Copied from Neverending Games?



I feel that there could be a 'trustworthiness' variable. For every shared victory, honored deal, tribute and ten years of peaceful borders, the trustworthiness bar goes up an increment. For every broken deal or alliance, blockade or encroachment of INFLUENCE (not expansion, as all civs will be expanding - right now, you can take relations hits for expanding even if the other players don't know you!), embargo or snide comment at the ambassadors dinner, the trustworthiness bar goes down an increment. It's fairly neutral from 5 / -5, but every 5 past those extremes, you start getting bonuses to relationships. 25 would be 'blood allies' or something, whereas -25 could lead to a Hundred Year War scenario where your dominions would engage in petty struggles even long after both sides were exhausted. This would encourage the AI (hopefully) away from bipolar scenarios where they are your friends one minute, and hate you the next.



- Metric could be affected/influenced by previous strength of deals (taking of several planets lasts a long time in collective memory, is much more heinous than disagreeing to a trade proposal)

- High levels of trust are much more likely to forgive minor breaches of conduct/forgetfulness on the part of the AI (perhaps sending messages if the player parks their fleets 'accidentally' on a world)

- Low levels of trust give the AI bonii to help them keep together if at war with a hated player (some approval, some ship cost reduction) to encourage them to aim their wrath towards enemies.

- Conversely, the AI (and players?) might receive a temporary dissatisfaction level for attacking long-term allies.

- Ceteris parabis, all things trend toward neutrality at which hostility and kindness are equally unlikely.



Please comment and add suggestions. I feel that, with some tuning, this could help a lot of the complaints leveled at the AI at the moment.

Added to the suggestion summary thread /#/endless-space/forum/28-game-design/thread/13003-summary-of-suggestions-so-far-discussion

This thread is the origin of discussions about a trust mechanic for diplomacy AI.

Other threads about this have been archived:

/#/endless-space/forum/29-archives/thread/13525-suggestion-poll-trust



Feel free to post here about this topic here and inform yourself about the variations the threads listed above have already discussed, but don't open new suggestion threads for anything alike.