aokoye wrote:
I wouldn't be so sure about that. I did the math and, from the numbers that Ethnologue has, the languages you've chosen make up around 3.928 billion first language speakers. This is actually very generous because I'm assuming you're not looking at most of the Chinese dialects nor most of the Arabic dialects.
If we go with Wikipedia's tally of L1 and L2 speakers (most of which also comes from a Ethnologue) then we get 5.628 billion speakers of the languages that you're using. That still gives us at least 1.872 billion people (so more than the population of China) who don't speak any of the languages that you've chosen. That's nearly 25% of the world's population (if we go with the idea that there are around 7.5 billion people living in the world right now).
edit: so yes - "most" of the population speaks those languages but it isn't hard by any means to find someone who doesn't. An easy example of this is the issues that hospitals have in California (among other states) have in finding interpreters for people who they think speak spanish but in reality only speak a native Mexican language.
Covering 75% of the world's speakers would be an achievement for a language that only has over 500 roots, considering that the 16 languages in Atlas have pretty much an equal participation in the vocabulary (I would not say that the language is represented if only two or three words were coming from that language).
As I said, it is difficult to establish a system in order to create the vocabulary, but as I said I think the one used by Atlas is fair. It could have been done in other ways but I don't think it would have been a better solution, just a different one.