Until now I though at switching base to trueos-core just like it was a sort of switch to freebsd-current, but I quick realized it is not so simple:
TrueOS base has implemented some improvement:
- libressl as default
- packaged freebsd base
- other
That means that some ports will need options set accordingly to those choices, and we can't simply use trueos-core + ghostbsd-packages, we would need at least the same options set, that's going to require lot of time.
At least for a start I'm inclined to use the TrueOS builders, with its own pro and cons, (I just need to just learn more about trueos before being truly able to make proper decisions).
The thing I don't like about trueos (more precisely about "Continous integration") is that we do not have a "fixed/stable environment, time wise," to be used to build ISOs concurrently, each of us on its own ...
Anyway, may be I have some pre-conceived idea and cannot see the exact benefits of "CI", therefore I would give it a try for a month or so.
If "CI" will work, that would be a good thing, because we will get increased overall testing from the whole TrueOS userbase, and at the same time it will unload us from most of the building workload.
The implication I see is that the ISOs should be always be built from a build-server, not locally, so that we can all use the same ISO when performing pre-release testing.
But, more important, while we might take advantage of trueos build infrastructure, I still want GhostBSD differentiated enough from TrueOS Desktop, and not only by a different DE, we should be able to make our own decisions about nearly all aspects of the OS, at least those things we can manage ourselves.
Your comments will be greatly appreciated.
