Background: ecosystems management practices for improving community memespaces
One can model individual human minds, as well as a community of minds, as an “ecosystem” of “memes”. These memes might be things like:
- Bayesian epistemology
- a habit of checking Facebook when one wakes up
- wiggling one’s fingers to indicate agreement with a statement
- prefacing things one says with “My model of this is that...”
- Doing calibration training
- Referring to blog posts in conversation
- Taking silent pauses to think mid-conversation
Calling this set an “ecosystem”, seems to me to be mechanistically very close to what’s actually going on. At least, this is because:
- Memes mutate as they are transmitted between minds
- Memes undergo selection pressure as they are transmitted
- The underlying topology/geography of social, cultural and geographical networks of people influence their spread
- Memes can be in equilibrium with other memes
- Memes can act as “invasive species”
Now there is an emerging European rationality community, largely driven by efforts from the Prague rationalists. This community imports many memes from the Bay area rationality community. For a high-level, historical examination of this memespace, see Julia Galef’s map of bay area memespace.
At a recent CFAR workshop, we discussed how we can ensure that this interaction is successful.
Five of us (Ales Flidr, Elizabeth Garrett, Nora Ammann, Adam Scholl, Jacob Lagerros), felt that the ecosystem model carried sufficient mechanistic similarity to the actual situation that it would be helpful to read up on things like: protocols for deliberate introductions of new species into new environments, invasive species regulations and protection programs, pest control, and more.
Collection of background notes
We spent 1h researching this. Now the outside view predicts that if we were to leave it at that, the 16-page Google doc would never be used again. Hence we’re experimenting with releasing our notes together with a LessWrong question, in order to allow others to benefit from and build on our progress.
You can find our notes here.
These notes are provided “as is”. I (jacobjacob) briefly went through them to make them more readable, but apart from that this should not be interpreted as something the authors endorse as being true, and despite originating at a CFAR workshop it is not official CFAR content.
We’d be interested in using further research to answer questions such as:
- What are warning signs of a memespace/ecosystem being harmed?
- What are best practices for introducing a new meme into a memespace, and what can we learn from actual ecosystems?
- What are some useful models for thinking about this problem?
Ecosystems do not have a goal
Ecosystems are not optimised for diversity, they produce it incidentally
Ecosystems do not cross-breed distant members
Ecosystems have no one overlooking the transmissions being made and deciding whether they're good or not. Memeplexes have all humans all doing that all of the time
I do share an intuition that there are relevant insights to be found by studying ecosystems, but I think you'd have to go really deep to get it and extract it.