Game Ploy

Games and the G.U.I.

Buckminster Fuller said we are children born of three parents; mother, father and television. Caught in this rationale it is cute to think that we are raising the Sony beat generation and panicking at the thought of what the third parent is telling the children. It is not surprising, given the rate of divorce, that everyone wants to keep the children to themselves. Like the parents who forbade their child from watching Thomas the Tank Engine because they thought he (Thomas) was state-owned, it is easy to be premature.

The desire to write games and so explore the nature of computing has driven some of the innovations pushing forward the industry. Unix, the operating system running most of the workstation and up-size machines (read: your Bank), is rumoured to have been invented to allow its program­mers to write a multi-player version of a StarTrek game called Space War. Curse the software which allowed the first full screen text editors (forerunners of word processors like WordPerfect), which, apparently, was created for the character based maze/adventure games' rogue and hack.

A game has, by its very nature, only to be true to itself. As such, user interfaces have not had to conform to the user interface standards laid down by particular operating systems. Games are the vanguard of future interface design. Games were the first with graphics, particularly hand-helds. The `Sonic' interface, with sounds played as you manipulate icons, was based on the observation that players perform 30% better with the sound on. Games introduced the idea of a real world metaphor created in software. Animation from disc first made its appearance on a game, and the present 3D games foretell the 3D user interfaces to come. Given this freedom it is surprising to see the lack of variety in game design. While there are many great games, the majority are sadly limited by the imaginations of those people who write and back them. Most games are a disappointment.

It might seem picky to try to differentiate between a toy and a game, but this is important to do. Toys are things less than abstract and unfettered by social conventions. Children uninhibited by narrative can turn a doll into an aircraft by the addition of a runway and an engine noise. Games alternatively are societies - there are rules, penalties, limits, goals, agreements and players. Most of all, games share narratives, scores are kept, there are winners and loosers - it is impossible to cheat with a toy, it is what you want it to be. The role of play in society swings between the private (toys) and the public (games).

A spreadsheet is programmatically identical to code written in a programming language like Basic. This similarity has confused and still confuses many programmers. Both systems are interpreters, they are both primarily numeric and symbolic. The spreadsheet invites the user to explore what-if-senarios, there are no dreams stronger. Code written in software like Basic is highly sequential, spreadsheets are not. Spreadsheets are continuously recalculating; the Basic code narrative is calculated once. Similarly, a typical game has a high narrative content - collect all the magic. rings to release the Princess, follow the bomb's fuse to stop it exploding.

Apple pioneered the exploitation of human-centred computing to the mass (read: post-military) market. Apple came down from on high, from the mountains of PARC, and brought with it the tablets of stone. These tablets were then given to the people in the form of the Apple User interface guidelines. The difficulty of programming for a window based machine is to do so without a narrative. The Ideal program for Apple exists without modes, sequences, preconditions; everything should be accessible for as much time as is meaningful. The practical consequence of this was that any program which asked the user to enter say the page size before they could start typing was labelled as confining (read: junk). Narratives in software need explanations and stories need story tellers (online help manuals, training). Having taught this kind of software to undergraduates for a number of years, I have realised that all I have to do is give the students permission and time to play. To many adults the notion of play comes hard, and given the current design of software this gives the advantage to the children.

Doug Elbert (from his hand the first mouse sprang) described computers as potential toys for thinking with. Toys, in his mind, developed thought: speadsheets, word processors, image manipulation programs, drawing programs all fall under this category of facilitating thought. The desktop interface of the Macintosh Finder is, as people love to point out, unlike a real office. It is as similar to a real office as a Barby doll house is to a real living room; the Finder or the magic cap interface is a toy office, it only has to simulate the real thing in as far as it invites you to play with it. This miss-reading of the function of the interface held back the introduction of direct manipula­tion in computing for many years. Most hyper-intelligent Management Information System executives - spending millions repurchasing computers capable of running Windows - don't like being reminded they used to call such interfaces toys.

Toys invite experiment, toys function as creative aides; like paint to an artist the content is undetermined by the paint manufacturer.

Compare this to games: games invite winning stratagies, sides (human-human, human-machine) imply agreed rules and rules imply a deliberate simplification of an identifiable limit to fairness. Rules however breed inherent weaknesses. Programs are a number of symbols and rules running together, and just as software has bugs (unforeseen and disastrous combinations of symbols), rules systems have fundamental points of vulnerability. The game player (like a network hacker) finds weaknesses and exploits them; Kasparov in a battle against the world's best chess computers looked for consistent faults in the software [Wired 1995 Jan]. In 'Elite' it was possible to flee from an enemy and fire backwards, by reversing the joystick you could shoot at the attackers and still remain out of their range. 'SimCity has similar inherent limitations; industry creates crime, housing does not, wells are more productive if they are stuck on piers. Success in a game like 'Battle Zone' (an arcade tank simulator), which involved strategic reversing, could not be transferred to combat a real tank. Toys roll towards emulating reality -games conversely push the player away from it and towards game success.

Initially, what attracted me to 'SimCity' was its apparent openness, you could experiment with cities and urban planning. `SimCity' was a toy, there was no score, you never got killed, you could even print your city out. I realised that games which appealed to me were those which allowed some creative contribution. If anything, this was as close as anyone could get to an interactive experience. The narrative in these games was being created as the player interacted. 'Adventure', 'Elite', 'SimCity, all had a common quality of being flexible enough to play in the user's head (MUD -Multi User Dungeon - a true virtual reality if you must). By comparison, most games take place in the game's space. The fundamental question is who owns the rules, the player or the played? To reappraise Doug Elbert, a toy is a tool for computer assisted fantasy. Games like these are hard to write and even harder to think up, but it does seem they become very successful.

If the computer has an example of shining good design, it should be these games. If all software can reach this level then perhaps the three parents don't need to divorce.