
"I endorse this message" — Plato
The strategic design of our next app, Languages, continues apace.
In the last post we gained an understanding of our personas’ experience without an app. We now need to imagine how an ideal app could make this experience 100% better, without worrying about bothersome constraints such as money, technology, time, or reality. It’s time to build a Platonic app (Plato, as you’ll recall, thought that all physical things are fail and that their is an ideal, perfect form of all things floating in heaven). Construction materials include aether, fluffy-golden clouds, and joyful thoughts.
It is essential to have a vision of an ideal app so that we know to what our imperfect app should aspire. You have to have a clear understanding of the ideal if the real is going to approximate it.
Here’s how it works. We look at our three experience domains (solo, social, and travel), look at all of the experience nodes in those domains, and then yell out ideas for how an app could improve those nodes 100%. Jeremy then writes these ideas on a sticky and posts them in our “ideal assistance” section beneath each experience node. The aggregate of all these ideas is our fantasy app.

Please pardon the calligraphy and the shaky-cam.
For example, when we examined the “social” domain, we determined that when you’re in a business meeting in Milan and an Italian colleague says “sono molto dispiacentissimo” you want to be able to nod your head and find out what that means without him knowing that you have no idea what he said. You don’t want to look like a fool. So we wrote down “incognito,” indicating that the ideal app could feed you definitions on the sly (of course this sounds impossible; but stick around). We also determined that an ideal app would feed you definitions extremely quickly, because you don’t want to spend five minutes in the street figuring out what “Achtung! Achtung! Ein Auto kommt!” means.
Once we have posted all of the characteristics of our Platonic app on the wall, we go through and draw a big, red x over all features that (1) are beyond our means or (2) would bloat 1.0. We call this step “killing the baby.”

Killing the baby is a traumatic experience for some.
So, of the two feature examples above, we put a big, fat, red x on “incognito” due to technical constraints. But we kept the idea of quick definitions because we felt that it was doable and essential.
What we’re left with is the list of features that will appear in Languages 1.0.








Three years ago Jeremy began blogging about how to design great iPhone apps. Since then he and we have been flailing around, trying to discover the perfect process. After three years of reading, researching, head-scratching, and good old-fashioned trying, we’ve developed the recipe for great iOS design—our secret sauce. We’ve learned many lessons from the 






