Metadata

[jadhvaryu]

Hi!

I recently started maintaining project gutenberg ebooks (html formats). I realize that metadata of the books is extremely limited.

Rarely i find proper authors, genres, or ISBN#

I not using Calibre.

I am hoping to get guidance on best way to complete the meta data. I am also open to consider epub format if that has better metadata quality and completeness.

(For full disclosure: the end objective of the project i am part of is 'for profit' from the reading device.)

[Quoth]

ISBN applies to particular editions / formats on paper of the source of the content, so more than one per Gutenberg text. An ebook can have an ISBN. But it's not relevant to ebooks from Gutenberg or Amazon.
The ISBN is used by retailers (or the public) to order a specific version of book. Library programs can store an ISBN, but should be using a unique per copy ID (barcode, QRcode or RFID etc) that isn't EAN/ISBN type of barcode on the book.

Genres are subjective and not all published works have an officially assigned genre.

Either use Gutenberg or a professional Library program.

Also titles are not at all unique. There is no copyright on book titles. I've two unrelated books called "Dancer's Luck".

[Hitch]

Originally Posted by jadhvaryu

Hi!

I recently started maintaining project gutenberg ebooks (html formats). I realize that metadata of the books is extremely limited.

Rarely i find proper authors, genres, or ISBN#

I not using Calibre.

I am hoping to get guidance on best way to complete the meta data. I am also open to consider epub format if that has better metadata quality and completeness.

(For full disclosure: the end objective of the project i am part of is 'for profit' from the reading device.)

Well, I said this on your other thread, but...fwiw: The sort of people that go to Gutenberg for reading material pretty much already have their own readers. OR, you can just click "HTML" and read in the browser, if you wish. OR, download whatever format and put it on your own device or phone.

Not to mention, some huge percentage of them are already available on Amazon, as (somewhat) formatted eBooks--click and they magically appear on your device. It's not just Pride & Prejudice. We're talking about the same business that has 90% of the English-speaking marketplace.

I'm delighted that anyone thinks that all the FREE books in Gutenberg are somehow going to have enough of a marketplace, a buying demographic out there, to invest money in building a dedicated app, but...I mean, we're living in a world in which few people under the age of 50 watch movies filmed in B&W or before the 80's. So...???

{shrug}. It's your time and your nickel, but damned if I see how you'll monetize this in any sensible way.

Hitch

[phillipgessert]

Love PG and hate to use this expression for it, but it's garbage in, garbage out. If the data's missing, there's nothing for it. Hypothetically you could cobble something together that scrapes some other free resource for the missing data, but it really just moves the problem--someone somewhere had to list it. And it's another potential point of failure.

You might be better off scaling down to their top 100 or something, and having a human add the missing info. There are some real obscurities on the site and I would be surprised to learn there's a hidden trove of rich metadata for everything, or even most things, on PG.