FW: [MUD-Dev] Interesting EQ rant (very long quote)

[holding99 at mindspring.com]

First, please excuse me if this is somewhat haphazard; I tend to do
most of my thinking on such things as I am writing with them.

At 06:44 PM 3/11/01 -0500, David L. Smith wrote:

> Separating the Player from the Character would in effect make all
> characters in the world identical. It removes (mostly) the
> distinction between PC and NPC. Ultimately, the PC representation
> would most likely be some sort of NPC AI that takes "hints" from the
> player as to desired actions, etc.

This was one of the reasons I was wondering if anyone had implemented
such a system. It seemed to me that it would solve some of the
problems discussed about player vs character knowledge. I am not sure,
however, that it would make all characters identical; each could still
have different skills, and therefore have different types of
information available.

> This would let you (among other things).

>> Easily have *persistent* characters (characters who do things while
>> thier players do not) Enforce character traits as they affect
>> character action (alignment, quirks, etc) Assume more correctly
>> that the characters "fit in" to the world they're a part of Treat
>> PC and NPC in similar manners (anyone want a PC that can be easily
>> scripted?

I may have missed a few emails, but could you tell me where this quote
is from?

> Overall a really d*mn cool idea, and one that I'd love to hear more
> thought about. Basically I think that the more closely you can
> combine the concepts of PC and NPC in a system, the easier it is to
> program effects that treat them both logically. But the AI system
> that interprets player suggestions and creates character actions
> would have to be fairly advanced. In addition it should probably be
> able to adapt the "personality" of the character to the play style
> of the player over the long run. For instance, the good paladin
> character shouldn't possibly attack the old woman the first time the
> player tells it to, but if the player spends some time making the
> paladin do increasingly evil things, there might come a day when the
> paladin's AI relishes the suggestion to attack the old woman.

Actually, I wasn't thinking anything so sophisticated, although that
would be a long term goal. Instead, I was considering a simple system
that would enforce basic life effects, which would be easiest to
code. Most of the player directions would be followed. However, giving
the character some autonomy could reduce some of the more tedious
aspects of games (such as eating and drinking), while still retaining
their importance. For example, your character has food, he's hungry,
therefore he eats, _without any input from the player_. However, once
he's run out of food, his behavior (and cooperation) would change.
Perhaps starting with thinking-type "I'm hungry" messages ("Hey, that
food sure looks good! Wish I could have some." "Oh, my stomach is sooo
empty!") moving forward maybe even to the character spontaneously
taking actions to get food (such as attacking a deer/rabbit/game
animal, or randomly picking berries off of a nearby bush and eating
them). If the player could, as you said above, script a little bit of
AI into his character (IE, if you get low on hit points, RUN!), then
things that could adversely affect the player (such as connection
speed, dropped connection, etc) would not adversely affect the
character. I am not sure such a system would be feasible, but it would
be interesting to see in use. I would especially like to see a system
that modelled character personality as described above. Then a player
truly could play a character with moral issues (such as a paladin
falling from grace), and would have to edure all of the consequences
therein.

T.H. Cooke

_______________________________________________
MUD-Dev mailing list
MUD-Dev at kanga.nu
https://www.kanga.nu/lists/listinfo/mud-dev