Ella has been playing the Dora The Explorer: Animal Adventures game for a few months now. The game is supposedly for ages 3 and up, but Ella has pretty much mastered all of the games at the highest level (and she’ll be three on Halloween). If it sounds like I’m bragging, I apologize… but I am.

The one thing Ella hadn’t been doing is clicking on the items herself. She would point and I would click. Now I’m starting to see where the age minimum comes in. Ella has started pointing and clicking by herself with the mouse. However, the software developers made sure to make it as hard as possible for a small child to navigate to and click on the button.

Take the example below. The area that is clickable is so sensitive it is ridiculous. Also, if the “star” cursor is on the corner of the eyes they are trying to click on, it won’t trigger. The middle of the star has to be within the clickable area. Ridiculous. Talk about a guaranteed way to frustrate a little kid. She meticulously tries to center it until it lights up. Then about have the time when she goes to click, she nudges it a bit and loses it again.

It is pretty obvious to me after sitting with this game that—while it is cute—it was never tested with little kids. If so, the hovering and clicking frustration would have been immediately apparent.

So, the lesson is… don’t just test. Test with your actual user base.

I just upped a post over at the BatchBlog called “Step Outside the Code: Be Your Own User.” Let me tease it a bit, then I’ll elaborate a bit more here about the process I followed.

As soon as I joined BatchBlue in May, I was thrown into the fire—we had a beta to get out, after all. There were lots of bugs to be fixed, display issues to be resolved, cranky browsers to accommodate, and architectures to learn. A lot of things initially impressed me about BatchBook, but it seemed like all I was doing was fixing issues. It becomes easy to lose a sense of the whole when you’re arm wrestling with tiny pieces of code day after day.

Then, a wonderful thing (on so many levels) happened—I took a couple weeks off to welcome our second child. For two weeks, I didn’t fix any bugs… I fixed makeshift dinners. I didn’t clean up code… I cleaned up belly button regions. I didn’t change class names, I changed diapers. Yet, I still used BatchBook. But I finally used it as a real live user.

It was very interesting finally jumping into BatchBook head first as a user. I had a lot of “hey, this works but if only it did this” moments. I jotted those down in a notebook. I kept the notebook open for the entire two weeks. Nolan has a knack for being wide awake between 11 pm and 2 am and requiring me to pace around the house with him, listening to music.

I brainstormed. A lot. Every time I passed the notebook, I would jot down the idea. I kept building my vision of how this thing should work. Then, the Sunday before I returned, I cranked out a bunch of mockups based on what I was thinking about. And hell if they weren’t some of the best work I’ve ever produced.

So, why is it that when I’m at my most tired, I’m feeling my most invigorated with my work? It could be a lot of things. I think one main reason is that since I didn’t design the first iteration of BatchBook, I felt very overwhelmed by the amount of areas I felted needed to be changed. Well, I’ve now mocked up a totally revamped Contacts section, which is really the backbone of the app. So, maybe it is because I can finally feel the application getting a sense of… me.

I honestly think another reason is that I just backed off of the development for two weeks and just used the thing. I organized my own contacts, made edits that weren’t just dummy data, and actually started recording new information in it as it came, leaving my other applications behind. I like the app, but I love what the app can be. And I think I’m just pumped because I finally have an idea of what that is.

Aza Raskin, Humanized

Son of Jef Raskin, “Father of the Macintosh”

Design towards human frailties and interface will work well.

Cognetics

• The ergonomics of the brain
• Tells you you can’t press two buttons ten feet apart at once
• Can’t hold 7 (¬±2) in your short term meory at once
• What people are capable of

Forgotten tools of Interface Design

• GOMS Modeling: Lets you take an interface and figure out how long it will take an average user to use it, a priori
  • Most UI designers don’t know it exists
  • When paper prototyping, can be used
  • Different than how easy, how efficient
• Information Efficiency (Developed by Jef)
  • GOMs doesn’t say what the best interface is
  • HOw much information do you need to put into a system vs. How much is absolutely minimally needed.
    • How efficeint you are being
  • Humanized blog, how to calculate efficiency

The Death of the Desktop is near

• What’s an interface.
  • The way that you accomplish tasks with a product — what you do and how it responds — that’s the interface
• To the user, the interface is the product
  • If it can’t be used, it can’t be used.
• Shovel Analagy
  • Cutting piece gets all the R&D
  • Interface is wooden handle
• We spend an inordinate amount of time fiddling with our computer, when we should be getting tasks done
• Keep simple things simple
  • Analog vs. Digital watch
  • • Digital is more complex (cites Timex instructions)
  • Write a manual when you design an interface
• Other simple things that aren’t
  • Cell phones
  • Rounded corners
  • Adding an entry to Google Calendar
    • When an interface is simple, you usually don’t notice that there is an interface at all
  • Making a simple website mockup
• The Problem: Applications (Like Isolated Cities)
  • Word/Photoshop/Mathmatica
    • Each does the same things, but has a specialty in one
    • You have to go back & forth
    • Programmers are re-programming
    • This is why programs are bigger and bigger
    • And our hard drives have to be bigger and bigger
    • 7 different implementations of spell check with seven copies of the English language
  • Applications always try to step on each others toes, they want to horde their functionality

So, what does an interface do?

• Create content
• Navigate content
• Select content
• Transform content
• Question: What about sharing content? (He’ll have to come back to that)

Raskin’s Rules of Interfaces

• An interface shall not harm your conent or, through inaction, allow your content to some to harm.
  • Why does the computer not save by default?
• An interface shall not waste your time or require you to do more work than is strictly necessary.
  • Copying from one app to another
• An interface shall not allow itself to get into a state where it cannot manipulate the content.

Content is everything

• First and foremost
• Everything you do has to do with the content

What will doom the desktop?

• What work do you really get done on the desktop?
• Starting with a flawed metaphor—paper gets lost on your desk
• When you have a bigger monitor, productivity goes way up because your windows are not hiding each other

What does the desktop do?

• Lets you get the computer into a state where you can enter content
• Lets you categorize your content
• Lets you navigate your content

There are better, faster, more humane ways (The web is a tresure-trove of examples)

Language has untapped power (“Thhis is the futture”.)

• 30Boxes, Google Calendar, leverages power of language
• Type what you want. Have two numbers? Type “calculate”.

Again, what does the desktop do?

• Lets you get the computer into a state where you can enter content
  • Command Line interfaces
  • URL= Command Line that everyone can and does use
    • Imagine a dropdown of all 100,000,000 web sites
  • Spotlight/Google Desktop search
    • Another command line interface
  • Humanized’s Enso Launcher
• Lets you categorize your content
  • Categorization
    • Tags
      • Doesn’t force a hierarchy on to the user
      • Not an either/or choice
      • Desktop needs to learn this lesson
    • Search
      • If you have a good search, you really don’t need to categorize very much
      • Nobody uses the Yahoo categories anymore, searching works better
    • = The death of forced hierarhy
• Lets you navigate your content
  • Navigating: Don’t really know what you’re looking for
  • Let content be content
    • GUI touted as having direct manipulation
    • You’re moving around a graphical representation of that item (when moving a document)
  • Let search be search
  • Let 2D content be 2D content
    • Windows being 3D causes problems
    • Expos√© for Mac helps by making it all 2D again
    • Microsoft’s Rolodex goes the other way, making you sift through one at a time
  • Let the user’s structure be
    • Don’t force them into thinking about the content the way the developer thought of it

Zoom interface

• Everything on 2D plane, user zooms in.
• With smaller window: With no changes in interface, everything still works
• Aza’s Desktop of the Future

Maze/Web

• A maze is a series of interconnected rooms connected by doors — when you go in one, you can’t see where you came from
• The web is a series of interconnected pages connected by hyperlinks — when you visit one, you can’t see where you came from
• Snap is the first step towards fixing this
  • Show you where you are going before you go there

The Desktop is Doomed

• Why the Stagnation?
  • The Toolkit Straightjacket
• We have a unique opportunity
  • We must not return to the desktop
• How can we overcome this?
  • The solution:
    • Services
      • Make mashups possible
    • Universal Access interface
      • Type what you want to get and it comes up
    • • Current solutions are not scalable
      • URL bar is limited
      • Need a fast, semantic method

An example: Enso

• Harnesses power of language to access what you need
• Similar to QuickSilver on the Mac

Design The Big Picture

• Challenge: Can you think of other solutions?

Take Home Message

• Content
• Language
• Services
• Unification

Found these notes from another of Aza’s talks.

What a fantastic talk. I will be thinking about this and covering it more on The CogBlog.

Update: I was running low on battery during the Q&A session, but there was a lot of great stuff discussed. Chris Messina brought up how Zoom didn’t account for the social aspect of interface design. Aza said that developing the content comes first, sharing it is an add-on. Chris disagreed, saying that technology starts with connecting people first. I tend to think that both are correct, but they are different problems. Aza’s approach does need to account for social sharing, but we do need establish just what we are sharing, too. More to think about… I’ll write more at some point.

Well, I guess I’d have five cents. Brian Oberkirch recently interviewed Dan Saffer of Adaptive Path for Weblogs Work. (Link soup?)

Dan talks about Interaction design not just how it relates to software, but to Disney rides, coffee shops, and other everyday situations. Dan’s book, Designing for Interaction, hits shelves this August.

Worth noting is that Dan & Brian were both on the insanely great “Let Go, Jump In” panel at WebVisions last week.

The interview is here.