[game_preservation] Descriptive terms for Video Games

[Jan Baart]

Hello Richard,

I guess I have to be the one disagreeing a bit with what others said, 
although not completely. First let me explain that I have been looking 
at genre classifications for a very long time and have recently had the 
"pleasure" of doing my own for a database project. Due to this I've read 
almost every paper or article on the matter and have found most of them 
to be quite inconcise and contradictory.

I'm with everyone in that a multi-tag approach is the most modern and 
flexible way to handle genre classification. What I disagree with is the 
idea that a traditional genre classification approach (as in, a game 
"belongs" to a specific genre) is obsolete and not useful. This idea 
seems to stem from the inadequate nature of all existing classifications 
and the general mess publications, especially the big websites, make out 
of it. Let me thus adress a few typical areas of criticism towards a 
strict genre system:

"Games can belong to multiple genres!"

My answer to this is, yes and no. Of course a game can be both a 
Strategy game and an Action game. But does that mean a genre 
classification is not possible, or does it mean the chosen genres are to 
broad and not well-defined?

"Creating more specific genres results in a lot of genres which looks 
ridiculous!"

Yes, having 100,200 or even 500 specific genres looks ridiculous. But of 
what relevance is that. All that matters is how useful the taxonomy is. 
If you have 50k games in your system then even at 500 genres you still 
end up with groups of 100 similar games. Are 100 existing games of the 
same specific gameplay worth their own genre? Ask any Tower Defense fan ;)

"What about oddball games like Pirates!"

Well, there'll always be some of those. Or games that define a new genre 
where you'll have to wait for more games like it to make calling it a 
genre worthwile. But, does having a few games you can't properly 
classify (stuffing them into "Other" or similar catch-all categories) 
having a working taxonomy for the rest worthless?

"Placing a game into one genre is not needed in times of internet 
databases with tagclouds etc."

The problem here is twofold. Firstly, even if that tag-based database is 
perfectly designed and tagged, that does not mean it allows you using 
the tags in a search as you had hoped. Secondly, tags don't have 
priorities, nor do they have a value. Simple example: An action game 
with RPG elements (Actraiser?) gets tags "action" and "rpg" and a RPG 
with real-time battles (Secret of Mana?) gets tags "rpg" and "action". 
Great, if that's all you've got there's now no way to distinguish them 
in your database. That's not useful for a user at all.

"Modern games blend too many genres and defy classification"

Do they? Yes, lots of genres have visible influences on a multitude of 
games outside of their core genre, like RPG elements in FPS. But does 
having stats in an FPS not make it an FPS anymore? Hardly. And if there 
are indeed games using an even mix of various existing genres, do they 
maybe define their own new genre by this? Again, how does this make a 
taxonomy less useful, especially if you're looking at a huge number of 
historic games not affected by new occurences?

Now that I've responded (admittedly, no full answers but rather "food 
for thought") to a few common complaints about genres, let me show you 
an example of why I think a clear two-dimensional taxonomy is still useful.

Let's assume I want to find games similar to Final Fantasy Tactics, an 
pretty easy example. Obviously, I use the biggest game database on the 
internet, MobyGames. Here's the entry for the game:
http://www.mobygames.com/game/final-fantasy-tactics

Now, we have two genres and one perspective tagged, as well as a 
turn-based mechanics tag. Even if I filter for all that in the database, 
I still end up with games like Heroes of Might & Magic or Master of 
Magic. Now, admittedly, those while not being quite the same are at 
least similar. But then again, it was a very specific type of a game, an 
easy example. Try doing the same for StarFox and you'll end up with 
Comanche, Battlefield 1942, Falcon 4.0, LEGO Star Wars, Spyro the Dragon 
and Tribes 2, partly due to their tagging and partly due to the fact 
that apparently you can't filter for two "non-sport" tags at the same 
time. Though none of the tags that StarFox has make it apparent in any 
way that this is actually an "on rails" type of game anyways, so there's 
no way filtering using the tags it has would ever "get rid" of games 
that are not on rails, thus never resulting in a selection of actually 
similar games.

Again, this does not mean that we shouldn't pursue approaches like 
multi-category tags (setting, perspective, concepts, ...). I'm all for 
it. But in my opinion there are good reasons not to abandon a 
traditional genre taxonomy. Which is why at our database website we use 
both. We have tags (not visible yet but they're there) but we also have 
an editorially tagged genre. One per game. This might seem outdated, but 
in the end it serves our users.

Sorry for the long and controversial read. I'm sure a lot of you 
probably won't agree with a lot of what I said but there you go ;)

Jan

P.S.: If you're interested in our current taxonomy I can provide you an 
internal document that gives game examples. There are plans to make this 
public along with definitions and write-ups exploring the specific 
genres, but that'll still be a while!

On 09.06.2011 23:14, Pugh, Richard wrote:

>

> Greetings, all.

>

> Some of us at NAVCC are building an archival collection of video 

> games, both console and platform based.  We are cataloging the items 

> using conventional machine-readable cataloging records, and the 

> standardized Americal Library  Association descriptive rules for 

> computer files.

>

> While the descriptive aspect of this project is doing very well, we 

> are having some trouble with subject analysis and genre selection.  We 

> have found the Library of Congress subject headings to be sorely 

> lacking, and the new Library of Congress genre form headings are (so 

> far) a minimal improvement.  We're looking into using a different 

> thesaurus of descriptive terms, or, developing our own list.  Has 

> anyone in here found or developed a thesaurus (or term list) for 

> describing video games?  If so, I would be interested to know more 

> about such lists.

>

> Thank you,

>

> Richard J. Pugh

>

> Motion Picture, Broadcast, & Recorded Sound Division

>

> National Audio-Visual Conservation Center, Library of Congress

>

> (202) 707-6636 / rpugh at loc.gov

>

>

> _______________________________________________

> game_preservation mailing list

> game_preservation at igda.org

> http://six.pairlist.net/mailman/listinfo/game_preservation


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://six.pairlist.net/pipermail/game_preservation/attachments/20110614/82afb394/attachment.htm>