DesignAday

In the Details: Too Simple

I’ve written about simplification many times in the past. Here’s a perfect example of simplifying an interface to a fault.

In Mac OS X Mountain Lion, iCal was subject to several changes. It’s alarms have been integrated with the new Notification Center. Previously, when an alert displayed for an iCal alarm, it included the option to set a snooze period. I might set an alarm to go off 10 minutes before a teleconference, but then set it to snooze for 5 minutes when it goes off. The new alerts, however, only provide a Snooze button. There is no way to specify the length of time, nor does it tell you how long it will snooze. I pressed the Snooze button once just to find out how long it would go. It’s a fifteen-minute snooze. That’s rather long. I doubt I’ll use it very often.

Certainly, calendar alerts have been simplified. The user doesn’t have to select a time period from a menu. It’s a binary decision: close it or snooze. But in this case, simplification is a red herring. It doesn’t save time if I puzzle over its behavior. It’s not easier to use if it isn’t useful.

In the Details: Share

As part of the Mountain Lion refinements and the unification of Mac OS and iOS, Safari received a new “Share” button, just to the left of the address field. It displays a menu of options, including Twitter, Message (the desktop version of iMessage), and Email this Page. Those all make sense, as they are all methods by which a webpage may be shared with other people.

I first discovered this button when I was trying to bookmark a page. There used to be a “+” button that would allow you to add a bookmark, but that was gone. I couldn’t find a bookmark button anywhere, but I noticed the Share button and clicked it. Somewhat unintuitively, the first two options in its menu are Add to Reading List and Add Bookmark. These aren’t methods of sharing at all; they are means by which to save a page to view later. To be perfectly fair, the same icon presents the same menu containing the same options in Mobile Safari, so it is now consistent across platforms. It’s easily learnable, so I don’t consider it to be problematic, and I can’t think of a better label for the button. It’s a sensible combination of similar functions under one UI control—simplification through categorization.

Choices

I’ve played a lot of computer games, but there are a select few that have captured significantly more hours than the rest. Diablo II is one of them, so it’s no surprise that I am currently engrossed in Diablo III. While I am very much enjoying the game, I am questioning the design of the leveling mechanic. Diablo II was famous for introducing skill trees, which are now staples in RPGs. Diablo III does away with skill trees, and further, seems to have removed all decisions from the player. With each level, your character earns either a new skill or an upgrade to an existing skill. There is no choice. 

The UI provides seven action slots to which skills may be assigned. Each slot corresponds to a key or button for activation. One slot is for health potions, so there is no decision to be made there. There is one slot for the left mouse button and one for the right—the primary and secondary attacks. Then there are slots mapping to keys 1 through 4. Regardless of how many primary attacks you have unlocked, you can only have one assigned to the primary attack slot. The same goes for secondary attacks. Keys 1 through 4 are customizable, but limited to tertiary skills. It doesn’t feel as though I can make any decisions that significantly differentiate my character from anyone else’s.

I’ve written before about simplification and clarification of complexity. At first blush, Diablo III seems to have reduced complexity through simplification. As a result, the game seems shallower. When there are less decisions to be made, I feel less vestment in my character. Then again, I’m only in the middle of the first act, so I may feel differently once I’m deeper into the game.

Simplify

My wife needed to make two copies of a DVD that I had created for her from an old VHS tape. She spent some amount of time finding the original video project I had created in iMovie and was trying to figure out what she needed to do to get it into iDVD so that she could burn the discs. I wasn’t home, but she called me to ask for assistance. I knew there was an easier way, but I wasn’t going to try to figure it out over the phone, so I told her to wait until I got home.

1. Right-click on the DVD and select Duplicate.
2. Once the copy is done, insert a new DVD, right-click on the copy, and select Burn to Disc.
3. There’s no step 3.

It wasn’t all that long ago that you required specialized software to do that. I’m quick to point out that simplification is not always a good goal—some things are inherently complex, in which case we should strive for clarity. But, sometimes simplification is the perfect goal.

Simplified to the point of complexity

Since the beginning of DesignAday, I have made a single post per weekday, minus holidays and days I was traveling. Now, I don’t always get my daily post done on the day it’s for. As of this sentence, I am thirteen minutes late for Tuesday’s post. I have always, however, made sure to date the post for the day to which it was intended. Last night, however, I was surprised to find the time and date field missing from Tumblr’s UI. I looked again, and it still wasn’t there. I stared hard at the sidebar where it was supposed to be, but it stubbornly refused to reveal itself.

So, I sent email to Tumblr support asking what happened to the ability to change the date and time of a post. I received the following response:

Hi Jack—

I’m sorry, but that isn’t supported at this time. Our apologies. We will share your desire for this feature with our team.

Thanks
Doug

I don’t understand why such a basic feature would be so casually removed after it’s been in place for over five years. I can appreciate the desire to keep things simple, but is the ability to change the date of a post so costly?

At any rate, Monday’s post is dated Tuesday, and Tuesday’s post is dated Wednesday. I’m sure there will now be some days with two posts, and some, like Monday, will appear to have none. Maybe nobody else will notice or care, but their simplification has become my complication.

Mission Control

Spaces was the multi-desktop feature in Mac OS X Leopard that has now been rolled into Lion’s Mission Control, which also incorporates Dashboard and Expose. I have heard that Spaces wasn’t used by a lot of people. I used it extensively and loved the flexibility of it. I had Spaces set up as a grid of six: three on top and three on bottom. I had certain applications pinned to specific spaces. For example, space number four (bottom-left) always had Mail on the left display and iCal on the right display. Space two (top-middle) contained iTunes on the left and TweetDeck on the right. Some applications, like Safari, were left to open in whatever space I was looking at when they were launched. Then there were other applications, such as Skype, that I set to display in every space. Using the control and arrow keys, I was able to quickly move between spaces, each space being no more than two keystrokes away.

Mission Control isn’t as robust. I assume Apple decided to simplify Spaces in an effort to make it easier for more people to understand. All spaces are now in a single row, rather than a grid, so it takes longer to navigate between them. There are keyboard shortcuts to jump directly to them, as there were in Spaces; I just need to learn them. They didn’t retain the preferences for assigning applications to spaces, and there is no way to make an application present on every space. Mission Control allows me to move windows from one space to another, but unlike Spaces, I don’t have the freedom of dragging a window from any space; only those windows in the current space can be dragged. Furthermore, I can’t drag a window from one monitor to the other monitor in a different space. I must first drag the window to the other monitor in the current space, and then drag it from the current space to another space, requiring two actions instead of one. This all makes the rearrangement of applications and windows a bit tedious, and I find that I’m doing it more often, now that I can’t specify where I want applications to open.

Hopefully, Apple will return some of Space’s functionality in future updates.

Correction: See post for 8/8/11, Desktop Underpinnings

Clarity

A couple days ago, I wrote about Eric Raymond’s Rule of Simplicity. He also puts forth a Rule of Clarity: Clarity is better than cleverness. He applies this rule to the generation of graceful code that is easy to understand, easy to maintain, and less likely to break. But what is the difference between simplicity and clarity?

Simplicity is a lack of complexity. It is easy to make the simple clear. It is difficult to bring clarity to the complex. Design isn’t about making the complex simple—it is making the complex understandable.

Edward Tufte elucidates in his book, The Visual Display of Quantitative Information:

What is to be sought in designs for the display of information is the clear portrayal of complexity. Not the complication of the simple; rather the task of the designer is to give visual access to the subtle and the difficult—that is, the revelation of the complex.

And how do you go about bringing clarity to the complex? Tufte has a very clear answer. “To clarify, add detail.”

Simplicity

Another of Eric Raymond’s rules from his book The Art of Unix Programming, explains that complexity in the code makes software difficult to maintain, troubleshoot, and build upon.

Rule of Simplicity: Design for simplicity; add complexity only where you must.

Similarly, complexity in the UI will make it harder for the user to learn and tricky for the designer to build upon in a consistent fashion as new features are added. The simpler the UI, the more easily a user will be able to comprehend the way it works and form a mental model of it. However, simplicity is not always a realistic goal. There are tasks that are inherently complex, and no benefit is gained by dumbing them down—in fact, this may result in a loss of functionality. There are also experienced users that can work more efficiently with a tool that, while complex, gives them the flexibility to do things the way they need to do them.

There is a balance that must be found between simplicity and complexity. Albert Einstein said it best: “Things should be as simple as possible, but no simpler.”