Skip to content

Instantly share code, notes, and snippets.

@tmcw

tmcw/library-json.md

Last active Sep 12, 2020
Embed
What would you like to do?

Hi Tom!

Here's what the Markdown for a book looks like on macwright.org:

---
categories:
  - book
layout: book
title: 'My Year of Rest and Relaxation'
publisher: 'Penguin press'
published: '2018'
author: "Ottessa Moshfegh"
isbn: '0525522115'
lccn: '2018006199'
oclc: '1007559349'
openlibrary: OL26959514M
goodreads: '36552920'
isbn13: '9780525522119'
rating: 2
---

So I started reading this before things got bad, globally-speaking, but kept on going despite reading a book about a woman staying inside all the time as I was staying inside all the time.

I don't think it's that good. It's not terribly insightful or cutting. The micro-structure - the sentences, metaphors, asides - aren't especially interesting. Where I found myself highlighting bits of _The New Me_ or _Conversations with Friends_, and I can still offhand remember a few truly amazing bits of those - I just didn't, for _My Year_.

Now, I read this as the world was collapsing outside my window as I calculated how many weeks of Clif Bars I had stockpiled, so take this with a grain of salt.

This novel was most worthwhile when it was the least realistic: when it's quoting Dr. Tuttle, her psychiatrist, whose quack-therapy suggestions are hilarious, and near the end, when the plot edges into magical realism. But the schtick of the self-loathing, unlikeable beauty queen is always in-between: not fleshed out enough to be understandable as a person, not stylized enough to work that well as black humor.

Thoughts on the post:

Book index

Yes! That's one of the core problems. I built and still use bookish.tech to give me cross-reference IDs in a number of systems. The hope is that there's be this sort of thing "as a service" as one of the components of a good system, and it would allow people to quickly match books across systems. It does pretty robust matching between goodreads, oclc, lccn, isbn.

Amazon is a trap for this problem imho: the only reason why I think Amazon-scraping should be a part of any system is in ASIN-to-ISBN conversion.

library.json

I feel a little divided about this. For interchange, It seems like there already at least three competing standards:

  1. Atom feeds, which can represent lists of books pretty darn well
  2. The review schema, which is implemented on macwright.org and afaict nobody ever wrote parsers or aggregators for
  3. JSON-LD which is also implemented on macwright.org and feels mainly like another trap that google set to make web people bend to their will and then they promoted the same sites that they were going to before

I'm not quite as opposed to RSS/Atom as you are, but also I probably haven't used them to nearly the same extent.

Re: the proposal itself -

  • My highest priority has been 'if two people review the same book, how do we interlink those things?' My general concept has been that we'd use library ids, like oclc ids, to do this. At least to my tastes and current levels of outside-world angst, amazon urls as a matching strategy would be a hard pill to swallow. I'd kind of assume that this part of the spec is open to change, though. This whole ID concept I think is both essential and hard to get right: you aren't guaranteed an ID of every type for every book.
  • Structuring this data is a rabbit hole but imho some part of the hole is worthwhile. For example, seeing worldcat's structure, which allows for multiple authors, seems like it preserves some essential details that otherwise would break most UI concepts - you'd have a toggle for "Douglas Adams and Mark Carwardine" as well as one for "Douglas Adams". Even stuff like date formatting - as in the comments already - is stuff that if you extended something else, you wouldn't have bikeshed risk.

That said, the thing that kills the indieweb is too much enthusiasm for specs, too few parsing implementations, and, ironically, too much focus on the 'indie' (building complicated self-hosted everything-machines) and not enough on the 'web' (noticing if anyone's using any of the things you built). So if there's a killer implementation and good content at the start, then momentum would potentially just carry you through.

@tmcw

This comment has been minimized.

Copy link
Owner Author

@tmcw tmcw commented Apr 16, 2020

Oh, one more thought: one way to hack around the problem of "a book aggregator being valuable" is to make this behave as both a webring and an aggregator. So like, I'd love to cross-link other websites that review the same books that i've been reviewing. Or to link to a page identified by the book's isbn that shows other reviews, or show an aggregated review next to mine. Which would both be valuable for me, as the humble proprietor, and eventually a source of interest for the aggregator.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
You can’t perform that action at this time.