Interesting angle. It strikes me, however, that the kinds of things people say to each other face to face might differ substantially from the kinds of things they find useful to say over a network. A target that better matches your goal might be "the set of things people say over the telephone". For example, Uber doesn't really replace asking your friend for a ride - it replaces "phoning a taxi".

I'd also like to note that not all apps are about communication (calculators, note-takers), not all communication is synchronous (flyers, newspaper ads), and not all communication is person-to-person (libraries).

All that being said, I think there's a lot of merit to the approach you're taking. I'd like to see a similar approach taken to the design of computer interfaces, particularly programming languages.

> All that being said, I think there's a lot of merit to the approach you're taking. I'd like to see a similar approach taken to the design of computer interfaces, particularly programming languages.

That's actually what I'm doing. I'm tackling the big picture. It all started with my frustration with the current state of computing, the technology and UI fragmentation. I find the current "application paradigm" very inefficient, and I'm convinced that we can do a lot better.

At first, I approached the issue by trying to build a "database of everything". Then I discovered the semantic web and integrated that to my vision. I then found a lot of problems and limitations to the semantic web, and worked on improving it. I've explored linguistics (especially semantics) and languages like lojban. I've done a lot of research in UX and programming languages. I'm currently working on a new language that will change the way people think and communicate. The language will be mostly computer-assisted, in order to make the limitations of human memory a non-issue. It will be a language like no other (unlike Esperanto or other "natural languages") and will be communicated and consumed through many senses. In most instances, this language will be used like you would use an app (little or no text or vocal input).

To demonstrate and test the flexibility of my solution, I need as many use cases as possible. I started by asking myself what people used technology and computers for, what people used the internet for, what applications are most popular and what applications are used the most. I've now reached the point where I need to understand what people really communicate at any level, either through apps or through face to face.

The promo video published by Google mentions people translate 100 billion words everyday with Google Translate.

According to the video[1] the most translated words in the world are

"Thank you", "How are you" and "I love you"

[1] https://goo.gl/epGctM

I'd say there is a lot of location-based stuff:

"Where are you?"

"How long until you get here?"

"Where should we meet exactly?"

A system that tied into GPS/Google Now/navigation apps/train schedule apps could help giving quick&accurate answers.