If the X1000 flops, it is likely that it will spell the end of the line for OS4.
I suppose this is why A-eon is a seperate comany, so that Hyperion can live on after A-eon has folded, but for the sake of argument - what will happen to OS3.x and OS4 sources when Hyperion folds?
Two words: Driver support.
Without it, all the "potential" is just talk.
Driver development can keep the few developers busy for years and years. For these small developer groups it only makes sense to support a very limited set of cards. I remember when QSSL was trying to push neutrino to the "geekhood", they failed for several reasons:
* lack of drivers
* closed developing model, "anyone" could not just develop drivers
* way too few developers internally at QSSL to deal with above problem
* way too little hardware internally at QSSL to deal with above problem (I helped the pcmcia ethernet card developer, appearantly I myself had more cards availably than he did)
* some clever person ported gtk to photon and everyone got busy bringing gtk apps to neutrino instead of creating native apps, with very mediocre results. Users quickly realized that there are better platforms for running gtk apps and left.
Thanks to all the fighting, the vast majority of developers left the scene a long, long time ago.
That, and some other reasons
* limitations in the OSes that makes development painful
* limitations in the OSes that makes alot of software irrelevant to implement
* closed source OS with all that brings (for development of subsystems, drivers etc)
* the license model, you cannot do proper testing of your software on different hardware without paying new licenses (30 minutes is not enough)
* the "social contract" developing for these systems implies
