Follow-up Comment #2, patch #4997 (project freeciv):

My apologies for any confusion: the *current* AI is omniscient, so there's no
increase of want for embassy creation based on other conditions (like
relations with the other player, wish to understand the techs / money / map /
units of the other player, etc.).  The current AI also has hardcoded knowledge
that embassies are player-scoped, so doesn't bother considering an embassy if
one exists.

A different AI could certainly not have this knowledge, but there are
performance benefits to understanding that embassies are player-scoped,
whereas most other actions are city- or unit-scoped, so I imagine that other
AIs would also embed this knowledge, either in code, or in datasets used for
initial knowledge definition.

For humans, it's a UI question: establishing an embassy in a city owned by a
player with whom one already has an embassy should be indicated to be useless
in the UI (and possibly disabled): I'm unsure how having this in the ruleset
vs. the code makes a difference.

Now, if embassies were moved to city-scope (so that if one has an embassy, and
the city is conquered, one loses that embassy and gains an embassy with the
conquering player), then the above no longer applies, but unless that is a
real plan today, it would be sensible to deal with a hardcoded restriction
during that effort (as there are so many other hardcoded assumptions one needs
to adjust for such a change).


Reply to this item at:


  Message sent via/by Gna!

Freeciv-dev mailing list

Reply via email to