Re: Some Serious 1.3.90 Issues, Again.

Andrew E. Mileski (aem@nic.ott.hookup.net)
Fri, 19 Apr 1996 15:19:43 -0400 (EDT)


> On Thu, 18 Apr 1996, Andrew E. Mileski wrote:
> > Why do people expect consistency with _EXPERIMENTAL_ kernels?
> > It is akin to expecting politicians to keep election promises.
>
> NO!
>
> You _should_ run the experimental kernels, as much as possible. And
> you _should_ expect it to work correctly and consistently. The experimental
> doesn't mean "may break - don't test". It means "may break - complain to us
> if it does, but don't expect us to feel sorry for you".

I _NEVER_ wrote anything remotely like "may break - don't test it",
so please don't attribute that statement to me.

I clearly stated that Linux kernels are not consistent, but especially
the experimental ones.

Just because kernel version X works, doesn't mean X + N will work.
Surely you don't dispute the logic of this?

At the very most I implied, "may break - don't use on a heart/lung
machine" (this is meant to be slightly humourous _TOO_).

> So Simon is doing exactly the right thing.. Don't flame people for
> reporting bugs.

I _NEVER_ wrote that Simon was doing "a bad thing". I wrote that
it did not make sense to expect a bug-free kernel, in an experimental
series of kernels, on the sole basis of the previous kernel working.

Linus, that was _FAR_ from being a flame. In fact, I attempted to
soften it with humour. I didn't have a stupid smiley, because
I thought it was obvious. Seems you have honest politicians in
Finland :-) (That was a joke _ALSO_ in case anyone missed it).

An intentional public berating to humiliate someone is a flame,
which is closer to what _YOU_ just wrote IMHO.

> As to why Simon sees the problems, I don't know yet. Duh.
>
> Linus

I have faith it will be solved. Maybe not in the next version, but
eventually. But I don't think even you can GUARANTEE every bug won't
come back, as the kernel is quite complex with subtle relationships
that even you, myself, and thousands of others sometimes overlook.

-- Andrew E. Mileski --
mailto:aem@ott.hookup.net
http://www.redhat.com/~aem/