Friday, December 18, 2009

Toboldian Content: A Useless, Poorly-defined Measure.

Tobold has a fatally flawed and useless definition of content. I usually don't carry over comments from his blog to this one [removed an unnecessarily strong phrase, my apologies -Ev], but he’s brought up a topic that I would like to address.

A Poor Definition of Content

Tobold begins by stating his definition of content in a sound-bite form:

I would say that any MMORPG has two fundamental parts: A repetitive part, for which the base rules are always the same, for example combat. And the non-repetitive part which creates the conditions for all those combats, which I call "content".

There are wrong assumptions here. Tobold assumes that all repetitive action in MMOs is combat, which is patently untrue. Crafting may be even more repetitive and deserves a mention. Aside from this semantic contention, t (Oops, I misread "example".)The “base rules” for combat are not always the same. The easiest example are the vehicular combats in WoW: they certainly do not use the same rules as normal combat. Also, playing as a different class engages you with different rules in combat. Tobold could be talking very abstractly here, but I doubt he is—he has never talked so abstractly before. The “base rules” for combat should also include the various special abilities for whatever foe you’re fighting, so those base rules do change between combats with different kinds of monsters (I’ll go into this further and make stronger assertions related to game rules in a future post).

Tobold fails to consider player action-related rules as an element of content. Certainly what a player can do in the world, as dictated wholly by the rules that govern his character’s action, should dictate some part of a measure of content. We can judge the content of games based on game rules alone if we want: which game has more content, the one where you have 2 attacks or the one where you can build houses and dig holes as well as engage in diverse combat?

Tobold’s assumptions are off and incomplete, but perhaps his definition is still salvageable. He claims that the context for repetitive action is the content in an MMO.

Content can be quests, landscapes, dungeons, scripted events like boss fights, monster models, loot tables, lore, and many other things.

This seems reasonable, actually. But then he demonstrates a lack of understanding by attaching no content value to the times where these aspects of content are not hand-made.

What I don't count as content is the number of square miles of procedurally generated landscapes, or the near-infinite number of randomly generated dungeons in games like Diablo or City of Heroes.

What about procedurally generated quests that generate new dungeons that otherwise would not be present? Much content in roguelikes would be disregarded by Tobold, even though it’s entirely illogical to do so. Certainly generating what he’d otherwise call “content” with a machine instead of by hand should count somewhat towards the total content value of a game!

Compare a game with 5 hand-made single-floor dungeons that are always the same to a game with 5 hand-made dungeons that each have 4 more randomly-generated floors. Clearly a game with 25 dungeon floors, even if they have the same art assets and monsters, has more content than a game with only 5.

He then goes on to demonstrate superficial thinking by comparing two games that have just about nothing in common, European football and WoW. This example illustrates the weakness of his definition and its uselessness to describe something worth describing.

A Useless Definition of Content

What do we care about in games?

Do we care about how many gigabytes of textures and sound files are included on the DVD? There is no correlation between the aggregate filesizes of a game’s graphical assets and how much fun it is. And when we’re talking about games, we discuss them based on fun. (Except for the few people who discuss them as art, but they would also agree that sheer size does not dictate artistic merit.)

If we only take quantity of assets as a sign of amount of content, we’re saying nothing useful about the game from a critical perspective.

When players talk about content, they usually are discussing, perhaps indirectly, the amount of time it will take to complete the game. Clearly it doesn’t matter if quests and dungeons are hand-made or procedurally-generated, they add to the time it takes to finish a game in the same fashion—even if you’re only looking to beat a final boss of some type and not looking to collect every collectable and achieve every achievement.

Tobold’s Definition of Content Should be Disregarded

Both as a practical exercise and a critical exercise, Tobold’s definition of content is wildly flawed. His assumptions are incomplete and incorrect, his conclusions are not consistent with his other conclusions and assumptions, and the definition he seems to arrive at does not give us anything useful as game critics or gamers.

We cannot accept a definition so flawed on its own merits and counter to the common understanding of the word defined.


Siqi said...

"Tobold assumes that all repetitive action in MMOs is combat, which is patently untrue. "

He said nothing of the sort - "for example, combat". How are you reading that sentence?

Green Armadillo said...

"When players talk about content, they usually are discussing, perhaps indirectly, the amount of time it will take to complete the game."

The point here is perhaps a bit more subtle than you're suggesting. Let's imagine that you have the rogue right click on the castle door and the non-interactive lockpick progress bar takes an hour of real time to complete. That would add to the amount of time it takes to complete the game but very very few players would argue that the developer has added additional content by increasing the lockpick cast timer to such an extent.

Part of the success of WoW's quest model is its ability to mask the grind. Changing locations after killing the requisite 10 mobs in furtherance of a storyline adds a perception of variety to gameplay that - at least until Wrath added phasing and vehicles - basically consists of run, kill, and loot.

The perception of variety matters as much as the reality. If it did not, your hypothetical developer could save themselves a bunch of time by simply copy+pasting the single floor of their dungeon four times (replacing the boss with a staircase to the next copy on the first four floors) instead of coding and debugging a random content generator.

Eric said...

In my view content is what helps to immerse me in the game. If content were programmed to change over time that would be a huge step forward in MMO gaming and immersion for me. Weather, Seasons, buildings that build and decay, garbage that piles up until someone picks it up, etc. Changing landscape today is very limited. I want more immersion, more persistence and less static art.

The content that WoW presents is brilliant and awe inspiring. Just last month I went into a dungeon to find a map room displaying the land in a globe format. It was amazing and I took the opportunity to observe it closely. It was as if I were in a museum looking at ancient artifacts. Sadly 90% of WoW content is inaccessible to me since I do not raid much. It seems wasted effort that they would spend so much time creating static content that only a few experience. The content that I do experience is static and highly repetitive. Like Arathi Basin. I could draw the entire map from picture memory.

Nils said...

I like to read both your blogs. Tobolds analysis is not complete, but neither is yours or Syncaines.

I know that contentious posts increase the number of comments, but please discuss in a civilized way. I don't want my favorite blogs to start some stupid blog-war about who is 'better'.

In their own way all your blogs are fun to read and sometimes I can even learn something.

Dblade said...

I think you totally missed his point, and how right he is, at least in soundbite form.

It is linked to your earlier post on meaning as well. The content is not the mechanical rules and repetitive actions-it is not the grinding to level 75 itself, but the reasons that drive us that are the true content of MMOs.

I was thinking of EVE reading this. In EVE the repetitive, fixed part is stuff like mining, and pvp. It is complex, but once you get the sense of it, it is fairly fixed.

The content is the reasons why you mine, pvp, and do other things. That part always changes, and is dynamic because it is player generated meaning.

Without it, the other part becomes purposeless. You grind for a purpose, that purpose is content-the grinding itself isn't.

Andrew said...

Tobold's example of something which is not content - i.e. combat - is actually a form of content. There are a few levels to this:

1. PvP is combat between players.... sort of like the soccer example that Tobold brings out. This is, in Tobold's own words, "player-generated content".

2. Changing combat mechanics generates a tonne of new "content" for players who enjoy the metagame. Suddenly they have a whole new set of research to do, experiments to run, and numbers to crunch.

3. This theory, obviously, extends to all game systems and mechanics, not just combat.

Tobold alludes to, and then ignores, the fact that content is a SUBJECTIVE thing. Instead he papers his own preference for content overtop of everyone else's and claims victory in the content debate. So while it is correct for Tobold to claim that WoW has more content that he personally enjoys, it is incorrect for him to assert that it has more content.

evizaer said...

What may be confusing about this article is that I appeal to the players' tacit understanding of content to show that Tobold's definition doesn't account for all content--or even most of the important parts of content. I'm not proposing a definition for content, I'm simply rebuffing Tobold's attempt at defining it.

Nils: I'm not going to start beating up on Tobold and creating a feud from nothing. I doubt I'll mention his blog again for a while. His posts usually aren't relevant to the kinds of articles I write for this blog. The last thing I intend to do with this blog is turn it into a place to whine about personal vendettas and take backhanded shots at games that I don't like.

Dblade: You're just inventing another incomplete definition of content and doing an even worse job of expressing it than Tobold did.

Green Armadillo: I understand that time isn't the primary concern--fun is. But time spent in game does correlate to how much fun you find the game (even if it's only due to social factors keeping you around). Content measurement does depend on variety, yes, but we now have to define what is varied enough to be considered new content.

MLW said...

This is some very venomous writing. Tobold has always made it very clear that he is merely sharing his thoughts and nothing more, and to call him a "charlatan", to claim that he is a deceitful or manipulative writer, is a gross misrepresentation of his writing and a needless insult to the effort and thought that he puts into his writing. He may not be trying to achieve rigor or exploring subjects as deeply as you would like, but comments like yours only make it harder to achieve dialogue.

As for his definition of content, you are making a big stink about a semantic matter. He defines content in a relatively narrow manner, yet still acknowledges the value of mechanics and environment. It is very short-sighted to pretend that he was trying to rigorously define content and offer some sort of fundamental game design principle.

Anonymous said...

Just a passer by here who has never read you blog before.

I agree with the very first comment which was my reaction as well. When you start off a post with an assertion which is obviously factually untrue you lose most of your readers intellectually and all of your credibility. If you can't even get the facts straight then "what you are trying to do" becomes for the reader irrelevant.

You may have great point to make. IDK. I don't care because you lost me on the first sentence.

Jayedub said...

People still read Tobold, who knew...

evizaer said...

"I'm going to refuse to read more than one sentence because that one sentence contains a mostly irrelevant misunderstanding."

With such an attitude, it's probably best that you don't take part in the discussion. One small (as I noted, "semantic") misunderstanding that is quickly passed by has no effect on anything else I say here. You can read it and think for yourself, I'm sure.

evizaer said...


You're right in your first paragraph: I came on way too strong calling tobold a charlatan. I didn't mean it to be as venomous as it sounds in retrospect. I editted that sentence in the interest of not personally attacking Tobold.

With regards to the second paragraph of your comment: Tobold reasoned poorly to produce a useless definition. I don't care if he wrote the post on his deathbed, it's probably a good idea to challenge such sweeping attempts at the redefinition of already tacitly understood words. I've made the mistake of redefining important terms in the past and I went back and noted my errors and took responsibility for my loose use of language. I'm sure Tobold is capable of doing similarly. I doubt he cares enough to actually make an argument on this topic, though. I now have the impression that his post was throw-away in nature and not thoroughly considered. I'm not trying to attack Tobold--I'm trying to argue that his definition of "content" is poor and useless.

scrusi said...

I had a bigger comment on attacking other bloggers with ill supported arguments here but I'll scrap that and get to some actual points of discussion instead.

You rebuke Tobold's point that automatically generated content isn't actually content but I don't see how you can actually believe that. If procedural content was just as good as handcrafted content, Torchlight would be one of the best games ever. Instead, the game gets boring quickly once you realize that you've seen it all before. Every level in the endless dungeon is different, yet you don't get any more content. Instead you get to replay the same old content over and over again.

This doesn't mean that procedurally created content is a bad idea in general, but if you don't mask it between a lot of hand-crafted elements players will soon realize that they are essentially doing the same thing over and over and over again.

I also think Tobold's point about player generated content is a good one, if not fully fletched. A game like soccer (or, if you want to stay somewhat closer to what we are discussing here, Go) relies completely on the interaction between players to make games different. The difference might not actually be the player, however, but the way individual instances of the game differ from each other. Most procedurally created content in games makes no meaningful differences to how the game plays, while "player generated content" often does.

Especially in RPGs, procedural generation of actual content is extremely hard to accomplish. Art (and voice) assets are obvious examples, but story is equally hard. Despite Chris Crawford's best intentions, interactive story telling is still very much in its infancy.

I think it was Chris who I first heard the term "chinese menu quest" from, but I can't find the quote right now so I might be mistaken. Many quests in WoW feel like they come from a chinese menu quest generator, but personally I don't consider those quests content - they are grind. Then there are the quests that are actually interesting and fun to do but those are very obviously hand-crafted. Those quests can be considered content.

Your analogy of time spent = content is quite flawed. (You attribute it to "players" in general but then use it as an argument, so I declare it yours ;))

Actions that I have to perform to get to the parts of the game I enjoy cannot be considered content, or you will have to consider the extra clicks we need to start a game (see Sirlin interview) content as well. Unskippable repeated cutscene? Content! Loading screens? Content! Lag? Content!

I'm by no means saying that Tobold's definition is perfect or even complete, but for a rebuttal like this one - even after the edits - you have remarkably few actual points.

MLW said...

I think you just have very different goals in your writing: you are aiming for a very strict definition of games, while he is grasping for one that's sufficient for a casual discussion and rumination. I think it's useful to highlight the occasional ambiguity of his comments, but I also think it's worth keeping your respective goals in mind.

Anyway, I think the edits have improved the tone greatly :).

evizaer said...

Jor: Great point and one that I do appreciate more after writing my post and seeing the reaction. If he was a little more vague I probably wouldn't have taken issue with the post.

I know about attacking other bloggers and how it is bad. That's why I took out the offending sentence once I realized it went over as flat insult (i honestly don't remember what i was thinking when I initially wrote it). This post is attacking Tobold's point, not him.

"You rebuke Tobold's point that automatically generated content isn't actually content but I don't see how you can actually believe that."

I'm not making a strong claim here, and that's why I don't think it's unreasonable. I'm only saying that procedurally generated content has more than 0 marginal value when added to a game. Procedurally generated content will not be as good as (most) hand-crafted content, but that doesn't mean that it can't stand in at certain points and still count as content. I would consider procedurally generated content (or, more accurately, the procedure that generates content) to be a part of the game's content, though it might be less so than a piece of hand-crafted comment.

Also, the mere fact that you use content in the phrase "procedurally-generated content" indicates that I've got a leg to stand on.

"Your analogy of time spent = content is quite flawed. (You attribute it to "players" in general but then use it as an argument, so I declare it yours ;))"

It's actually what people generally consider to be content. Players and reviewers often say things like "this game has 15 hours of single-player content". So content can clearly be used meaningfully as measured by time spent. I'm by no means saying this is the ONLY measure, just that it's a commonly used one by players.

"I'm by no means saying that Tobold's definition is perfect or even complete, but for a rebuttal like this one - even after the edits - you have remarkably few actual points."

You've barely taken issue with half of the points I've made. I don't see how making four points is "remarkably few actual points." If there was ONE major flaw with his conception, I could have written a post about it justifiably.

Tolthir said...

Raph Koster defined content as "statistical variation in systems," which I always thought was a pretty good definition. The idea is that repetitive activities don't add much "content" in the normal sense of the word. For example, old-school arcade games like Pac Man take a long time to complete, but they don't have very much "content" because you're really doing the same level over and over again.

I think that may have been what Tobold was getting at. Randomly-generated dungeons, for example, only add content to the extent to the extent that there are significant variations among them. Once you've discovered all of the variations, you're just repeating the same activity over and over again.

evizaer said...

The question becomes to what degree does a player memorize gameplay. A randomly-generated dungeon is memorized through the understanding of the pattern used to generate the dungeon. It becomes boring to play through procedurally generated dungeons because you become familiar with the universe of possible dungeons. There may be differences between dungeons, but you have perfectly accurate expectations of what you will (and can) see.

So there is variation when the player is learning the game, but any degree of variation in procedurally-generated content can be picked apart into patterns if the player is given enough time. Unfortunately for game designers, the player will find a way to trivialize any PvE-style content, procedurally- or hand-generated, by exploiting holes in the system or gaining and using near-complete knowledge of the strategies required to succeed. So we cannot safely say that procedurally-generated content is not content, just as we can't say that hand-made content is not content. They both present the player with a learning curve sufficient for them to be considered content. A hand-crafted level should simply be counted as more content than a randomly generated dungeon level.

We also see here how subjective content is. Depending on the player, different kinds of gameplay yield different amounts of content. It's probably best, as I say in the article, to stick with the tacit understanding that most gamers have, because trying to pin such a subjective measure down (as tobold has tried and failed to do) only results in irrelevant or incomplete measures.

Anonymous said...

I am thinking that the basic premiss for discussing "content" in a game seem fairly straight foreward. As it is fairly self-evident that if "something" is in the game, it can be called content.

Similar to a glass containing beer.

To have a bias as to what is and is not to be called "content" seem pointless, insofar as no sensible justification is given for wanting to appropriate so called "content".

Verilazic said...

It seems more like Tobold put forth his own definition in order to aid the main point in his post. No real need to initiate your own post by saying his definition is bad. Say you disagree, that's fine. Saying someone else's beliefs are bad is a fast track to conflict.

Meh, why am I even bothering with this, this is the internet, but I also know you're reasonable about things. There's no real reason to worry.

evizaer said...

A definition can be objectively poor and useless when it does not do a good job of defining anything meaningful and when it cannot be used to reliably understand or predict what it pretends to. That's why I make the strong statement in the title and the headings. I don't usually talk like this and maybe I should've let Tobold's post pass by in silence--but occasionally it's fun to take a stand against something you find ridiculous and take the kid gloves off.

Brian 'Psychochild' Green said...

As mentioned above, one problem here is that content is subjective. For example, kids are fascinated by simple board games like Candyland, but adults who see there's no meaningful choice find it dull; the game is content for kids but not for adults. Or, in the realm of MMOs, I might find PvE content in a game where I enjoy PvP to be empty. I suspect this is the source of the pissing contest between Tobold and SynCaine because they enjoy completely different games and seem to be dismissing what the other would call content.

I think that evizaer is on the right track when he mentions time spent. Tie this into the class definition of a game as "a series of interesting choices" and you start to see how the definition applies. The "one hour picking a castle gate's lock" example that Green Armadillo gives above isn't really content because there's no decision being made. This is one reason why WoW is able to "mask the grind", because they give you what appear to be decisions on how you want to approach the content; although once you've done the content a few times and have optimized it, you realize how little choice you really have....

scrusi wrote:
You rebuke Tobold's point that automatically generated content isn't actually content but I don't see how you can actually believe that.

In the past I would have agreed without hesitation. But, go play the game Splunky. That game is entirely procedurally generated and has a ton of content because of the way things interact. Not to say that all procedurally generated content is great, but it's not all empty fluff that you can figure out the pattern given enough time. The game will continue to surprise you with the interesting events and combinations that happen in the game. Of course, I suspect one reason why the game is so deep is because a lot of those interactions were foreseen and specifically implemented by the game's creator.

My thoughts.

scrusi said...

"You've barely taken issue with half of the points I've made. I don't see how making four points is "remarkably few actual points." If there was ONE major flaw with his conception, I could have written a post about it justifiably."

I don't mind you writing a post about it at all. Tobold's definition is flawed and incomplete and definitely warrants further discussion. What I object to is the "Ignore everything Tobold said, it's wrong" attitude that at least I get from your post.

As for "remarkably few points": I meant remarkably few points that supported your claim that Tobold's definition is worthless. Yes his list of what constitutes content is incomplete, but he actually says that. Some of the things you claim him to be missing are actually there (i.e. "he various special abilities for whatever foe you’re fighting" which is clearly part of "scripted events like boss fights").

As for procedurally-created content (there, I said it again), I agree that it is content - but so does Tobold. He says he counts "for example the number of different tile sets used" which should obviously include the algorithms for that content's creation as well.
The amount of actual "content" that procedurally created content includes, directly depends on the quality and amount of its hand-crafted elements: algorithms, art-assets, and pre-implemented foreseen interactions as mentioned by Brian. (There may be more, please don't disregard my point because I missed something ;))

A pure time-based definition of content is certainly technically valid but of next to no use. Brian is right when he mentions interesting choices, but I would personally count non-game bits as well, as long as they enhance the experience. If it takes me 40 hours to play through Dragon Age, I wouldn't go and artificially reduce that number to 35 because the rest was non-interactive. (Cut-scenes, me listening to an NPC's monologue, etc. Numbers made up.)

Likewise, if parts of the game are not enjoyable (i.e. due to being repetitive), I wouldn't count them as content. This is why Torchlight's content isn't infinite. You are still making the same "meaningful choices" in level 236 of the endless dungeon that you made in level 2 of the normal one. (Maybe even more due to more available abilities.) The algorithms are simply not sophisticated enough, and the art assets and hand-crafted bits not plentiful enough to make the game interesting for that much procedurally-generated content to work.

Longasc said...

"the context for repetitive action is the content in an MMO"

Hehe. Yes, it is part of the content. It can be story or environment or both. But what do we mean when we talk to each other about the "content" of a MMO? We could talk about the number of quests, then we can take in the quality of the quests, and then we could start to argue how much content there is in a sandbox of user generated content and... it gets really complicated. Often content is also taken as the amount of "landmass" and "dungeons" of an expansion, this was usually the use of the word when people were debating whether LOTRO's mini expansion "Siege of Mirkwood" offers enough content or not.

Tobold's definition is flawed for sure, but almost everyone is not exact or precise when we talk about "content".

Is re-skinning half of the bears in a certain LOTRO area new content? Is getting a new horse model with new textures content? Is adding some higher level mobs and balancing mob levels and mob spawns in an area content, can it be called new content?

Language and common use are the problem here. What can be called "content" changes a lot with the context in which the word is used.

It is VERY hard to measure, you can count how much "stuff you can do" in a certain MMO and say it has much or a lack of content. EVE is the premium MMO out there at the moment that I would call a "sandbox"... but how do you measure the content of EVE? There are not much obvious things to do besides mining and running missions, this is not a themepark style MMO.

Don't blame the poor guy for the poor use of a word that is about as dilatable and meaningless as "freedom".

But I agree to your verdict: Talking about content is not talking about hard facts, as it is indeed extremely poorly defined.