My brain’s in a knot. Any way I push or pull it draws the knot tighter, leaving me ever less capable of unravelling. Here’s how it happened:
Captain Jameson, in its current form, is fun. It’s fun to explore the galaxy, hopping from station to station and nav point to nav point, picking a path between laser beams and oblivion. Unfortunately, this exploration doesn’t serve any purpose. I usually invent my own goal by setting off in an arbitrary direction, but for most players that isn’t enough. People need a reason to journey into the black. Initially I thought the problem trivial, and I set out to build a quest system. I figured it’d take a few hours. All I’d need to do is create a station visible from a long way off and tell the player to go there. I haven’t done it yet.
The reason I haven’t built this system is that I’m not sure I should. I want Jameson to have procedurally generated quests, so that each time I start a new world I have new things to discover. Unfortunately every generated quest I’ve ever seen in a game they have been utterly, utterly dull.
“Hi! My name is charactername. I need your help! Please deliver itemname from placename and I will reward you with lootname. Thanks!”
No thanks. This led me to think about generated narrative content, about which there are many articles suggesting it should be possible, and very few examples of it working in any way. Most of the articles focussed on the Hero’s Journey, otherwise referred to as the Monomyth. Here’s an overview:
“A hero ventures forth from the world of common day into a region of supernatural wonder: fabulous forces are there encountered and a decisive victory is won: the hero comes back from this mysterious adventure with the power to bestow boons on his fellow man.” – Joseph Campbell, The Hero with a Thousand Faces
The idea is that most myths follow this template, with other elements (such as supernatural guides, vision quests, death and rebirth) often contributing to the tale. Working from this formula it seems trivial to automate the process of building a good narrative, but despite all the talk nobody seems to have managed it. Thinking more, I’m not even sure I’d want to do this. After all, this formula is tailored to non-interactive storytelling. I’m sure there’s a lot I could learn from it, but I don’t think it’s a good idea to just dump it into a game and assume it’ll work. No, I think I need to understand how and why it works, then consider how that effects the design of interactive systems.
Now all I need to do is learn narrative design, discover some incredible way to make it procedural and interactive, and code it. People have been trying this for years without success, so my expectations are extremely low. Still, I have a giddy feeling about it so I won’t give up just yet.
<3; Farbs
This is what i was saying. The quests could kill Captain Jameson, or make the game. Whatever you do, be careful not to ruin the game, we are relying on you. IMO, the game should still work without quests.
for the love all that is fun dont put baby sitting escort missions in. that kills the flow of a gmae for me like nothing else. except maybe for metroid like revisiting of explored areas.
having procedurally based quest would be great. i usually dont go back to games after i beat them because i remember all the quest and if there were a few i hated i wouldnt play the game again. do you think you can sneak in some achievements? like kill x enemies in x time? or build a ship with x girders? or is all of that alittle to mainstream?
Achievements? My question is, how would achievements find into the story of the game? there has to be a reason for them.
One way to keep the game open and yet allow for quest experiences would be to introduce the idea of a large war.
You could be begin by randomly generating two or more factions.
Then introduce character ships with various roles to either side: “Fellow Hero”, “Initiator”, “Jester” etc; multiple of each but pretty random.
Set the character ships to attract themselves to points in the war where major events are taking place – perhaps you could set the mechanic up in a way that meant that at particular focal points player or character ships would be required to do very important things – destroy a station using only an item available at x random location, provide a particular type of crystal etc. etc. – allow the dialogue to flow from the character of the “character ships” and a list of “story” requirements – close a warp gate to prevent resupply from the enemy, establish a minefield, seel an enemy station loads of narcotics to reduce its efficiency etc.
If coded cleverly there would be no need to create set stories, but the player would find themselves naturally drawn into quest-like behaviour. In fact, choosing the right thing to do would be made part of the game. It would be similar to IF. Perhaps all else has failed and the station just cannot be destroyed and it is holding a particular sector for the enemy. The decision and creative leap to decide to try to flood the station with narcotics would be very rewarding for the player, and the massive size of the Jameson universe would mean that any one unsolvable puzzle wouldn’t ruin the game for the player.
Perhaps all of that’s a little too far out, and I’ve not done very well describing it, but it could be fun. Making the player a protaganist in a vast cosmic drama would be, well, pretty cosmic.
Ever tried Wurmonline?
It’s another game – the beauty is that THERE ARE NO QUESTS. A player starts out in the middle of a town and ventures forth. He may decide to build a house – or be a nomad. The game allows all this, and more. The saving grace is the ability to multiplayer – the ability to trade, to interact.
I wonder if Jameson would have this ability, but if not, Ben’s idea is pretty beautiful.
There’s quite a lot of interesting stuff on the Procedural Content Generation Wiki (http://pcg.wikidot.com/), but nothing I can find that’s specific to quests. Still, there might be something.
Ben’s idea sounds pretty ideal, to make the quests relevant to the game world, so it doesn’t feel like a quest in the ‘Yellow exclamation mark I have a quest for you’ way.
I always kind of thought that the Hero’s Journey focused on the wrong parts of the story. I mean, yes many great stories follow it, but that isn’t WHY they were great, you know? Having characters you care about, exciting situations, interesting twists, god dialogue, etc. seems to matter more.
Just continuing dumping random thoughts as I wake up this morning: I think one problem with quests is that they are giving the player assignments rather than a situation to react to. If instead the player was told about interesting situations that were already going on and could use their attention, I think that might serve the purpose of giving them direction without making me feel like I’m doing someone’s chores the way MMOs do.
Don’t most MMOs have some kind of large war as setting? I think it would be more interesting to see an MMO that went with something a little more peaceful, at least on the surface. You can still have polarizing elements, such as politics and religion, without open conflict. I think that would force a more nuanced approach to the story.
Speaking of cults, will those weird symmetry cultists be in this game too? I love those guys.
Farbsy I had this exact same issue with my game Venture the Void (www.venturethevoid.com) I think it was pretty cool in terms of exploration and procedural generation but it fell really flat with respect to quests. I basically used a template system like you note above. Some friends have said that it’s an awesome game– but with no soul.
Here’s what I would recommend:
- Quests need a personal touch, they really do.
- You’re going to have some kind of template “goal” system in place to define goals (NOT quests, per se) and being able to chain them together; something very simple.
- You’ll also write some kind of object “query” system; i.e., something to describe different types of objects, space stations, etc.
- You’re going to write out quests in this goal language/query system. This is where the personal touch comes into play, and it’s what will give the game soul. You’re going to be able to specify goals, motivations, characters, etc.; the query system will draw together objects from your procedural world to fit your quest and the goal system is going to provide all the logic.
- Next, you’ll set it up so that even on repeated plays, the player doesn’t see the same quest twice, or at least has decreased probability the more often they have seen a quest. I.e., this is something that persists between games.
- Finally, you’ll have a simple website that people can create their own quests on for the Captain Jameson universe.
Well, that’s one idea! Whatever you do, good luck and charge bravely onward!
Sounds fantastic! Can’t wait to see it in action.
@ryan: Escort missions would be fine as randomly offered, optional quests, allowing the player to pick what jobs they want to freelance.
ben’s idea sounds awesome. and it does sound like it could be set up to be randomly generated and still be really interesting.
Well, you could also have “undefined” quests, or jobs. An example of this is the “scrap hauling”. Nothing could tell you that you can make a profit for scrap hauling, you just discover the option to sell modules at stations, and thus, discover and “undefined” way to make money. Just don’t have info buoys come up to you and say “[Name] has a new quest for you, [questname]!”
Kiaora amigos.
I think i mentioned something similar in a previous post, but as a non quest based incentive to get players searching the dark and hopefully seedy nether-regions of the Jamesonverse, you could simply put random rewards (I dont know, maybe an new class of building materials left from a precursor race or random ship heart attribute bonuses) ..umm. oh yes, put those out in the depths of space. The futher the distance from 0, 0 – the greater the potential reward, but the greater the threat of becoming laser seperated space junk.
Quests sounds awesome too.
-B
Just some thoughts…
You can make factions be specialists – one makes the best shields, the other makes the best weapons or the fastest engines. And they need raw materials, and build and sell their specialty based on what they have. The factions can also build their own scavengers, defenses, etc. They either “like” or “dislike” each other and the player, and their prices depend on it.
You can now have quests like “kill faction X’s scavenger”, “provide me with quality X raw material”, “provide me shield/weapon/engine with quality at least X – to build scavenger/defense/etc.” that really affect the universe and the available components.
Getting the factions to a random level at the start of the game can also make each play more unique.
You can also throw in some more stuff like crew members from the factions that the player can get by being nice to them (give them a face) so that not everything can be obtained by force.
Me doing quests is a stupid notion. I’m more awesome than any NPC. Make them do the quests. If I have the rewards to give, they will do my bidding.
I’ve been poking around the concept of procedural narrative off and on for a while now, and the Dwarf Fortress adventure mode method really intrigued me. It’s basically a completely bottom up system that simulates conflicts and other systems, and needs bubble up from those interactions.
For example, a goblin camp is close to a human village. As such, the goblins end up raiding the human village one day and kidnap a villager’s child. When the player discovers this village, that character need has been established and if the player talks to him, he’ll express that need and potentially offer a reward based on his wealth or belongings or whatever.
The end result is similar to what you might find in a scripted game, but the motivations ring a little more true, and the nuanced variances will definitely make each one of these play differently than the last.
Farbs
Wow- it’s great to see more devs interested in applied storytelling theory – You’ll be at Freeplay right? I’m running a workshop on “Story Driven Game Design” & I’d be honoured to reserve a spot for you.
A lot of my research into story & experience design involved the work of Joseph Campbell (the guy who identified the Hero’s Journey & coined the term “Monomyth”) so it might prove helpful. Infact, the problem of procedural quests you mention here is really poignant & it’s something I’ll try to incorporate into the workshop.
cheers!
@Dan: Yep I’ll be there. Sounds fun! I’m not sure what day/time my panel will be, but if scheduling allows it I’d be very keen to attend your workshop. Actually I’ve never been to a workshop – how do they work?
Given the timeframe it’s a very rapid ‘mini’ workshop, but essentially it’s a interactive learning session where, rather than talk at people for 50+ mins, I lead some activities that firstly involve retelling a familiar story & playing a game together, then reflecting on & analysing the experiences on varying psychological levels & then to finish up we apply what we’ve learned in a quick design + pitch exercise.
It’s just a primer – nothing too in-depth, but it aims to demonstrate a useful tool – an interpretive paradigm for how story & gameplay correlate.