The container could be whatever. If there exists a text, there probably exists a book (the content, in fact, is not «floating around»). If that is magnetic tape, so be it. And if that is an abstraction for a physical container - a file, i.e. a labelled pointer to an array of data -, again, so be it.
The file is the book; its physicality - irrelevant and interchangeable, but given - could be the magnetic tape, cards etc. that host it.
https://www.etymonline.com/word/book
It's also worth noting that what we now refer to as a "book" is more properly a codex, also ultimately from a reference to a tree trunk, but defined as a collection of leaves of paper or parchment (or folios or signatures) sewn into a binding with covers, as distinguished from a scroll.
https://www.etymonline.com/word/codex
https://www.etymonline.com/word/code
And of course, scrolls and tablets were the original forms of books.
The notion that a book must be a specific physical representation of paper, binding, and covers, or even physical is ... at odds with origin and usage.
I wrote that explicitly ;)
Note also the other side: something is a "text" when it is textile in quality, in the context of elaborating the expression of ideas: you "weave" the expressions of your thoughts into a "text".
Also note that "liber" means bark like 'book' does (though on the technical, craft side of "peeling"), and "paper" clearly refers to papyrus.
You have terms for the container and terms for the content. The "book" is not a «specific physical representation», but is the side of the container.
"Textile" (a woven fabric) is another great one, yes, thanks.
Diving into other etymologies and origins of other words based around information (literally, to form an impression), truth and trust, etc., is interesting. Many of the words for the most ephemeral concepts are deeply ... rooted ... in the profoundly physical, tactile, and material.
To my "school", a necessary duty :)
> in the profoundly physical, tactile, and material
That depends. Some scholars identify a number of key concepts as construction blocks of their many derived terms. Not all of them are concrete - some are extremely subtle (also to be productive of so many different, yet related concepts).
For example, you write 'physical': some identify the root of that term in an immaterial concept of a dynamic principle, as if expressed by the explosive phoneme, that pertains immediately to "physical", "future", "foliage" in this construction assumed to express a "spawning". Similarly to 'material' ("measure", "mathematics"...), that some see as coming from the key root concept of limited extensions.
In the development of the terms and idea, concretion and abstraction may follow each other ("measure", "moon", "month"...).
You don't like it.
The problem doesn't lie with the answer.
Here's a rhetorical question: When I read a Kindle-format book from Amazon on the Kindle Cloud Reading app on my laptop, what did I just read? Was it an eBook or not? If not, was it a book in any sense? If someone asks, "Did you read the book, The Cuckoo's Egg?" I'd answer yes, even though I read it electronically -- which is, I presume what the e in eBook stands for. Or do I answer, "No," because it was just digital data on a computer -- absent a standalone eBook device?
I have an eBook reader that came as part of the Calibre package, at least they call it an eBook reader -- it's just a program to read epubs. Is it is, or is it ain't an eBook reader?
If you separate the content from the hardware, you hardly have a "book", e- or not. You have some content floating around, and big machines that are difficult to use and even more difficult to access (cost, scarcity, location) to read it.
That's probably the genius of inventors: they see things that aren't there.