[freeciv-ai] Re: (PR#9808) AI should remove shared vision more often

["Per I. Mathisen" <per@xxxxxxxxxxx>]

<URL: http://rt.freeciv.org/Ticket/Display.html?id=9808 >

On Wed, 25 Aug 2004, Benoit Hudson wrote:
> On Wed, Aug 25, 2004 at 07:49:08AM -0700, Mateusz Stefek wrote:
> > AI doesn't share vision with you if it's not safe, but once shared
> > vision is accepted AI will never remove it even if it became dangerous.
> > For example if AI's ally give shared vision to someone, who is neutral
> > to AI. This may result in vision shared beetwen enemies (and is in fact
> > very often in auto-games).

Yes. Thanks for making a patch.

> A related problem is that the AI often gives shared vision without
> demanding quid pro quo; then asks for it repeatedly, without
> consequences to refuse it.

It doesn't really need it. I agree it should be reciprocal. Patch would
appreciated.

> Also, human players don't automatically withdraw vision from enemies,
> which is odd. There should at the very least be a popup asking to
> withdraw vision, if it's not done automatically (I could see giving
> vision to an enemy in some strange circumstances).

A dialog on war declaration asking if you also want to withdraw vision
would be ok.

> Finally, a more general problem: the AI very rarely drops alliances or
> sues (or even accepts, at any price) for peace. An alliance lasts
> pretty much forever; a war does too. Macchiavelli would shake his head
> at this.

This is a bigger issue, and I agree with you on this. The AI should be
more evil, and seek victory alone, not an allied victory, if it deems this
possible.

Personalities for AI, especially nicer personalities for easy AI, would be
nice.

Feel free to code it. If not, I will do it eventually, but not likely
before xmas.

  - Per