Experiment the second (and a considerably longer play this time than Doofus vs. Delores). To introduce the cast:
Harry. A Good, Frugal, Neat Bookworm whose life goal was to become a secret agent.
Lisa. A snobbish woman with no sense of humor whose life goal was to have a net household worth of more than 100,000 Simoleans.
Lars. An evil, mean-spirited character with a good sense of humor whose life goal was to become emperor of evil.
These were all siblings, and I figured that the good/evil dichotomy between Lars and Harry, and the funny/unfunny between Lars and Lisa, would produce plenty of conflict (always the basis for a good plot, I figured). And indeed they did fight a lot. Lars enjoyed picking on Lisa, but also found her so boring that they didn’t interact for long. Lars and Harry enjoyed one another’s company at first, but soon Lars’ evil nature made it impossible for them to get along. Rather pathetically, for a while Lars had a goal of becoming friends again with Harry, but I eventually cancelled this when it became clear that it was simply impossible.
Because they were all created at around the same time, they all died at around the same time too. Lars went first, then Lisa. Unfortunately, Lisa happened to die while Harry was cooking. The pan was forgotten on the stove while Harry watched Lisa being Reaped. The inevitable fire ensued. The Reaper noticed the fire but said he’d be back later because he didn’t feel like doing any more reaping on an empty stomach. And come back he did: the fire got extinguished, but Harry died anyway later the same evening. He was the only one not to get his life’s wish fulfilled, incidentally — possibly because he’d spent so much of his neat-freak life making beds and cleaning toilets.
On a low level, there were a large number of entertaining mini-interactions: Lisa vamping in the mirror only to have Lars sneak up and make faces behind her; Lars mopping up a puddle and then cackling evilly, I think because he knew that Harry had really wanted to mop that puddle himself. And there was an entertainingly large number of evil actions available for Lars to try, from undermining charities to trolling internet forums. (Lars was a success: he did get all the way to Emperor of Evil status, complete with a sinister red aura, before the Grim Reaper came to collect him.)
Things I’m liking so far: I can’t fault the thing for diversity of content. There is clearly a huge amount of the game that I haven’t come close to seeing yet. I tend to play Sims with a focus on their careers, so they don’t go out and see other people much, and there are large parts of the town that I’ve never used at all.
Something I particularly like about the social interactions is the way that many activities are keyed to character traits — so the player can select certain actions because the character is flirty or good or mean-spirited or whatever. This contributes a lot to the sense of individual characterization, and communicates to the game player a lot about how the traits are affecting the game.
On the other hand: Despite some attempts on my part to push the Sims into big conflicts, there wasn’t much by way of arc. Lars and Harry both advanced in their respective careers. Lars went to prison a few times, but it didn’t seem to generate any particular complications that Harry was at the same time rising through the ranks of the police. When Lars died, Lisa inherited his partnership in the crime syndicate, but she apparently had no trouble taking over. And very often the Sims avoided brewing trouble by avoiding one another or managing not to be in the house at the same time.
Even in the small interactions, it isn’t always clear to me, when the Sims interact autonomously, exactly what they’re doing or saying to one another. Often it’s hard to distinguish on the basis of tone and gesture which of the many social options they’re using on each other, which means that they can get into fights without my knowing what they’re fighting about. I suppose that’s realistic in its way, but I find it a little unsatisfying, and sometimes I find myself wishing for an IF-style transcript or scrollback where I can see: “Lars implies Lisa’s mother was a llama. Lisa slaps Lars.” …or whatever.
I wonder how much of the arcless feel comes from the relative lack of major changes in the game, especially negative changes. I mean, characters do die, but generally of old age. (I gather it’s possible to starve them as well, but they’re self-sufficient enough to prevent if you give them a vaguely normal set of resources.) But, for instance, though Lars and Lisa got into multiple fist fights when they were young, neither of them ever moved out of the house, or did the other serious long-term harm, or took any action that fundamentally altered the game state.
I remain interested in the idea of a structure in which the player sets up a conflict using the built-in tools and traits of the system and then the game determines what the characters do in response, though. (Or partly determines. The player always has the opportunity to meddle.)
If it’s one thing the Sims is known for (even in the very first iteration), it’s a multiplicity of ways to die.
Based on some of the events of the Alice and Kev narrative, it’s certainly possible to have negative outcomes and have one character to try desperately to stay away from home, even to the extent of being arrested for violating curfew.
However, just like real life, just because people are in conflict, doesn’t mean there will be some interesting narrative as a result; sometimes it is just malaise as two people who hate each other just trudge on.
Which suggests that attempts (in IF an otherwise) to simply reproduce realism and let the chips fall may be misguided, and that some kind of narrative forcing may be required.
Sandboxes can be inherently interesting without any kind of external shaping of the narrative, it’s only their current lack of sophistication which makes this fact non-obvious.
To take an extreme example, imagine a computer program which was able to simulate an entire world, with all of the physical systems behaving correctly, and all of its inhabitants behaving in a believably human manner. The “player” is then able to make any changes she wishes to this world at any time. How could this possibly fail to be an utterly compelling toy with which to play?
Obviously such a system is far beyond current capabilities, but the difference is only a matter of degree.
But would it be ethical to toy with such accurate simulations of thinking beings? Frankly, I think the situation you describe here’s going to cross beyond the line of true artificial intelligence.
How could this possibly fail to be an utterly compelling toy with which to play?
This would only be compelling if you are looking for a power trip. Generally, people play games because these give them challenge and meaning. Your “utterly compelling toy” gives us neither.
Viktor: I think people generally play games for fun, rather than challenge or meaning. Challenge is usually a good delivery mechanism for fun, whilst meaning is often entirely lacking. I’m not suggesting that this is an ideal state of affairs by any means, but your statement that players are searching for meaning in their games strikes me as an odd generalisation.
I am intrigued by your lack of empathy with my point of view here, and I would genuinely love to comprehend why we view this so differently.
Take my extreme example above, where the player has ultimate power. Reduce the power of the player a little at a time, as if this variable were controlled by a slider in the game design interface. At what point on the slider does the experience cease to become merely a “power trip”? At what point does it become a worthwhile challenge?
If the player were cast as a mighty emperor (Emperor of China in the Song Dynasty, say) is that still of no interest? If the player is a homeless man in London trying simply to survive from day to day, have we now provided challenge? How about meaning?
Can a worthwhile and meaningful gaming experience only come about via limiting the player to a predefined narrative path set out by an author? If we now place this “sandbox / authorial control” scale on a slider too, at what point on the slider does “meaningful” kick in?