On Mon, Apr 10, 2000 at 01:14:42PM -0700, Gerald Gutierrez wrote:
> >At full functionality, there's nothing stoping any sysadmin from using
> >the Hurd in *any* way they see fit, from RealTime applications to
> >security, etc.
>
> Hard (and soft, depending on your definition) real time operation of the
> Hurd is impossible; there is no way to bound times and there is no
> determinism. Fortunately, many users say "real time" when they really mean
> "fast enough", which the Hurd can provide.
RT isn't an area that I have looked into much, but I beleive that Posix
supports an RT set, and that there's an RTMach kernel that takes care of
the time requirements when requested to do so.
> > When the Hurd can reasonably be seen as the best answer
> >to the question at hand, it will gain in popularity - and I don't think
> >that will be far off.
>
> Again, this is only amongst the technically inclined. "It can do this
> really cool under-the-hood thing." A less technical audience would not
> care. They would ask "for my uses, how is this better than Linux?" and
> there would be no clear, uniform answer.
No, but one of the answers might be that the technical folks can support
it better. I also beleive that the potential development speed is higher
once enough documentation is available to let programmers begin
developing faster.
> I realize that most people in this list are hackers with a CS or EE (i.e.
> technical) background, but there is a need to go beyond the
> technically-oriented thinking and ask how the project will benefit other
> people ... your wife, your kids, your church, the guy next door. Without
> support from them, I fail to see how this project can become self
> sustaining in the long run. Think about it; if half the kernel hackers
> quit, would there be others to take their place right now? Would the above
> mentioned people care?
But the same could be said about any free software project. If Linus
Torvalds, Alan Cox, and a few others had the building fall down around
them at a trade show could Linux continue succesfully? Not in the form
that it's in. If the same were to happen to Thomas, Roland and Mark we'd
have some serious troubles, but all the code continues to be owned by the
Free Software Foundation. rms would step in and declare a new project
leader, and it could continue.
But the starting value *is* under the hood. I wouldn't put my belle, my
mother or anyone at the console of a Hurd box right now, but we've
already demonstrated some amazing resiliency(sp?) that cannot be
discounted. There have been instances of X running, and will be again
shortly. With a graphical system we have access to all of the Gnome and
KDE suites minus some driver support. When we get there we'll be able to
continue making the Hurd do whatever's appropriate for us at the time and
I don't beleive that making any serious plans now will help us get to
that stage any sooner.
--
There is no sin except stupidity.
- Oscar Wilde
