Re: Let's do some coding :-)

[olafBuddenhagen]

Hi,

> The problem we are facing here is that every major design decision we
> are making _now_ must be correct down the road in 5-10 years, in fact
> much longer _if_ we are successful.  If we _want_ to be successful.
> This is a huge responsibility, and it requires great care, and a lot
> of studying.

Now come back down to earth. I was sceptical about the Hurd/L4 project
for quite a while now, with the ever increasing "let's throw everything
away and start all over again" notion.

But now, it is getting downright silly. Not only throwing away almost
all existing code, but also all the design work and experience from the
past 15 years.

In your quest for the perfect OS, you are overlooking a very fundamental
point: The computing world changes faster than you are able to create
perfect operating systems. If you come up with a perfect design for
today's requirements, and have it implemented in 25 years or so, you
will realize that nobody wants it anymore, because requirements have
changed.

If there is a design lesson to be learned from the slow progress of the
original Hurd, than it is the simple realization that it was *too
ambitious*. Now the right conclusion from that is certainly not to go
for something even less realistic.

Sure, it's harder to fix stuff once you are popular. But it's way better
than spending your life in a vain attempt to avoid any possible problem
up front, delivering nothing at all.

What you are doing right now looks to me like an attempt to prove the
conclusion of our great friend Eric Raymond: "[...] it was alreasy clear
that HURD had become an exercise in intellectual masturbation [...]"

I hope you are at least enjoying it :-P

-antrik-