Nerdolo - new theory team member, self-proclaimed anti idea-girl

Welcome to the team, Nerdolo!

I’ll gladly meet your request to throw topics at you! As Thrive is such a huge and multifaceted project there are always a thousand things which would be great to do/think about, but it’s always a matter of setting priorities and approaching things more or less in the order of their planned implementation.
Combining this first descending list of priorities with your descending list of personal focal points, I’ll try to give you a list of things which might interest you and which are being discussed in some manner atm (hopefully to the satisfaction of the priorities of the team).

Given your background in cognitive science, and your main competences in, among others, simulation, I’m thinking that the auto-eco system might be one thing which could be particularly up your alley. @Thim has recently posted a fascinating video about an auto-evo algorithm he has been developing as a possible alternative to our current one: https://www.youtube.com/watch?v=Ka4btpy9jlk
Main story short, our current auto-eco system is based solely on how well which microbe can obtain which food. Thims prototype is asking the broader question of how well which microbe is adapted to which niche. Atm both algorithms don’t really account for how much a microbe is being predated upon, which should be the other big question asked of them next to “how well can it get which food”.
Now atm Thims prototype (if I understand correctly) has somewhat predefined niches. An important question for the future which Thim and others have discussed on Discord is how a niche is determined and how new niches can emerge from existing niches. I’m sure this discussion could benefit from your perspective!

Another important issue which isn’t really being actively discussed atm but which I reckon needs to be investigated sooner rather than later is how the microbes AI works and how it can utilize new organelles as they are added. If I remember correctly there are some organelles (definitely the binding agent, possibly the chemoreceptor and others?) which the AI doesn’t know how to use at all. Others (like the toxins) are used by the AI but not with an adequate proficiency (see this issue for example: AI shoots toxin too early (while still turning) · Issue #3938 · Revolutionary-Games/Thrive · GitHub). Right now the AI consists of five value sliders like aggression vs docilness, curiosity vs carefulness etc). Is this system enough to account for all the organelles which we have implemented and are planning to implement? If no, what do we do about it without overloading the roadmap (a general AI rework for the microbe stage isn’t planned atm).

You mentioned on Discord that you’re exited about the interactions between microbes and geology, so the planet generation system as well as the dynamic environmental compounds could interest you. While this is also one of my main interests, it’s a bit further down the line I’d say.

If I missed any discussions/plans where a theorist with your main interests/skills is needed which are more urgent/immediate, other devs can surely correct me!

edit: I just realized that Buckly specifically stated that auto-evo and microbe AI are some things he’d like to work on for the next release, so this might actually be an opportune moment to contribute to these topics:)

1 Like