Being as the Sturgeon's coefficient on internet writing is particularly high, it's valuable to learn to assess pieces as quickly as possible. Doing so is simple, if time-intensive.
Is this worth reading? Within the first paragraph, you will have a constellation of feelings about reading the rest of the piece. These are accurately correlated with whether it is worth reading, but unfortunately untagged. Hence, you must guess what each individual quale means. Then, read the piece. Check the accuracy of your predictions. Repeat as necessary. When you reach a level of accuracy you're happy with, stop reading the ones you don't feel like reading.
I trained myself until actually reading the first paragraph was unnecessary. Glancing in its direction gets me 95%+ accuracy. I can only guess which concrete qualities I'm using for assessment, but it works.
(A fragment of "The Mind; A User's Manual.")
Subscribe to:
Post Comments (Atom)
3 comments:
Would you be willing to submit to a controlled trial of this ability? If it's effect outweighs confirmation bias placebo, you could further train an automatic classifier (no l concrete features required a priori for e.g. a neural net) and rank/filter your graphome entirely outside of conscious effort.
it's effect -> its effect
no I concrete -> no concrete
If you can design a controlled trial, sure. The metric is inherently game-able and not open to negotiation. Namely, did I find this rewarding? There's no incentive to lie to yourself, but lying to others could easily be more rewarding than not having to read some mediocre writing.
Post a Comment