First off let me say that I really enjoyed this game. I thought Budget Hero took a very complicated problem, the federal budget, and simplified it in a way that is understandable. Plus they give extra insight into the different spending options, giving information on both sides of every issue, and provide a way for people to find more information about budget related subjects. With that said let me move on to the review.

Overview

The game allows players to drag and drop spending/cutting cards into their overall budget area on the right side of the screen. Each card has a value that will add to the budget or take away from it. The player inherits the current US budget so the cards values are based on the current budget (the game cites resources for all of its value assumptions).

When players choose cards they affect the following areas:

1. Badges – Players choose three “badges” that represent their policy. I chose energy independent, health and competitive advantage (for America). Other badges include national security, being Green, efficient government and safety net. Certain cards in the game will contribute to a player’s badge score. From what I can tell there are three cards that help each badge. If a player places these three cards in their budget then they get awarded that badge.
2. Deficit/Surplus – How much is a player’s budget is taking in verses spending (max is $600 billion).
3. Size of government – Is the player’s budget a big or small government budget (based on percentage of GDP) (min is 12% and max is 24%).
4. Budget Bust – When will the player’s budget go bust or fails (max is 2070)
5. Debt – What is the country’s debt based on % of GDP (min is 20% and max is 40%).

The player can choose cards from 8 government departments, the country’s taxes, and the I.O.U. does not have any cards (only showing the interests that occurs on the overall debt). Each department has its own set of cards related to that department and the tax cards obviously have to do with how the government taxes its citizens. There are a lot of cards so I will not go through them but each provide more information beyond their title. Double clicking a card will give you an expanded card screen with the total budget effect for the next 10 years, a general overview of the card, the pros and cons of the card, the impact of the card and the real world source that corresponds to how the card was balanced.

The player gathers cards on the right side of the screen, I ended up with 27 cards. A player then can see how their budget will play out and get a summary screen. Here the player is given the following information:

1. The player is matched against the current budget: Did the countries debt change, when will the budget fail, and how did the size of the government shift.
2. What badges the player was awarded.
3. A player’s biggest cards.

A player can compare their budget with others that have submitted their budgets or a player can print out their budget (which I thought was a really nice feature). Comparing goes by US residency, zip code, year of birth, gender, income level and political affiliation. A player can also go back and alter their budget if they wish, which is also nice so a player can go back and shift around a few things to see the difference.

That is basically the overview for the game. Now for some of my insights.

My Strategy

I took away a bunch of Bush’s tax cuts (I started taxing the rich though) and I was able to put every spending card in my budget from the smaller departments: education, science, home, misc., and infrastructure (though I took out the ‘pork barrel’ spending in the misc.). I cut some of the bigger things from the military (cutting the spending by 10%) but I kept foreign aid and created new military units so in the end I didn’t take away that much from the military. I was able to give health care to everyone (I think that was the largest single spending card in the game) and I even increased Social Security spending.

My surplus was over 600B, the size of the government was under the middle range, and budget wouldn’t bust until after 2070. My budget print out can be found here. It really can’t be this easy, just by cutting those tax breaks and taxing the rich, can it? Plus I met 2 out of three my badges or health and competitive advantage. I didn’t get energy independence because I refused to drill in Alaska (kinda BS if you ask me). The other cards “increase mass transit funds” and “fund congestion study” were the other two major energy independent cards which I did fund. If there was a “give money to research alternative fuel programs” I would have gladly went for that.

Card Woes

If cards depend on one another I would like to know why? Some are a little more obvious, e.g eliminate ‘pork barrel project’ means you cannot cut ‘pork barrel spending if half’. Some of the tax stuff however was weird since there were a couple different configurations for the same tax decision. Maybe each card could have extended options that could expand down or cards that can’t be used should be taken away.

Technically the whole middle section of the screen is under used when working with the cards since players do not need to check the overall budget for each department often (each building allows the player to scroll over the building and see the overall spending of that department). Making the card area bigger and turning the card area transparent when a new card is selected or having an option for the player to turn the transparency on when they wish to view the overall budgets would be other ways of using the middle space more efficiently.

+ and – for the budget numbers are used differently for the tax cards (+ gives you money) and the department spending (+ takes away money). A little confusing at first.

Last, the game needs more cards. Being an academic I wanted to see a ‘Give lots of money to NSF’ card

Summary Stuff

The game allows a player to print out their budget but the print out is not in color. This makes it hard to see which policies go to which department within the game.

Comparing the budget to other people was great however the data needs to be put into proper visualization software because the game screen is too limited.

Although one flaw of the compare data is that it is not screened. I forgot to take a screenshot of the compare screen so I quickly reloaded the game and did a no card budget. Meaning I didn’t change the budget or play the game at all. Not thinking I typed in my real information again into the compare screen. There were only 2, male, age 20-29, in Georgia people (really only 2 from Georgia in all) when I got to the compare screen. I take it the first one was me and then the second one was from me again ( in the screen shot), where I didn’t play the game. Thus the ‘average’ score was then in-between my previous score and the one without playing the game. Obviously this doesn’t help when comparing across people because you can’t tell how long people played for or if they are telling the truth when they put in their demographic information. Kinda stinks cause I liked my first score.

GUI errors/concerns/annoyances

These are just things that I have to say about the GUI in general.

1. The cards cutting off on the top of the screen while players cycle through them seems weird. Should have put a border around the area.
2. The badges take up space and could be tighter.
3. The badges get filled as a player picks cards but this should be more explicit. Players should know how much more they need within each badge to achieve the badge, especially stating which cards work towards which badges.
4. The textbox size for the card cost is just a little too small at the top of expanded card view. The bottom of the numbers get clipped ever so slightly (I believe on the 1,000 and over numbers only)
5. The ‘Impact’ box in the expanded card view should be placed under the pros and cons list or in a tab since the text is too squished together.
6. The sources that are quoted in each expanded card view are nice but should be click-able and linked to the source material (if possible).
7. When clicking on a building the left arrow for the card cycling flashes on the screen but disappears because the player cannot move the cards to the left. This is extremely picky but I hate it when GUI elements draw themselves on the screen when they should not appear.

Conclusion

All in all Budget Hero is a great game. I agree with Ben Sawyer, who sent out an email about this game on the serious games email list, that more budget games like this should be created. Since games usually teach players how to manage resources anyways I don’t see why this cannot be achieved.

This is turning into quite a blogging filled day. I really should spread these out, make it look like I’m writing more.

I’ve started using Python recently because everyone was saying that it is easy. It’s actually pretty sweet, reminds me of the lingo scripting language used for Adobe Director. I also stumbled upon an AI blog, AiGameDev.com, written by Alex Champandard that also highly recommends Python as an AI language. So I think I will be sticking with it for the time being.

On another note about AI I’ve been recently thinking about AI and user interfaces for games. I was talking with a fellow PhDer Jichen about it and she pointed me towards two sides of this debate (though not really about games just AI and UI in general). On the one hand there are people like Ben Shneiderman and Donald Norman who love transparency when it comes to designing interfaces. Then there are people like Pattie Maes and Nicholas Negroponte that think users should not have to worry about what is going on in the AI just that a person and the AI function well together (Ambient Intelligence), I would call this concealed AI. Both sides are all fine and good for everyday design, for different purposes of course, but what about for games?

I started to look for examples of both transparent and concealment in game AI. First, the most notable transparency example would have to be The Sims franchise. The interface allows for a lot of information to reach the eyes of the player. Plus turning off the freewill of the agents in the game gives users complete control over all of their sims.

Now contrast this with a different game, Façade which doesn’t even have a UI. Here the notion of concealment is more noticeable the characters are supposed to act like real people holding a conversation with the player using natural language without the player having a lot of information about what is going on in the heads of the characters. However the player is supposed to recognize how the characters feel by evaluating their responses, such as tone of voice. The dialog becomes a sort of transparency output for the player. This is a little weird though because if we want to take AI to the next level, especially in games, does this mean developers would have to make sure that each spoken sentence a character says needs to carry an emotional overtone? I’m not saying that it’s a bad thing but deciding whether the characters could flat out lie to the player based on their tone would be a decision that would have to be made.

Then the middle ground between these two extremes s concealing the UI until it is needed or requested. Black and White 2 and Viva Pinata do this well but also rely on very little UI to begin with. B&W2 put sliders over the head of a player’s creature in order for players to teach their creature how they should act in a given situation. The player can also direct the creature around if they choose or they can give them high level commands like “be a builder” or “be a soldier”, in addition to freewill.

Viva piñata on the other hand has less intense creature AI but it has better looking creatures . In VP the player only gets a happiness UI bar for the creatures. This really seems all that the player needs because otherwise the creatures just walk around and hang out. There is a creature “information” page where the user can set a creatures name (which makes them happy) and where you can find their current romance status (creatures need to fulfill preconditions in order to be able to mate and reproduce). Both Black and White and Viva pinata take a middle approach to transparency and concealment, trying to give the player a sense that these creatures are real but allowing the players a little more control.

So which of these do players like the most? We do know that The Sims is the best selling game franchise of all time, B&W2 and VP did okay in terms of sales, and Façade wasn’t commercial but did get over 100,000 downloads. However I’m not going to pass judgment so blatantly using only these four games especially since Façade wasn’t even a commercial game. The Sims is highly customizable and has multiple expansions compared to the other games so that is just one reason it has done so well.

But I think it is important to note the differences in what each of these games try to accomplish. Developers that want to take a ‘real’ approach seem to use concealment, in real life we don’t know what other people, or creatures, are feeling or thinking unless we use our senses. The Sims approach gives players complete control so they can plan each interaction down to the last detail, a more management style of gameplay.

If I remember correctly the original Black and White was a much more open system not giving the player a lot of control over their creature, that changed in Black and White 2 to some extent. I’m wondering if they did that because players did not like having so little control over their creatures. Do games, which are just abstractions of real life, condition players to want abstracted, controllable information? Why is it so necessary to try and bring realness into games by hiding information? These bring up a lot of questions concerning emergent behavior and storytelling too.

I don’t really have answers to these questions, I’m writing this to get my thoughts down and to move forward with them. But I do think it would be an interesting study to look at how players react and play differently when faced with varying levels of AI transparency.

I’m probably late to the game when noticing these things but I can’t help but think, I like Office 2007’s UI. At least from a graphic standpoint. The use of fades and transparency just make moving my mouse around the menus a fun activity (I am amused very quickly). Everything is instantly gratifying and visual.

It’s interesting because I went to a talk by Jasper Juul last week and he mentioned this very same design principle. He looked at casual games and how the ones that do well tend to have a “Juiciness” factor, or excessive positive feedback. It makes sense, hardware capabilities do not limit developers any more from using a lot of animations and what we would have called “frivolous” UI details just a decade ago. I’m still waiting for my fully animated intelligent OS desktop though.

Another point about Office 2007, when I first started using it of course it was strange. More graphics, no main roll out menus, and some tools had moved. But after only a few months of using it I’ve gotten used to it. The other day I was trying to instruct someone, who was using Office 2003, where to find some ‘track changes’ feature and since in Office 2007 it’s just right in front of me when I have the review pane open I couldn’t tell them how to find the feature.

It reminds me of the studies on how users get used to an old interface and don’t want to move to a new one, or how it’s in a developers best interest to keep a ‘tried and true’ UI scheme since users have become experts with that layout. Maybe I am used to diving into new interfaces so this ‘legacy’ effect didn’t hold true for me. Or maybe Office 2007’s interface is just better even if it is only the graphics .