Research: How many "Want To Read" books are available to borrow?

[mekarpeles]

Feature Request

Problem / Opportunity

There are likely many books that patrons want to read that are in the library (i.e. borrowable).

https://openlibrary.org/search?q=readinglog_count%3A%5B1+TO+*%5D+AND+ebook_access%3Aborrowable&mode=everything
readinglog_count:[1 TO *] AND ebook_access:borrowable
Shows there are 843,233 books on reading logs and that are readable.

There are 45K books (English, 9+ reading logs, borrowable):
https://openlibrary.org/search?q=readinglog_count%3A%5B10+TO+%2A%5D+AND+ebook_access%3Aborrowable+AND+language%3Aeng&mode=everything&sort=want_to_read

Reading Log DB:
SELECT COUNT(DISTINCT edition_id) from bookshelves_books where bookshelf_id=1;
2,039,463 distinct editions
SELECT COUNT(DISTINCT work_id) from bookshelves_books where bookshelf_id=1;
2,775,731 distinct works

If 3% of patrons act on a message that a book of interest is available for free, that would be 1350 patrons returning to the platform.

Proposal

Breakdown

Related files

Refer to this map of common Endpoints:
*

Requirements Checklist

Checklist of requirements that need to be satisfied in order for this issue to be closed:

• [ ]

Stakeholders

---

Instructions for Contributors

• Before creating a new branch or pushing up changes to a PR, please first run these commands to ensure your repository is up to date, as the pre-commit bot may add commits to your PRs upstream.

[nexpectArpit]

Hey @mekarpeles , I'd like to work on this research. 1350 returning users sounds great.
I can do the analysis between "Want to Read" books and borrowable ones.
What would be more helpful : more data or breaking it down by categories? Also, could you give me some guidance on the approach since the breakdown section is empty?

[Pahari47]

Hii, @mekarpeles, can you breakdown it little bit, it would be more help full. and I made a PR on solr test integration, can you review it. and also, can't find any communication platform like if slack is available can you add me. If there any other issue you help me to find It would be helpfull.

[nexpectArpit]

Hey @mekarpeles! Following up on this. I've been looking at the codebase and here's what I'm thinking:

Approach:
1.Use Bookshelves.most_logged_books(shelf_id=1) to get all "Want to Read" books (shelf_id=1 is "Want to Read")
2.For each work, query Solr to check if ebook_access: 'borrowable' (which indicates books in 'inlibrary' collections)
3.Cross-reference the two datasets to find overlap
4.Generate statistics and breakdowns
Script Output:
•Total unique works on "Want to Read" shelves
•How many of those are currently borrowable
•Percentage overlap
•Breakdown by:
Language (English vs non-English)
Popularity (based on want_to_read_count)
Maybe time-based trends if useful
I can create this as a standalone script in /scripts/research/ that can be run periodically or modified for different analyses.

The main question I have is about scale - should I pull all "Want to Read" entries or would it be better to start with a sample (like top 10K most wanted books) to keep the analysis manageable?

Does this approach seems right?

[davidscotson]

A related feature idea would be to let users sort their Want to Read list by availability, so that available books are listed first.