[game_preservation] Descriptive terms for Video Games

[Jan Baart]

Very interesting replies everyone. My excuses in advance though, this 
turned out to be a much longer mail than I had hoped!*

Richard Pugh

*> Absolutely!  Pipe them along!

Will do so in a private mail, but in case others are interested too, 
here are our "main" genres, which contain further granular subgenres:

Action-Adventure
Adventure
Arcade
Compilation
Construction and Management Simulation
Fighting
Life Simulation
Music
Other
Personal Development
Platformer
Puzzle
Racing
Role-Playing Game
Shoot 'em Up
Shooter
Sport
Strategy
Traditional
Vehicular Simulation

A few notes about them as some will no doubt be questioning these 
already, skip this if you are only interested in the actual discussion 
we were having ;)
- Compilation is obviously not an actual gameplay based genre but it is 
a technical necessity to have it, unless you accept certain "games" not 
to be classified at all. Our database contains compilations as companies 
released them, no way around this really.
- Shoot 'em Up should be a subgenre of Shooter, we're aware of that. But 
again, usability creeps in. People want to differentiate between Shoot' 
em Up subgenres and we didn't want to introduce a third level of genres 
(we only have genre and subgenre now) just for this one exception. 
Usability trumps strict definition in this case, although you can of 
course write both definitions in a way that excludes the other.
- Traditional might not be intuitive, it refers to conversion of 
traditional real world games like board games, card games, slot 
machines, pinball etc.
- The controversial "other" exists, yeah. Please note thought that it's 
actually just a grouping for distinct and well-defined subgenres that do 
not fit the definition of any of the other main genres. If you read 
through my whole mail you'll see examples and reasoning for why I think 
even a real catch-all-oddities "Other" is okay for a taxonomy.
- Action-Adventure might stick out as a hybrid between often used terms 
Action and Adventure. You might notice we have no Action main genre as 
we think it is entirely useless label. Almost every game contains action 
elements and there's absolutely no way to properly define what 
constitutes an "Action game". E.g. the "Action" definition at MobyGames 
fits sports games perfectly as well. As a rough definition for 
Action-Adventure you might consider something like this: 
"Action-Adventures are all games mixing adventure game elements like 
exploration, story and puzzles with the physical challenges of an action 
game without either type dominating". That definition has its problems 
for sure, but it still gives you a good idea of what it encompasses.
- Arcade might be controversial too but I don't think going into detail 
here helps, let us just say it encompasses, among others, ball & paddle 
games and maze games.


>We don't want to exclude LCSH of the LC genre headings from this 

project; that would be throwing out the proverbial baby with the 
bathwater.  What we're trying to do is >develop something to add to the 
existing structure.  Also, we want to have the terms follow the general 
structure of LCSH/LCGFH, to make it easier to integrate with the rest of 
 >the catalog. If you've ever used the LC genre form headings for motion 
pictures, you'll find that a given movie can have three or four genre 
headings; we expect similar results in >the game world.  If a given game 
gets over a dozen genre headings, then the usefulness of the record 
comes into question.  The cataloger's "rule of three" will likely be 
applied: >We will choose the two or three terms that most adequately 
describe the game in hand.

Not being American and not being an active archivist or similar I'm not 
familiar with the LCSH headings but a quick round of research (aka: 
typing a few words into Google) indicates that those are more like tags 
too. I see things like "Ghost films", which seems useful if you want to 
know about games featuring ghosts, but not so useful if you want to know 
what the film is like. Think of Casper vs. Ghostbusters vs. A Chinese 
Ghost Story. I have my doubts that mixing this in with classifiers like 
Thriller or Comedy is a good idea. But maybe I'm just looking at the 
wrong list of headings!

In general though, I wouldn't look at movie classifiers too much. After 
all they have a vastly different accepted used of the term "genre" than 
games. And they also have a history of mixing narrative, formative and 
structural terms into the same taxonomy, if you even want to call it a 
taxonomy.

 > My brain may be wandering here, but the "cloud tag" versus taxonomy 
debate seems to be just another iteration of an old, long-running 
battle.  Namely, free text versus
 > controlled vocabulary.  I remember that debate raging while I was in 
graduate school (now approaching 20 years ago), and it's still going 
on.  In the words of Mr. Spock,
 > "fascinating."

It does indeed boil down to that I guess.

*Jim Leonard
*> In other words, I blame the MobyGames framework for not being 
fully-featured enough, but I still think the concept is sound and true. *
*
Jim, please note that I did not mean to critisize the system itself, but 
rather the implementation. I love having a well thought of multi-layered 
approach to classifying a game in place and your work in that regard was 
certainly pioneering (as was MobyGames as a whole). I just think there 
is a need for a traditional genre taxonomy on top of that.

 > The problem with this is that there are some games that are equally 
two or more genres.  Your single genre is therefore subjective.  I 
maintain that the goal of any taxonomy is to > be objective, so that 
there is no debate or confusion what makes up a game.

This is where we disagree then. I think this stems from your approach of 
having a very limited number of very broad genres. You have only eight 
basic genres on MobyGames, and one of those is Educational, which is 
debatable in itself. Of course you can find thousands of examples that 
equally blend some of those seven basic genres, that's a given. What I'm 
referring to when I speak of genres are much more granular sets of 
games. And I maintain the stance that you can classify every single game 
into one of them, with two exceptions:

- Games that feature distinct levels with completely different gameplay. 
You had a lot of these on the old computer platforms. You know, three 
levels, one a racing level, the next a platforming one and a puzzle in 
between. You can never place those in a taxonomy other than giving these 
mixes their own "genre". C'est la vie.

- Games that do actually define their own granular genre but that no one 
followed up on, resulting in a genre with so few entries that it is 
probably not worth having its own granular genre. These do indeed end up 
in catch-all kind of classifiers, but where's the problem with that really?

 > By definition, all role-playing games are adventures.

This is of course true, but only if you enforce an outside world 
definition of "adventure". See the Tetris example further down as to why 
I think that is a problem in itself.

 > I think that some people are uncomfortable with a game not being able 
to fall into a single genre, so they create many "meta-genres" to give a 
game a single label, even if it's not > the best way to classify something.

I can only speak for myself but this is not the reason why I try to have 
a "single label" system. My reason is usability of the database itself. 
I want to provide users an easy way to find similar games. Be it because 
they liked the initial game or because they are researching a certain 
type of game. For this purpose, it IS the best way, in my humble opinion 
of course. Again, I'm all for a multi-layer and tag based approach, but 
I think it should be an alternative method, not the only one.

 > "Indie" tells me nothing, and is not a genre

I think we can all agree on that. It's a very subjective term anyways 
which makes it inapt even as a simple keyword.

*Rowan Kaiser*
 >Well, there are two aspects to discussing genre which have come up 
here, I think. The first is genre as a classification tool, which 
theoretically gives people the opportunity to >determine the essential 
characteristics of a game in a word or three or ten, which is, I think, 
the aim of the initial question here.

Yeah, that for me is the purpose of a classical genre taxonomy. Being 
able to deduce the actual gameplay (to a certain degree) just from this 
single term attached to a game. This might seem naive and overambitious 
but so far I've successfully (I think) done it for around 2000 games 
across all generations of console games so I still think it is possible.

 >On the other hand, there's genre as a social construction, which, 
regardless of accuracy (how is /Tetris /a *puzzle* game? It's an 
abstract action game!) is used and will be used. This, >Jim, I think is 
where you had problems with your RPG classification. You may consider it 
inaccurate, but the social construction is popular shorthand. It is how 
gamers understand >games, I think.

 > Still, it's part of that labeled genre and I accept that, even if I 
think it's kinda silly.

That's exactly what I am referring to above in the adventure vs. rpg 
point and below in the Tetris example.

 > I think you could do a pretty comprehensive labeling system if you 
base it around three categories: Gameplay Genre, Setting, and 
Perspective, in descending order of importance > (this system would not 
look so different from MobyGames, which is part of why I like MobyGames!)

Heh, believe it or not but when I started designing my database I 
actually had a table prepared with exactly those three columns. There 
was a fourth one though I think, let me dig out that document, I must 
still have it somewhere. Aha, found it, I actually had two more columns: 
Style and Technology. I'm not exactly sure what those were meant to be 
about but I think Style was meant to distinguish between realistic and 
arcade approaches and Technology between sprites, polygons, etc. Looking 
back I don't know exactly what I was thinking there, but hey, we all 
start somewhere ;)

As for perspective, this is a pet peeve of mine. People seem to attack 
any use of perspective as part of a genre name, like in First-Person 
Shooter. The reasoning I always hear is that genres should be based on 
gameplay mechanics and different point of views are just different 
representations of the same thing. To this, I can only shake my head in 
astonishment. While I can agree that a different point of view does not 
necessarily result in a different gameplay experience, I cannot accept 
that it is deemed impossible that a different perspective can(!) result 
in entirely different gameplay. First-Person Shooter does not equate to 
"Space Invaders from the cockpit perspective" and thus to two tags 
"shooter" and "first-person perspective" but rather labels a very 
specific set of gameplay elements that result in their own unique 
gameplay experience. Sure, you could probably come up with a fancy 
alternate label that does not incorporate the perspective in its name. 
But how is that more useful really? Again, I think the problem here lies 
in accepting the rather abstract meaning of a term within this limited 
universe of discourse as opposed to its meaning in general use of language.

*Tetris*
This is a really good example of the problems that can arise with genre 
taxonomies. Established genre names do not necessarily match up with 
their literal meaning. Even more so, a game can fit a genre name 
perfectly from an outsider's point of view and still not be part of it. 
Not every game with puzzle based gameplay is a "Puzzle game", see Point 
and Click Adventures. Not every game where you play a role is a 
role-playing game. Trying to ignore the existing genre names and, even 
worse, redefining a main genre while using the same name will not result 
in a taxonomy that's useful to most of its users. They might not make 
sense objectively, but they're there and established, we have to live 
with that.


*Andrew Armstrong*
 >I'd agree with whatever actually works; show me a system and if I 
can't find certain games with certain search, tag or category terms it's 
not encompassing enough (be that by >content genre, gameplay genre or 
abstract "social" genre).

This I would agree with, I measure the success of a system by its 
ability to actually use it for filtering and finding what I intended.

 > I think the ranking of different genres is the important point; the 
choice of divisions or inclusions of different ones is great, but having 
it priorities so at a glance you know if a
 > game focuses heavily on one mechanic or gametype over another would 
be even better (where it also notes secondary ones though).

To me, that is what the classical genre taxonomy provides. It does not 
state "this game only has role-playing elements" but rather "this game's 
dominant mechanic is role-playing". That for me is the beauty of 
combining such a system with a tag cloud. You have a (hopefully) clearly 
defined term that lets you know what you can expect in a game but you 
also have tags that tell you about all the other aspects of it.

 > For instance, many games include multiplayer (surely some form of 
genre we'd all agree!)

I would disagree, at least in the sense I think of genres. But even with 
a multi-genre tag approach (Strategy+Action etc.) I would not accept 
Multiplayer as a genre but rather as a mechanic or feature.

 >Knowing the focus and intent of a game, where perhaps it does contain 
a side quest where you race stuff but isn't really a sport game, but 
perhaps that side quest could be actually >a widely accepted unique 
selling point and gameplay feature especially at the time so important 
to note, having some division where it can be noted but not be the 
primary focus >would be cool.

Oh, definitely, tagging a game with racing elements (if only one 
subsection) is worthwhile. But if you only have a "racing" tag without 
an attached importance value this would mix it up with "real" racing 
games. That's the problem with simple tag clouds. It's not a problem 
though if you have classical genre taxonomy along with your tag cloud. 
The tag racing would show this game, the genre racing wouldn't.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://six.pairlist.net/pipermail/game_preservation/attachments/20110615/b94a8635/attachment.html>