The same way you know which C is "middle C".
Alternatively, color the keys differently.
Alternatively, shape the keys differently, or add a distinguishing part to some, like the nubs most computer keyboards put on f and j keys.
> We need some patterns in order for our brains to be oriented to the instrument.
Yes, but they don't need to be messy ones. Consider a guitar: excluding the two thinnest strings, each string is tuned exactly 5 semitones up from the last. Each fret is one half-step up. Guitar players don't get lost!
> You seem to think that making the keyboard more symmetrical, or more uniform, or different somehow, makes it easier to learn.
Yes, and for good reason. It means the student can skip learning 11/12ths of the required muscle memory. That's a very significant, and very measurable gain.
> The field of music and music instrument technology is alive and well and waiting for your invention.
That remains to be seen. In fact, the very article we are discussing shows how difficult it is to make the world-changing introduction of a piano with slightly smaller keys.
> Also, altered dominants and the various minor modes, in addition to the major scale modes I mentioned. Yes that adds up to probably 300 or 400 scales. Which takes a lot of work, regardless of whether you have a standard keyboard or a WWWWWWWWWWWW keyboard.
And then you multiply that 400 by twelve, and you learn just how much muscle memory is required to play the piano.
Since you seem to think this is such an impossible task, note that it has already been accomplished: https://www.youtube.com/watch?v=cK4REjqGc9w
The only way Janko's keyboard is not fundamentally superior is that it isn't widely available.