A book is not a file.

[transfire]

Would you be happier if we called files "objects" instead? It's just a name given to a on-disk data structure. Call it what you will. In the end it is just a stream of bits.

However I agree with you at a higher level. Our computer "file" systems have not evolved. And now they are being displaced by hundreds of isolated apps the manage out content. e.g. You might own the Jungle Book in your Nook account, but your copy of Peter Pan is on Amazon. Worse still, you don't even own these books anymore, they are instead a lease, which can be revoked at any time.

[IpxqwidxG]

Actually it seems correct to me to put books and files in the superset called "objects", but not name that superset "files" by itself.

Ownership and portability of a book is one thing that becomes relevant when price includes mark up to be paid to own it.

[Turing_Machine]

I guess technically the blog posts the author touts aren't "files". They're (usually) database entries stored in a database file. I'm not sure why this makes a difference, or why a database entry has more "creative discourse" than a file.

I'm pretty sure that old-school publishers and authors kept (paper) files of their works, too.

[zrail]

A hosted system where you can write and edit a book-length document and then one-click publish to various platforms, including direct sales, as well as dead-tree prints, would be pretty amazing.

[IpxqwidxG]

Yup! And that system need not have two hundred features and formats for the author to learn and master. I've been stoked by the idea for quite sometime.

[Turing_Machine]

That's kinda what leanpub.com does, isn't it?