Agreed. Bedrock is interesting, and it's fine to interpret this thread anyway anyone wants. However my purpose in starting it was to get a feel for how many don't mind changing very basic structures of Linux and how many recoil against it. On one level I embrace the freedom inherent in Linux to do whatever one wants with it, but I will likely never go there.
GoBo is IMHO yet another attempt to make deep and fundamental changes to the structure of Linux without much return on investment. Maybe such small steps evolve into benefits we can't see yet, but if so, that's for others, not me. I'm perfectly happy with Linux that stays as close as possible to Unix ways. If others want to see it evolve into a sort of Free Windows that's fine as long as that doesn't isolate and minimize what is available for vanilla.
Major changes like this are a sort of double-edged sword. Example - no one can reasonably deny that Ubuntu has had the very positive effect of bringing large numbers of new users to Linux increasing interest in use and development, good for us all. At the very same time it has also created a situation where it is viewed by many as some de facto standard, making it somewhat more difficult for other distros. Example - LinuxSteam was basically written for Ubuntu 12. Thankfully Steam handles "non-supported" distros reasonably well with system checks and updates but still must create a set of libraries for Ubuntu 32 and Ubuntu 64 on whatever distro you install Steam to do so. It isn't a step too far to stop bothering to accommodate once the numbers get big enough.
Systemd has similar deep changes threatening to cause more vanilla distros to redesign (such as with the udev issue) to stay vanilla. It is certainly interesting (and maybe a little comforting) that Ubuntu not only "plays" but "gets played" in return, (since they had to essentially abandon Upstart and embrace systemd) but nevertheless there is some risk that some distro will become so entrenched that developers will write for it and leave more traditionally *Nix varieties out in the cold. The odds of this may not be very high with so many players, but the stakes could be very high. In the marketplace it isn't always The Best that thrives. Often it's just the lowest common denominator that wins out.
What constitutes deep changes can also be somewhat subjective. Example - whether your vehicle of choice has carburetion or fuel injection, while an important difference requires no change in how you drive. Oddly, taking the shifter off the floor and putting it on the dash, the steering column, or a paddle just off the steering wheel isn't a big adjustment while perhaps changing the orientation of brake and accelerator might. Putting the driver in the backseat certainly seems like it would be a major adjustment. Having an auto-driver with no manual override most certainly would for most be a step too far, beyond mere taxi service.
The next couple years seems like they will see the biggest changes in Linux since it's inception, maybe bigger than all that have gone before combined. I'm just not in favor of "throwing out the baby with the bathwater" and watch such trends carefully.