The future of NetBSD

[Daniel Ouellet]

On Thu, Aug 31, 2006 at 08:03:04PM -0300, Marc G. Fournier wrote:
> On Thu, 31 Aug 2006, Miod Vallat wrote:

>>> I'd rather have Adaptec provide a source code driver for their cards
>>> directly, then have Scott Long have to fight with unavailability of
>>> documentation itself ... if the driver works, what do we need
>>> documentation for?
>> To fix the driver.
> If the vendor is supporting the driver, and working with the community, 
> then one would hope that they would also fix the driver as bug reports 
> come in about it ...

Dream and never going to happen. History proved it too many times over.

Case 1: IBM sales support contract for OpenSSH, makes tones of $$$ from 
it. Get support calls for bugs and send it to OpenSSH instead and tell 
it's clients that they can't do anything for them. But at the same time, 
they claim to support the open source... They cash in and never gives 
back. But they have the FULL source code and that's not even a driver to 
a black box that they don't have documentations for. If they said they 
can't do it for that fully open source code, how do you expect others to 
for example support IBM hardware BLOB drivers, even with the source 
code, without docs? The said they can't, but they sure have the men 
power don't they? So... NOTHING replace docs, nothing at all.

Case 2: Linux got huge market share, at the time RedHat got public and a 
few others got rich at it. IBM saw this as an opportunity and claim they 
support the open source, offer accounts on their hardwares to developer 
more of Linux by the community so they they can sale more hardware at no 
cost to them. They also could claim to support it and be the biggest 
supporter of it. Yeap, they go what, last year announce 1 billions of US 
$ of income from support contract on Linux. Well that's a good amount of 
money, so why don't they contribute back then. But it pays to keep it 
not to good doesn't it? But hey, they couldn't contribute to OpenSSH in 
$$$ for all that they got in no?

Proof 1 case 3: The same IBM used to have OS2 and it was a hell of a lot 
more stable that Windows ever been. Even Microsoft software ran better 
on it then on Windows. But it didn't pay as IBM didn't get support 
calls. So, they let it die, without telling anyone obviously. It was 
secure and stable for it's time anyway. Ever tried to by direct from IBM 
a server with OS2 pre install. I did and couldn't, but sure could get 
Windows on it with support contract as well. Doesn't pay to be too 
secure or stable, customers don't upgrade, or call for support and give 
you $$$.

So, it's all the same thing in the end and the history proved it many 
times over. Just look back and you will see it too.

Adaptec, IBM, and the like. They don't really want you to be so happy, 
if you were, they wouldn't sale you more stuff.

But all this is also cause by the same community that accept this in the 
first place. They accept BLOB, so they get them and then when they 
upgrade their OS and it doesn't work, they complain and change their 
hardware with new BLOB and the circle start all over.

Until all the *BSD act as one voice and demand documentations, this will 
not change. I wish I could asked the Linux to do the same, but that's 
even a bigger dream I think, but I hope I am wrong.

So, no, they will not work with the community, it's been proven, they 
will not support them either, proven too as well, even in case of 
service contract with all n their hands to do so. And as for giving up 
the documentations because it is working, well I guess no one is 
interested to be sure it is safe, secure, stable, efficient, but trust 
others to do so.

In short let the direction be taken by others and in the end complain 
that some project are going down. The lack of direction make it happen 
and the acceptance of other BLOB codes without questions accelerate it too.

Instead of having one *KERNEL*, may be it could start by one voice and 
demand documentations, freely and simply that.

That would already be a HUGE step forward and that would help ALL the 
*BSD including the *Linux as well.

But sadly, looks like only one *BSD project see it clearly and stick to it.

It's a sad status of affairs I tell you and I sure hope that NetBSD 
comes back with it's own good direction as well. If sadly not, then I am 
sure OpenBSD could use some more selected devs, but that's sure hell is 
not my place to say. There is a head for that and no one question it 
either. That's direction

Wish you the best never the less.

Regards,

Daniel