> You can build random access around tarballs, just need to index the header data.

> Granted, you do have to traverse the entire tarball.

So you can't randomly access a tarball, you can cache the linear access you've already done.

You can read all the headers without reading the whole file by just seeking over the file data....

But you would still have to decompress it. How much support for legacy systems do we need vs just making a slightly better version of a jar?

Decompression has nothing to do with tar, though. But I agree, it is painful to deal with tar+gz.

> Decompression has nothing to do with tar

Which is exactly the problem. Same issue occurred via volume and filesystem management resulting in ZFS. We need systems that compose and also elide.

So something that both archives as set of files and compresses them w/o losing affordances over the layer below it in the process.