Features Download
From: Sia Lang <silverlanguage-Re5JQEeQqe8AvxtiuMwx3w <at> public.gmane.org>
Subject: Re: What's the status of Haiku?
Newsgroups: gmane.os.haiku.devel
Date: Monday 18th August 2014 15:02:03 UTC (over 3 years ago)
On Mon, Aug 18, 2014 at 4:38 PM, Augustin Cavalier


> On Mon, Aug 18, 2014 at 10:23 AM, Sia Lang

> wrote:
>> I absolutely agree that these guys are fantastic and do impressive work,
>> although in a misguided direction in my opinion.
> What direction would that be?

A direction involving an existing kernel and set of drivers so my GPUs and
obscure devices can be used.

> Adrien's main argument (Linux's lack of unification and need for manual
> tuning) stands. I've never had to personally configure the Linux kernel
> I've had great compatibility, but the few times Linux crashed on me it's
> been no easy feat to get it unstuck. Anything that uses the Linux or BSD
> kernels will be stuck with this. Hence why we're writing our own kernel
> designing it for the users.

A BeOS layer would obviously be coupled with a finely-tuned, ready-to-go
configuration. Ubuntu and Android "distros" don't generally get stuck on
users (which are in the billions.)

Anyway, I'm not really rooting for Linux per se. I'm rooting for an
existing kernel with great driver, toolchain and library support. There's a
few options.

Also, I think you don't get how complex this stuff really is. You won't
> achieve a rewrite of the kits on top of Linux any faster than we'll reach
> 1.0.

Well, that's exactly what I'm trying to figure out - how far away is 1.0.

I'm a long-time OS developer and do actually understand the complexity, and
I'm one of the few who did commercial work on BeOS before it disappeared
from the public eye. I basically did close to three kits in two months. Not
to brag, but this isn't all that hard when you have a rock-solid and
comprehensive kernel with a massive amount of ready-to-use libraries
available. Really.

> And we're not *that* far off: as soon as we can get HaikuPorts
> we should be able to release Alpha5. And from there it's mostly bugfixes
> before beta1 and then release candidates. My estimate is 4-6 months for
> alpha5, 6mo-1year after that for beta1, and then another 4-6 months after
> that for R1.

So 1.5 years. Real numbers!

This is far enough away for me to continue on my project. I'll put up a
github repo when the source base is in good shape in case anyone wants to
chip in.

> If there are any issues that prevent you from using Haiku full time,
> please let us know and we'll do what we can to fix them. If you've been
> able to get somewhat far with a BeAPI reimplemenation on Linux, you
> be able to help develop Haiku too :)

Thanks for the offer, but I have no interest in working on an approach I
don't believe in.

CD: 3ms