What is the human unreadable in a stream obj of the uncompressed PDF

As It's supposed to be uncompressed by

qpdf i.pdf --stream-data=uncompress o.pdf

but then there's still some sequential lines which is human unreadable in some stream objects

So confused what it is actually or usually, so what is qpdf CLI to have a "100% real" uncompressed one?

Streams in PDF can contain either text or binary information. Examples:

• XMP information: textual
• Font: binary
• Page description: textual
• Image: binary

What happens when you uncompress them, is that you get the uncompressed data, but this uncompressed data is then either textual or binary depending on what the stream is used for.