OS choice

I have a direct bearing on the development of operating systems, so I can do some analysis of existing solutions on the market. The market of operating systems is full, it is filled to the brim. What? There are many solutions for different domains. Now, I'm faced with a choice of OS for the next project and I can't say that among the existing (Jari OS is a good platform, but aren't ready, and as a woman requires a lot of time) I can find a suitable, I considered the decision on free software, or are not expensive ones - I don't found suitable one. I will not describe my criteria, and I don't want to dilute the flame of useless discussions on an eternal themes.
My experience in the architecture design and development allows me to say - the development of OS is interesting at the beginning and in the middle is a boring job, it's one of the causes of OS market saturation product is covered only part of the requirements and products that are not so flexible.
I came to the conclusion that I can easily take the pieces of the finished OS components and quickly make a monolith in which the applications operate in the kernel mode - for the first time to write applications I need is a situable solution, and then choose right solution(if Jari OS is ready, I will take it).
I'm talking about the fact that I can't understand the meaning of many projects. These are two extremes as ever. The first - the development of general purpose operating systems, I believe that it is a useless idea. Second - the development of narrow-purpose operating system, the appointment is so narrow that except of the author of the project, no one can use it. I don't understand why nobody wants to create a platform for the OSes, flexible so that the possession of the editor (emacs for example), knowledge of the OS design and implementation you can get what you want.
The development of such platform, will not take a lot of time, if we have the necessary number of developers (1-3 years for production use).
I started to do this (since I've understood that multi-purpose OS isn't required now), and continues as the forces I have, but I do not have so much time and resources (just a few people).
I'm disappointed completely and forever, no - not in my projects and local fails, but in humans and OS market.