mtbc: maze K (white-green)
Mark T. B. Carroll ([personal profile] mtbc) wrote2022-03-13 02:06 pm
Entry tags:

Time travel plots and computer programs

A couple of years ago, I mentioned how getting Primer (2004) straight in my head is worse than Dark (2017). The more I consider such issues, the more I suspect that programming languages like Haskell, which make it easy to encode one's thinking on a scaffold of custom mini-languages, may be a boon to those wishing to make sure that their time-travel plot is consistent.

Today, I stumbled onto yet another diagrammatic attempt to explain what happens in Primer, why what we see makes sense. It looks a good try, even handling revisiting the party, which I think is the toughest aspect. It occurs to me that the way to be sure is to encode,

  1. one's model of time travel

  2. the observed events

  3. the hypothesized events

and see if the computer thinks them consistent. It could even try to generate that last item, the what really happened, rather than trying to verify a fan's. Perhaps languages like Mercury would beat even Haskell for this kind of application.