Thursday, February 07, 2008

The 500 Club

This post will nominally be about text sims, but it's really more about game design, so if you're not interested in text sims but are interested in game design, just hang in there.

I have a deep affection for text sims. I'll never forget the first time I played Championship Manager (97/98 edition), the legendary sim created by Paul and Oliver Collyer. I was near the end of a season, fighting for promotion, when one of my players got the flu. Then another. Over a period of about ten days, most of the players on my team got the flu, and I desperately shuffled line-ups, trying to save the season.

It was a moment that felt so real I forgot that it wasn't.

Since then, the concept of text sims has always been irresistible, because the world of a text sim, theoretically, is almost infinitely deep. And the possibility of getting lost in that depth is incredibly appealing.

On a side note, that's exactly why Dwarf Fortress is so wonderful, even though it's an entirely different kind of game.

Here's the thing, though: in spite of my conceptual love for text sims, and my desire to play them, I've played them less and less in the last ten years. A graph of the time I've spent with them since Championship Manager 97/98 would look like a ski slope, and at the left of the graph would be the top of the run.

I want to play these games. I'd like nothing better than to spend a hundred hours locked into some sports world. So what's stopping me?

And more to the point, what's stopping everyone else? Why do so many of these games never break 500 copies in sales? Sure, there are exceptions, but I'd be willing to bet that most text sims don't break the 500 mark for each version of the game. So what's going on?

Well, I think it's almost entirely due to the concept of push versus pull.

Do you remember the Pointcast Network? I was working for a computer company in 1996, and one morning a friend of mine just said "watch this." Then he clicked a target icon in system tray and all kinds of news stories popped up.

It was a miracle, at least in 1996.

That "miracle" was Pointcast--a push technology that would send news stories right to your desktop, based on the kind of news you wanted to see.

Of course, I went right back to my desk, downloaded the program, and started using it, and so did almost everyone else in the company.

Which, um, kind of crashed the network.

Two days later, the Pointcast Network was banned. In those two days, though, I'd seen the future.

You may not see the connection, initially, but that's exactly what's wrong with text sims. As players, we're trying to manage a world of data, but instead of having data pushed to us per our preferences, we have to go pull every single piece of information we want to have.

In a text sim, that means there are literally dozens of categories of data that all have to be manually managed by the player. Thousands of individual pieces of information.

Pull, pull, pull.

Even for an experienced player, this can be daunting. For a new player, well, forget it. Text sims have no new players, or very few, because they are just ridiculously intimidating. A few sims--the Total Pro Golf series, in particular--have made a focus of becoming more user-friendly, but the vast majority of text sims have interfaces that make feel like I'm looking at the controls of a German submarine.

There's a reasonable explanation for this, really. Text sim designers and programmers are data guys, not interface guys. These games are usually one-man shows, and the designer does all the programming. The design, though, usually focuses on incorporating all the different data types, not making the program easy to use.

That's a huge problem. Detail without accessibility might as well be invisible. Hell, the way most of these interfaces are designed, that detail is invisible.

I believe there's a much stronger appetite for the depth that text sims offer than is generally believed. The obtuse, difficult interfaces, though, make it impossible for new players to become converts, so people are misled into believing that the market is much smaller.

So what would an accessible text sim look like?

Well, there are two major design areas I'd like to discuss: navigation and information.

Many text sims have absolutely awful navigation systems. It's impossible to know where I need to go, or when I need to do something.

That doesn't work.

The vast majority of things you can do in a text sim are either events or activities. By "activities" I mean something the player initiates. "Events" are hard-wired into the world--games, recruiting deadlines, etc.--and they're scheduled to take place.

An accessible text sim would be based on a series of hubs that would facilitate both activities and events.

Centrally, there would be an office hub, and all game activities could be performed from the office. That would be the activities hub. Gary Gorski has an interface like this for his games--it's actually an office, and different activities can be performed by clicking on the appropriate icons. Want to answer an e-mail? Click on the computer. Want to make a phone call? Click on the phone.

That's the activities hub.

Additionally, there would be a calendar inside the office hub, and that calendar would function as the events hub. Everything that was scheduled to happen in the game would be listed on that calendar, and every listing would be clickable to take you to that event.

Oh, and beside every listed activity there would be a "help" link which would take you to the specific help section for that activity.

There's the events hub.

We need one more hub, though. In any text sim, there are a blizzard of reports available. The problem, though, is that we have to manually go and pull that report every single time we want to see it.

Pull, pull, pull.

Has that report updated since the last time we viewed it? When does it update? Who the hell knows?

This third hub would be the "push" hub. Inside the office hub, it could be the geeky looking guy sitting in the corner with a calculator. Click on him and you see a list of every available report in the game. All of them. And instead of going and pulling them one at a time, you can specify which reports you want delivered, and you can specify how often you want to see them.

When reports are ready, a fat manilla folder appears on your desk. Click on it and see all your reports.

Don't make me pull all that information. Let me configure it at will, but once I have, push all that information to me. Do that for every single category of data possible.

So we have a central activities hub (the office), and inside that hub we have a master events calendar (events hub) as well as a reporting function (push hub).

Every single screen in the game should give you one-click access to any hub. No backtracking through multiple screens, and if you ever get lost, you can immediately return to any of the hubs.

Text sims have a staggering amount of data, which means that it's even more important that the interface be logically designed. If it isn't, the data and detail are irrelevant. Most text sims have incredible depth, but if the information I need is two miles deep and all I've got to dig with is a spoon, isn't it kind of pointless?

Then there are the mouse clicks. Man! Reducing the number of total mouse clicks by 20%, or more, is far, far more important than adding a dozen more ratings categories. As far as I'm concerned, significantly reducing the amount of clicking necessary to play the game IS a new feature.

All right, this has gone on way too long, so for both of you who are still with me, thanks for hanging on to the end.

Site Meter