#concept #todo: review

purpose of a bit flip (literature-level) ? to ensure novelty of your contribution over the broadest coverage of papers possible

Think of Godel! His work completely affected a vast amount of work in the decade leading up to his finding, including from great mathematicians like Russell. Think of Wittgenstein! His work demolished a fuckton of philosophers who were waxing and waning about semantics, thereby creating the new field of analytic philosophy. Think of Darwin! He flipped religion on its head

rather than a paper-level bit flip, you want to go for a ____ bit flip because ____ ? literature-level, it’s making the biggest impact bit flip >> an inversion of an assumption that humans have about how the world is supposed to work

Recipe for a bit flip:

  1. Establish bit: Articulate an assumption, often left implicit in prior work: this is the bit
  2. Flip it: β€œNo, it should be this way instead:” argue for an alternative to that assumption

On the value of literature search: β€œan hour in the library…” ? β€œβ€¦saves you a year at the keyboard” After __ papers or so, it kinda starts asymptoting >> 5 CS PhDs usually look at like __-___ papers for their research >> 25-35### References 1.

Notes

They’re much more often pivoted off of today’s work:

  • Some constraint that exists but shouldn’t, or visa versa
  • A realization that an idea has been applied in domains like X and needs to be rethought in domains like ~X
  • A recognition that others have tried this technique in users of context A, or data of up to size N, but ~A or >>N breaks the technique.

Why this metaphor is good

Because it is clear that there are several design axes, and this focuses on flipping just one. Because more than one will be harder to track. Keep it simple silly!

Nearest neighbor paper

Nearest neighbor paper is a paper that is adjacent to your idea.

What assumption or limitation did this paper have, that you’re erasing?

Expanding to more papers from nearest neighbor paper

  • backward influence - citations in the paper
    • if it’s mentioned lots of time
  • forward influence - papers that cite the paper

Literature-level bit flip

Recipe

  1. Read the literature
  2. What assumptions underlie all of the papers? for all p with element papers:
  3. Which assumption are you changing? And why does it matter to the literature?

This is even better because more important ideas can be identified by bit flipping across a broader literature.

For example, if your nearest neighbor paper assumed X and you want to do ~X, but other papers in the domain already assumed ~X with slightly different setups, it’s a more minor contribution.

I think bit flip process applies to:

  • academic research
  • organizations
  • art
  • ideas / stories that we tell ourselves