Best books, many metrics, impossible questions ⁂ Antilibraries Analects ⁂

There are a number of themes I've been circling with the current iteration of this newsletter: reading communities, book reviews, the future of bookstores, modes of engaging with books, my own reading practice.

One area I've been trying to figure out how to write about more concerns books and quality, judgment, ratings, canonicity — dancing around the constellation of things that make a book a good book; that identify a book as one of the best books.

From the start, trying to make a list of "best" books is fraught: best to whom? Best by what metrics? Best for a particular context, or best — as best we can decide, so far — for all time?

Antilibraries is in large part about reckoning with these questions, about exploring where specific books live in the matrix of might/should/could-readability, as well as about wrestling with the meta-level conundrums of curation and classification and value.

I have scatted notes on ideas for potential book rating metrics, collections of particular best-of lists, places to source good summaries and recommendations… For now I think do best to split this up and write about one narrower slice at a time.

I'll start by sharing some ideas I brainstormed for book rating metrics. These emerged from conversation with David Laing, and roughly aim to answer the question: if you designed a book app that let you rate books numerically not just on a single vector (e.g. is this a 1 star or 5 star read?) but on 5, or 10, or 20, what would they be?

I first came up with a long list:

Plot, Craft, Fun, Beauty, Information quality, Information density, Novelty, Distinctness, Fit, Impact, Realism, Success, Truth, Clarity, Immersiveness, Memorability, Resonance, Inspiration, Timelessness

Some, like Plot, or Truth, apply only to fiction, or only to nonfiction. Some, like Success or Clarity, might be measured on a book's own merits. Others, like Fit and Impact, might be measured only from a personal, contextual point of reference. And some, like Timelessness, might not yet be measurable at all.

From these, I narrowed to a short list:

Craft (writing quality), Fun (enjoyability), Beauty (aesthetics) Information (quality and truth value), Distinctness (novelty and uniqueness), Personal (impact? memorability? kind of an x-factor)

With some further brainstorming, other possibilities emerged:

Cleverness, Playfulness, Perceptiveness, Economy, Profundity, Expansiveness, Quotability, Virality, Connectedness/Entanglement, Adaptability…

The possibilities of evaluation vectors multiply, many bleeding into one another, distinct yet overlapping, hard to define…

More important than "what metrics?", a follow-on question: assuming a perfect, personalized evaluation method, could such numbers ever really capture the true picture of what it means to call a book "good" or "great" or "must-read" or "one of the best" or "part of the canon"?

Is this an impossible exercise? And of course, assuming it is, how can we wrest something useful from that impossibility?

Finally, are there any metrics, processes, philosophies — any approaches to evaluating books — that resonate with you? Whether ones you've adopted personally, seen used effectively, or just think would be cool to experiment with…I'd love to hear em!

Brendan