3 posts tagged “academia”

View other items tagged “academia:” Photos, Audio, Videos, Books or Links. View “academia” on Vox.

Vocational vs Academic: The Struggle of the University

May 13, 2009
Post a comment

As a graduate student, I consume knowledge like I consume cake: ravenously and without thought to my general well-being. It's supposed to be a useful trait, or so I'm told. However, I am one of the very few people who are like this. It's believed less than 1% of the population gets a PhD, and for good reason. This is a career, and it isn't for everyone. That said, I'm not one of those people that think university is entirely for academic pursuits either.

There is a constant push and pull between academics and students. Academics believe that they're creating the workers of the future. Giving them the building blocks for fundamental understanding is much more important than giving them experience in the vocational tools they'll use in the short-term. Universities create engineers, vocational apprenticeships create car mechanics. There's nothing wrong with creating mechanics. The world needs all types of people. However, the farther up the university quality food-chain you go, the harder it is to convince anyone that a vocational course has a use.

With Computer Science, and video games especially, this tension reaches a breaking point. I've sat in many classes where I've felt the academics have severely impeded my learning... I needed to know the "hows" and not the "whys". I didn't care how the naïve Bayes algorithm was derived, just what it did. Even at graduate level, I run into the same things.

The added difficulty with video games is that game studios are constantly pushing on universities to be a training farm for them: teach undergraduates C++, XNA and 3D Studio Max and get them out (the irony being that game developers then say in Edge that the Computer Science graduates are preferable to those with "lesser" games degrees). This is what the academics rightly push back on, but it seems many go too far. It is important to give undergraduates an idea of what a commercial toolset looks like. Neglecting that for some idealistic wrongheadedness that forces students to always use open-source software, work by themselves (teamwork is hard!) and be damned with the real world is hurting everyone.

A moderate approach is needed: the strong basics that Computer Science has always had, coupled with an idea of what the hard road of video game development is like. A class on XNA can only benefit, an entire curriculum based around it hurts everyone. This sounds obvious (at least to me), but you would be surprised about the number of professors for whom even this is a worrying proposition.

I trawled Google looking for a relevant quote to end with, and found this by Lawrence J Peter: "Education is a method whereby one acquires a higher grade of prejudices."

That'll do.

How To Argue on a Subject You Know Nothing About

Jul 21, 2007
6 comments

The Argument of Narratology (possibly)

Did you know that, in academic circles, there are two points of view on how to study video games? No? Syntheticist doubts many gamers do. In a very condensed version, here are the two warring camps.

Hamlet on the Holodeck: The Future of Narrative in Cyberspace

Narratology
Narratology is the study of video games as "cyberdrama"; that all games have an inherent narrative element that cannot be separated. The classic text on narratology is "Hamlet on the Holodeck" by Janet Murray. Star Trek's holodeck is an example of a cyberdrama, the "player" takes part in a story.

Jane Douglas, who wrote "The End of Books - Or Books Without End?: Reading Interactive Narratives", which primarily studied interactive fiction, describes a section of "cybergaming geeks" (boys who play games) that are too obsessed with their "joysticks".

Many narratologists have a media studies background.

Half-Real: Video Games between Real Rules and Fictional Worlds

Ludology
Ludology is the study of video games which places great importance on the rules of the game. A video game is a modern way of playing a traditional game, and the interface and ruleset are very important. The narrative that comes with a game is a sideline to this. This does not exclude narrative as an important part of games, but just that games themselves cannot be meaningfully analysed without looking at the way a game plays.

Jesper Juul is oft-cited as an influential ludologist, and writer of "Half-Real", a book which attempts to create a framework for video game studies. Syntheticist is currently reading this book, and heartily recommends it for those interested in video game criticism above and beyond that of mere reviews.

Quite aside from the insulting stance of Douglas (which one would hope is not typical of all narratologists), this argument, from an external viewpoint, is utterly absurd. How many narratologists have actually taken the time to ask a video game player what they think? Surely 99% of gamers today understand that a video game relies on its gameplay, or in some cases, it's story. But games absolutely cannot be "reviewed" without looking at the gameplay. Or the social interactions it enables. Or the feelings they invoke.

Can one truly analyse genre-defining games such as Doom, Civilization, or Super Mario Bros. from a narrativist perspective? Absolutely not. It's nonsense. These games are fundamentally reliant on their rulesets and interfaces. To study Mario 64 without analysing its control scheme will miss everything that is wonderful about the game. It seems that by virtue of having a PhD, this is a qualification to study things that obviously you know nothing about. This is rather akin to Syntheticist releasing conference papers about "insular spotty bookworms" who read mountains of books, without ever actually reading any of the great novels that our culture has created.

Fortunately, this debate hasn't spilled out into the wider gaming society, but it is tying up conferences and research time that could be much better spent actually understanding where video games fit into our culture and how we can use that knowledge to create better gaming experiences.

Artificially Academic Intelligence

May 24, 2007
Post a comment

How much longer will we put up with zombie AI?

Synthetic spoke to a university professor yesterday in regards to artifical intelligence in video games.

"Oh, they don't worry about us [academic AI], and we don't take much notice of them", he replied.

Academia has never traditionally been concerned with the pay-day. Discussing why we must spend millions of dollars to study what would happen if two black-holes met (this is an actual conversation Synthetic has had) is a nonsense question; it is studied because we don't know what happens yet.

The same is true for artifical intelligence. It would be a fallacy to say that AI research is not concerned with speed, but the "big problems" are ones that will eventually be processable over time, it is safe to rely on other advances in technology or algorithm design to perform processes quickly enough. The important thing is whether these processes work.

Video game AI has no such luck; everything must happen quickly, and usually, is of a lower priority than something easily marketable. As Nareyek puts it:

Humans are very visual animals, and a beautiful sunset is much easier to sell than any particuarly clever reasoning capabilities of an opponent.

Is this about to change? Will AI become more important to developers, and will we see a convergence of academic and video game AI? Very probably (Nareyek also notes that this has been said for a long time). Multiple cores on consoles are offering real opportunities for paralellism, but that doesn't solve the marketing problem; just fill those cores up with yet another pointless shine effect and release the game.

If we presume that Fracture will be all it promises, we have reached the boundaries of physics sandboxing (for now). And we have been happy with Crackdown, GTA 3 and others to give us a sandbox and let us play with the rag-dolls inside; no-one will call the AI in these games spectacular. There are simply too many variables, and the player has too much power to shape the world around them.

By reaching this point, developers have forced their own hand. AI is the primary means of adversary in video games, and that is not likely to change for a very long time. Agents will now need to respond intelligently to the changing surroundings, and this is going to mean more than simple path-finding. Combine this with the sheer number of objects which now populate these worlds, from light-switches to blenders to vehicles to coffee-machines, the AI now has to do much more than run directly at the player to appear believable.

Assassin's Creed and Splinter Cell 5, with their focus on crowd dynamics, might well just be the tip of the iceberg, and maybe there will be some developer one-up-manship going on to see who can implement the most intelligent agents from the wealth of academic research out there.

And maybe, just maybe... publishers might find a way to market it.