Yes, I'm afraid I agree with all of this. Despite taking hundreds of pounds from users, I just can't see what the Select project has really done to take RISC OS forward. I'm sorry if this offends people, but having recently switched from the A9's OS to 4.03 on VRPC, there are only a couple of things I notice. There may be stability fixes, and some of the bundled apps have nice updates, but the fundamentals haven't been fixed. If I had been a Select subscriber for the life of the project, I'd be livid (and poor).
I hope that ROOL does at least offer some hope. Despite the forums being quiet, I know there is work going on (I'm involved with one project in a minor way). If at some point in 2008, it proved possible to build the entire OS, with some nice extra features, then at least it would offer a way to make a new RISC OS machine cheaper. For example, could VRPC sell for a lower price, if they didn't have to licence RISC OS Adjust? Would the politics allow it?
As for ARM-based platforms, it's so disappointing that Castle, having done the really hard work of getting RISC OS to run on non-IOMD and VIDC hardware, now seems unwilling or unable to produce an Iyonix II with a faster CPU. I realise it wouldn't catch-up with Intel, but I would imagine many Acorn-nuts would pay for a relatively fast ARM-based solution. My impression is that plenty more juice could have been squeezed out of the graphics cards on the Iyonix, if only the development time were available. Now it seems like the potential of the platform is destined to languish unfulfilled.