/amiga-os-command-reference-help... )
Keeping these two tasks separate allows swapping-out the implementation of each (e.g. I tend to use .tar.lz these days, since I'm mostly on Unix)
once you have a general file-system representation format (with all its complexities) adding compression of the blobs seems a minor addition.
On a related note I was surprised to discover that the Windows 10 backup tool was able to store about 240GB of various data in barely 80GB of backup; I believe it must have spliced most files to look for common fragment to deduplicate (with some NTFS magic behind maybe). .tar.xz will never be able to do that, if I try to compress 10 copies of the entire firefox codebase, it will never be able to recognize the duplicated files; only something like a file-system+compression could do that.
borg handles this just fine. I put all kinds of stuff into borg repositories: raw MySQL/PostgreSQL data directories, tar archives (both compressed and uncompressed), or just / recursively. You can do stuff like:
$ tar -caf - / | borg create …
$ borg create …