[oi-dev] Desktop Illumos Still Matters

[Nick Zivkovic]

> OpenIndiana aims to be a general-purpose traditional distribution usable on server or desktop, not a hypervisor
> (although kvm and qemu packages can be found in the repositories).

I don't see why OpenIndiana can't be _both_. This is software after
all. More choices means more users means potentially more
contributors. I don't see why we should support only one way of
sharing and installing packages. This is orthogonal to the NGZ
appliances I mentioned earlier. Either way, I'll test/implement my
ideas, and  make them publicly available.

On Sun, Sep 2, 2012 at 11:17 AM, Magnus <magnus at yonderway.com> wrote:
>
> On Sep 2, 2012, at 11:18 AM, Alan Coopersmith wrote:
>
>> On 09/ 2/12 07:00 AM, Adam Števko wrote:
>>> IPS is documented in the official IPS Developer Guide located somewhere in the OpenSolaris/Oracle page. I went it through lately and I find it a good source for learning to work with IPS, in general.
>>
>> http://hub.opensolaris.org/bin/download/Project+pkg/files/ipsdevguide.pdf
>
> FWIW, I think this sort of response is a sign of ill health for Illumos. We're still pointing to Oracle for our own documentation. And, as we should know by now, Oracle has no problems withdrawing things from public view with no warning.

Agreed. Oracle is not a friend of Illumos. In fact Oracle is probably
Illumos's biggest nemesis (however inept and impotent they may be).
This documentation can probably be rewritten on the OI/Illumos wikis.
I will do this.

>
> Additionally, I see us debating about putting a lot of work into supporting a small fringe of users (desktop) while nobody is really talking about modernizing hardware support for ubiquitous 4K sector hard disks (and I mean beyond crude hacks).
>
> We're missing the big important stuff.

Well the desktop is important to me and others. So _I_ will personally
put a significant amount of work into things like having a modern
Xorg, USB drivers, WiFi drivers, and so on. That's what open source is
all about. I am not saying that engineers that are preoccupied with
other things that are actually relevant to their companies' business
plans, should do desktop work pro-bono, at no benefit to themselves.
People should scratch their own itches.

I stand by my claim that having Illumos work _adequately_ on the
desktop, can mean fresh talent for the Illumos community. Much how the
Linux desktop helps recruit potential contributors. This is, imho,
very important for Illumos. I see no logical reason why Illumos can't
be an amazing desktop system given that it has best of breed
technologies. It won't be able to inter-operate with proprietary
formats and protocols (skype, MS Office, etc), but that's why KVM is
an excellent technology for a desktop OS. I think community
contributions to desktop technologies should be encouraged by the
Illumos devs even if they have no interest in writing the code
themselves.

I am not saying that Illumos can dethrone dominant desktop OS's; we
can't, nor should we, cater to non-techies. I am saying that it can
grab a fragment of those desktop power-users (those that run linux,
for example, but would like to use something better, because it comes
at no personal cost to them, but can help them significantly).

>
>
> _______________________________________________
> oi-dev mailing list
> oi-dev at openindiana.org
> http://openindiana.org/mailman/listinfo/oi-dev