Re: Progress on a "spiritual" derivation of GoboLinux

IMHO Gobo hasn't withstood 14 revisions. It made a splash with it's
filesystem layout in initial years, but has went progressively downhill
since. This doesn't mean that the Hierarchy is bad, neither is the way it's
handled bad. In fact, both are very good. Just that I think some major
flaws were made in it's inception that were never (or never able to be?)
fixed.

Two primary ones stick out to me.

The first being that it doesn't have a way to account for different builds
of the same program with the same version. Things like the vendor and
possibly compile flags do need to be taken into account. This would allow
the administrator to have a custom source build of a program running along
a stock build. (This may be partly to blame for why there was there never a
64bit Gobo.)

The second being capitalized folder/file names. The only reason I can
figure that this was ever considered to be a good idea was because the
designer was using zsh. Big fail. Bash is the industry standard, and it
does not auto capitalize. Don't get me wrong, I think zshell is great, but
many of us just want the standard. Basing a system around something that is
not standard is begging for failure. Others may think I'm wrong and that's
ok.

I'm just being honest in this, and I think there are a lot of great things
about Gobo as a file hierarchy and OS. I just also think there are problems
inherent to the core of the system, and they can't be worked out. They just
have to be replaced.

I agree that Ben's analogy to filesystem creation isn't great. (Ext3? :))
But I think Ben makes a great point with "accomplish your goals without
changing the filesystem hierarchy standard". That's something I think makes
a lot of sense.

_______________________________________________
gobolinux-users mailing list
[email protected]
http://lists.gobolinux.org/mailman/listinfo/gobolinux-users