Content?

Grand Text Auto (GTxtA) is discussing the role of content and its creation in game worlds  (Not/Non Feeling Content: The Unfairness of Programming). 

At the Game Developers Conference no doubt the "problem of content"  will be a rallying  cry again this year, as Andrew points out:

There is a serious problem, I believe, that doesn’t get enough attention: by and large, people are not enabled to create dynamic content for interactive experiences. Many people want to do it, I’ve noticed, but most are unable to. And, the path to allowing this to happen is not an easy one.

To recast it more broadly, and  in our terms:

1. The bottleneck preventing us from realizing more worlds, at least as currently framed in our discussions, is one of content
2. We'd have more content if we had more "programmers" engaged (amateur or profressional)
3. ...Yet, not everyone can or wants to program.

The GTxtA discussion suggests we throw the machine into the breach - e.g. greater use of Artificial Intelligence  (AI, in its many generative flavors, more broadly meant than described here, for example).

Who pwn's the MMO content relationship?   Are worlds, as they are commercially conceptualized these days,  mostly limited by  a "programmer" culture that, say, over-indulges in non-scalable (game) content designs  and  methods.  Or rather, is it that these worlds  have become such  monuments of over-investment (in category terms) that it leads to a dominant  perspective and culture that  is blind to other forms  of worship? 

MMOs blur the lines between the creator and consumer of content.   Mostly because I suspect a dream.  We all want to believe that if we could just build the right cathederal, players will engage themselves so completely.  Perhaps, to such a stage an AI future would be empowering:  more experiments, more worlds, and more opportunities to get slapped around a bit, by a better idea.

Comments

[Dave Rickey]

1.

Very simply the problem comes down to this: Designers who can't code (but think they can), and programmers who can't design (but also think they can).

Code is the fundamental substance of our medium. Code determines whether or not our grand notions actually work, or blow up all over the place. They are for us what pigments and binders are for a painter, what lenses and filters are for a director. Trying to design without knowing code at a visceral level would be like trying to write a book without ever learning grammar. *Maybe* you could pull it off, but could you do it twice?

And "knowing" code is more than just fiddling around with one-man projects of a few hundred lines. You don't know code until you've done it for a living. Until you've work as part of a team of programmers for the better part of a year or more. Until you've struggled with source that looks like an entry in the Obfuscated C contest, asked "Who wrote this crap?", and checked the CVS logs only to find out *you* did.

On the other hand, what it takes to "know" design is much more of a cipher, so it's easier for a producer trying to sort out an argument between a designer and a programmer to take the programmer's word for it. Especially since he usually doesn't know either one.

As far as AI being our best chance for a way out of this scaling trap we've fallen into, I'd agree completely.

--Dave

[Nick Cassidy]

2.

As an avid MMO gamer, I find this discussion to be rather intriguing. The ideology that is a never ending stream of dynamic content seems to be an intangible task. Players that indulge in higher end content aren't genuinely immersed due to any one advance in AI, but generally whole events as they are written. Being that each encounter is tailored specifically for a given event, it seems difficult to form a dynamically driven template that will immerse players at an ever increasing rate. I do agree that improvements in AI is a step in the right direction, but the key seems to be networking programmers and story board designers more efficiently alongside advances in AI.

[mike]

3.

Huh. As a programmer, I find myself drawn naturally to emergent and systematic designs, machine-generated content; the entire culture of programming is all about writing the program once and letting it go. The non-scalable parts are all the art assets, maps, stories, quests, &c. that need to be crafted and fed into the system by human beings. I think Diablo's neat; they find it soulless.

Which is to say: I'm all for making programming easier; it'd make my life better. But I don't think it's the bottleneck--your point 2 is at best a major oversimplification.

Write a better Max or Maya, though...

(dave: i'd argue that *rules* are the fundamental substance of our medium; having a programmer's brain helps, but isn't necessary.)

[Tess]

4.

Ironically, many of us don't have the time or resources to make our worlds extensible by the public. It may seem counterintuitive for this to be the case: Think of all the time and money you'll save in the future, as players make all your content! Consider, however, that you are often faced with a deadline and a list of requirements that do not include this feature. Also consider that tools made for the public need to meet an entirely different standard from tools you make for Joe on the other side of the room. Look how much time and effort Bioware put into their NWN editor! Good tools are hard, and rarely given the budget they deserve.

[Dave Rickey]

5.

(dave: i'd argue that *rules* are the fundamental substance of our medium; having a programmer's brain helps, but isn't necessary.)

Rules expressed throgh code. Yes, if we had these wonderful "Do what I want, not what I say" programming environments, then you could design games based purely on the rules. Painting would be a lot easier if the picture in my head didn't have to be espressed by my fingers (did I mention that I literally cannot draw a straight line without a ruler?).

But even that won't stop you from having to think in code terms, algorithmic efficiency requires that you devise rules that turn into clean, efficient code. You can't do that if you can't think in code.

--Dave

[Neil]

6.

An infinite supply of content! What a wonderful dream! Think of the endless surprises.

I find that surprises are rare in the popular MMOGs. For example, the WoW web page was updated today with some "under development" information. So I already know that I will find a very powerful blue dragon in Azshara once the next patch is released. I know that it requires a raid group to kill. I have seen a screenshot of it. So... I'm almost bored of it already.

What if we didn't get all this information about the encounters being added to the game world? What if we could actually DISCOVER this new content?

It's the difference between "Ah, there's that new blue dragon" and "What the...?!? This wasn't here last week! Guild! Come quick!"

If a game could provide the player (with a high level character?) with a way to "summon" a new monster, possibly provide a new skin for it, customize some of its abilities, and place it in an approved place (not a newbie town or anything silly like that), that could provide some surprises. There could be an in-game cost of creating these encounters so that it doesn't get out of control. Creating content in the game could be a reward in itself. Templates could simplify the implementation and ease-of-use. A series of drop-lists and check boxes, etc. It would be a start, and a very compelling idea for players looking for something to do at the end-game.

But most MMOGs are so tightly controlled right now, it doesn't seem like there's room for any surprises, especially nothing created by players, heaven forbid...

[Nick Cassidy]

7.

I find programming ethic without proper direction (management) to be counter intuitive. One of the biggest problems we're faced with in IT (here on campus) is the ability to hone programming savvy minds retroactively. Yes you can build dynamic content, but does your ability to design tailor to the needs of management or more importantly, consumers? This is not to say that intelligent minds aren't more than capable, it's merely a "better" approach.

As said above, the use of editors, much like what Never Winter Nights uses is a great tool that allows players to build concepts for dynamic content. The problem is that once all editor fundamentals have been exhausted, we're back at square one. Generally the way that MMO companies maintain content is by using game specific editors that also have the capability to work as open ended source. This allows for a wider array of possibilities when constructing events. The trade off is that as events become more elaborate, more time is necessary to create a truly new experience for players.

[Matt Mihaly]

8.

Neil wrote:

What if we didn't get all this information about the encounters being added to the game world? What if we could actually DISCOVER this new content?

Many virtual worlds work this way, including ours. We generally do not announce ahead of time that we're working on a particular new game system or group of new content, and let the players discover them. In fact, what we try to do with new areas is work a little plot in so that they aren't just dropped into the world without any sort of in-character action.

I'm including the plot summary of the introduction of a recent area full of gnoll slavers aimed at lowbies as an example (We try to write up histories of any plots that go on so that participants and observers feel like they've gotten the whole story. It's also nice for participants to be recognized publicly for their role in any event.) It's nothing particularly special in terms of events, but then, it's a lot more interesting than just slapping in new content. I'll add in PC or NPC so you have some idea which are the players and which are the NPCs being controlled by gamemasters assisting the plot.

----------
The 18th of Miraman dawned clear and crisp, a perfect morning to showcase the offerings of spring. Unbeknownst to most of Sapience, the cruel leader of a previously unknown race called the Gnoll had decided that today was the perfect day to lunch upon something other than his normal fare of Mingruk stew. With his muzzle salivating at the prospect, Lirthak (NPC) sent two of his best slavers(NPC) out with the instruction to return with something young, tender, and sweet.

The Gnoll slavers had heard of pixies in Minia, and they headed there first. Vellis (NPC), however, chased them away from the archway to Minia. Thinking it wise not to tangle with the aged, yet strangely confident man, the two turned to one another and plotted anew. "What of
Halflings?" he asked, "Do you think he will enjoy them as much as a pixie?" The other nodded and replied, "Yes, Halflings would do nicely."

The Gnolls quickly headed towards the Valley of Actar. They were not alone on this journey, however, for an Arcanist named Ethel(PC) was quickupon their heels, shadowing them from Minia to Actar. Ethel and two of her guildmates, Andreas(PC) and Aurielle(PC), stalked the slavers through the halfling village. The slavers managed to evade their followers by
splitting up. Finally a young halfling girl was spied, and she was snatched up by one of the slavers. She had but a moment to scream as she saw the toothsome Gnoll leering at her.

The slaver fled the valley, calling for his comrade to follow. Merrygold(NPC), the village elder, heard the scream and ran to investigate,
only chancing to see the backs of the Gnolls as they ran away. She frantically searched the area for her niece Blossom(NPC). It was not until
Andreas(PC) questioned if the Gnoll had taken her that Merrygold(NPC) finally accepted what must have happened.

"Please find her," Merrygold(NPC) begged, "Please find my niece."

They questioned her about her knowledge of the Gnolls, which was very little. Merrygold(NPC) told the Arcanists, now joined by Dreas(PC), that she heard one of the Gnolls mention heading back to the "Hills" and a "fortress" as he fled. Sobbing and light-headed from the shock, Merrygold(NPC) shouted out for the entire realm to hear her plight: "Help, please help!"

While Aurielle(PC) and Andreas(PC) left in search of the "hills", adventurers who had heard Merrygold's call arrived: Josalyn and Defia, followed quickly by Syth, Sethai, Reiye, Xel, Namino, Glithoniel, Dallam, and many others (all PCs). The distraught Halfling elder(NPC) explained what had passed. More search parties began to form to head to the Dakhota, Granite, and Shamtota hills.

The Gnolls were indeed in the Granite Hills carrying their captive towards home. As they stopped to rest, the noises of the first search
party reached them. In a panic they split up, trying to lead them away from the carefully guarded entrance which led to Manara Burrow, their home. As they blundered around the hills, Namino Silverwing(PC) came upon them. Fearful now, a Gnoll slaver yelled to his partner to take the girl and flee to the fortress. Blindly heading to the entrance of Manara Burrow in a panic, he flew headfirst into the tunnel, revealing the entrance for all in the area.

The first to enter into this newly discovered place was Namino(PC), followed by Rambaldi, and Taril(PCs). Shortly after the discovery of Manara Burrow, Taril(PC) found Lirthak, the slavers, and the frightened Halfling girl(NPCs).
Yelling his find, he was swiftly joined by Kereille, Acar, Namino, Larali, Kato, Josalyn, Aurielle, Madrick, Dreas, Eranan, Kelderos,
Syder, Ethel, Andreas, (all PCs) and others. They all piled into Lirthak's den, checking to see if the girl was alright and demanding her release.

Though Lirthak(NPC) was secretly quaking at the sight of so many seasoned adventurers invading his domain, he could not show fear in front of his subordinates. The girl was his property now, he sneered, and would make a nice snack once they left. Several of the adventurers, realising that Lirthak would not yield to their demand to release the girl, resorted to violent means. The dragon Aragon(PC) struck the first blow, assisted by others, and they slew Lirthak. As the Gnoll leader perished, the others dispatched the slavers with relative ease. Alas, amidst the chaos, a pair of innocent slaves were slain, mole-like humanoids called Mingruks(NPCs) who had been subjugated by the Gnolls.

These killings did not escape the attention of the terrified Halfling girl, who began to bawl. When the adventurers enjoined her to finally
leave with them, she fearfully asked, "Will you kill me as you killed those helpless slaves?" and refused to trust them. Josalyn and Arabi(PCs)
told Merrygold(NPC) her niece had been found. Overjoyed by the news, Merrygold was led to Manara Burrow and Blossom(NPC) by the Sentinel Glithoniel(PC). As soon as Blossom laid her teary eyes upon her Aunt she threw herself into her arms, holding on tightly as they were reunited.

In the end, all is well, Blossom and Merrygold have returned to Actar Valley without further trauma. But what does this mean for the Mingruk
and Gnoll race that had lived for years within the Burrow? Only time will tell if good or evil will triumph in Manara.
------------------

So again, nothing major as far as in-game plots go, but it's so much nicer to let players feel like they've had a hand in discovering something new or at least got to participate in a one-time-only opening of a new area.

--matt
P.S. Sorry about the length!

[Jim]

9.

As a long time player and current staff member of a MUD, I can bear witness that player-generated content would have definite problems.

Over the years, we've made additions to our game, and inevitably some have been of less quality than others. In fact, some have been of much less quality. The zones are not developed by players but rather staff, and yet bland, stupid, and unbalanced content crept in slowly but surely.

If members of staff can be guilty of this kind of low-quality content, what about players (in other words, a much larger group of people who are unproven at best)? What kind of quality can we typically expect of them other than low?

This isn't to say that players aren't capable of adding something worthwhile to the content. It just means that such players are the "diamonds in the rough".

If the power to add content were to be given to players, there would need to be severe checks on building options, oversight of a skilled staff member, or both. Otherwise you can only be guaranteed that your game will be inundated with low quality, unbalanced, and sometimes just unenjoyable content.

[Nathan Combs]

10.

Neil>

But most MMOGs are so tightly controlled right now, it doesn't seem like there's room for any surprises, especially nothing created by players, heaven forbid...

A great deal of this is likely 'rolled-up' in the world-y vs game-y tug-and-pull. As should be clear to those who track my posts, my bias is towards the 'world-y'... improv, surprise, and prose can imply a "first-class" fun experience, in its own right. And arguably it can be a scalable source of content - as it pushes back some of the responsibility to the players. However, as Jim implies ("player-generated content would have definite problems"), it only works if you can regulate it, somehow, within the world construct. One regulatory system can be the AI, its another way of looking at the role of AI in these systems.

[Biggles]

11.

How about actually employing those 'diamonds in the rough'? First, you need some powerful content tools which don't require deep technical knowledge. Then, you screen people for the ones able to make excellent content. Those who are successful contribute to the game and recieve a fee depending on the popularity and quality of their work (eg, 5p per player who votes 'excellent', 2p per 'good'). Work which recieves bad reviews is removed. Anyone can apply to be a content maker, but must pass certain criteria.

Only a rough idea.. any thoughts?

[Flatfingers]

12.

I'm a world-y type, too. (And a programmer, and a designer, albeit one with no game design credits.)

So I'm very much in favor of players-as-content-creators. But there are three caveats that need to be considered in any real-world discussion of letting players create in-game content:

1) Most players aren't interested in thinking beyond their personal interests. If they create content, it's solely to make the game more fun for them; if that reduces the fun of others or otherwise unbalances the game, they won't care.

2) Sturgeon's Law applies in spades. Passion to create and discipline to create well are not always found in the same person.

3) Power corrupts. Some players will abuse any creative ability given to them. If you run a small or experimental VW, this may not be a problem for you, but the owners of large commercial MMOGs can't afford to offer a tool that makes it easy for some players to create offensive material they can impose on other players.

What these three points tell us is that the number of participants in any virtual world who are capable of adding desirable content to that world will usually be very low. Handing out tools to everyone will merely insure that your world becomes filled with self-interested and rude junk.

Conclusion: developing general-purpose content creation tools for players of virtual worlds (in particular, mainstream MMOGs) is probably not worth the time and money.

That's the negative side. But there's also a positive way to look at these observations, which is to realize that if you could come up with a content-creation tool that rewards the capable while discouraging the opportunists and griefers, that could be worth doing. If the cost to develop and manage the tool is less than the value of the content your players create, then you win.

If this is possible, then it seems to me that the approach most likely to succeed (again, focusing on mass-market commercial products) is to make player content creation part of an overall framework of content creation. In other words, only allow player content creation within the overall developer content creation process. The developer defines and communicates their high-level goals to players (addressing issue #1), who develop low-level features that are selected or rejected based on how well they achieve those goals (addressing issues #2 and #3).

Offering developer guidance ("here's the story you'll be helping to tell this week: ...") isn't likely to be a problem. The controversial bit will be weeding out the self-aggrandizing and the offensive. There are two ways to go: let your players decide what's acceptable, or review and approve player content yourself.

The former approach relies either on players reporting undesirable content to CSRs (who then have to check it out and decide if it's acceptable), or on a self-controlling economic approach (make content creation have an in-game cost so that only the players who make content that lots of other players like can afford to keep doing it). My question: Does anyone think that letting players police each other can work in a mainstream MMOG?

The latter approach -- developer review -- means more work for some Official Persons, who will have to spend time assessing player-created content. Such persons will need to be able to discern the good from the bad, and be willing and able to communicate their rejection of undesirable content to players without upsetting them. This is probably the more effective approach, but it does carry an operational cost, and I suspect that people who could do all these things well would be very rare indeed.

...

The upside of enabling player content creation is that you get some of your creative work done for you, and you generate valuable goodwill among your more motivated players. The downside is that you either open yourself up to charges of censorship and favoritism (if you review content), or you get content that (to put it politely) doesn't support your development goals (if you rely on player review).

As much as I'd like to play a game that allowed me to create interesting content, when I think about it realistically I can understand why there are so few games that offer such a feature.

--Flatfingers

[Nathan Combs]

13.

Flatfingers>

Conclusion: developing general-purpose content creation tools for players of virtual worlds (in particular, mainstream MMOGs) is probably not worth the time and money.

I also think, relevant to a number of discussions here on TN a while back - it would undermine the coherence of the design of the world. There seems to be two exceptions:

1.) smaller "private" micro-worlds within a larger world - the 2L paradigm: coherence is managed by separation.

2.) worlds are "smart" enough and tools crafted restrictively enough that players can only create things consist with the underlying purpose and design... this gets to some of the GTxtA AI and "generative" tools suggestions.

To get to (2.), it seems you would have to get darn close to a (3.): why not just push this responsibility into the AI, parameterized as an "independent" agency, in the world?

[magicback]

14.

The structure and the stylistic aim of the world can be modeled after open software development.

I recently bought a DVD copy of Dark City. It's been a while since my last viewing, but I like the idea where the shape of the world each day as the inhabitants know it was changed by the collective minds of many.

If bad code or structures were added, the world can always revert back to the last safe copy. It becomes a collaborative world like SL.

[Jim]

15.

Flatfingers wrote:

>My question: Does anyone think that letting players police each other can work in a mainstream MMOG?>

It can and does work. I think that Achaea has shown this to be true. There will always be people who can't use responsibility properly, but there will also be those that can. Players are, I think, singularly good at telling the difference (as long as they don't have anything personal at stake).

As for how to find the players that can do what's needed, Biggles wrote:

>How about actually employing those 'diamonds in the rough'? First, you need some powerful content tools which don't require deep technical knowledge. Then, you screen people for the ones able to make excellent content.>

...and he's right. MUDs have done a good job of providing tools with a low learning tools, and the result is that people who have good creative ability now also have the technical ability (through the tools).

Having staff members watch and choose mature players is the best method, in my opinion. This is the most consistent way to weed out the hundreds or more of obviously unqualified people.

Flatfingers wrote:

>The downside is that you either open yourself up to charges of censorship and favoritism (if you review content), or you get content that (to put it politely) doesn't support your development goals (if you rely on player review).>

The worse of which is to have an enduring piece of content which lowers the quality of the whole game just by existing.

Nathan Combs wrote:

>To get to (2.), it seems you would have to get darn close to a (3.): why not just push this responsibility into the AI, parameterized as an "independent" agency, in the world?>

Because you don't get the added benefit of having it known that players get to be a part of developing the world. Whether or not that's a big point, it is an additional draw to your game. Plus, humans can usually create more depth plot-wise than an AI.

I don't think the biggest problem about using player content would be physical consistency. You could always just limit what kinds of objects can be created in that part of the world. On the other hand there's no way to limit dialogue, quests, etc. while keeping it open enough for creative work. Regardless of how you turn it, player content must be reviewed.

[Jim S.]

16.

I wrote:

>It can and does work. I think that Achaea has shown this to be true. There will always be people who can't use responsibility properly, but there will also be those that can. Players are, I think, singularly good at telling the difference (as long as they don't have anything personal at stake).>

Actually what I meant is that players are good at telling the difference if the inapropriate actions of others aren't benefiting them, the players (ex. players would tend to support equipment unbalanced in their favor).