[oi-dev] What comes next?

Nick Zivkovic zivkovic.nick at gmail.com
Tue Sep 4 14:04:52 UTC 2012


On Tue, Sep 4, 2012 at 8:19 AM, ken mays <maybird1776 at yahoo.com> wrote:
>
>
> The desktop for OpenSolaris is not an issue nor major desktop apps. Most of the major apps were ported and some specs live in SFE.
>
> You can look at distros like Shillix or even Solaris 11. Do you need Xorg 7.7 or Xfce 4.10? What specific hardware do you have and what does not work?

Having out dated software will only make people continue to see
Illumos as a thing of the past, no matter how amazing our kernel is.

>
> OpenSolaris distro was always a CORE OS distro in which most bloatware apps was moved to IPS. So X11 is the main thing to support besides parts of JDS.
> ------------------------------
> On Tue, Sep 4, 2012 4:38 AM EDT Damian Wojslaw wrote:
>
>>Hello
>>
>>So, all you OpenIndiana developers, whom I admire, what comes next? :)
>>I've seen few people declaring they would like to help, I think packaging was mostly mentioned.
>>Please, let me ask few questions.
>>
>>1. Is this link still valid? http://wiki.openindiana.org/oi/IPS+Knowledge
>>2. What would those people need to do, so start maintaining a package? Is there a process,
>>like in Debian, a person has to go through before they are given package maintaining duty?
>>3. Is the development sof OI till going at all?
>>
>>4. What you, OI developers, would want from this distribution? What your priorities are?

Below are things that I intend to work and code on. These were all
problems in OI-147. I just upgraded to OI-151a, and some these may
have been fixed. As soon as a scrub completes, I will reboot the
machine, get networking running and do 2, 1, 7, 5, 6, 3, 4, in that
order.

1: Making drivers for wifi; drivers for usb3 (as soon as I get HW);
improve usb-stack (major source of anguish and bugs).

2: Making NGZ appliances that users can use to get started on things
like building illumos. Or other projects. NGZ appliances can be
transported via USB device (because some machines may not be networked
[no wifi drivers, etc]).

3: Delivering app-bundles to the GZ using zfs datasets. Same rationale
as above, but some software (i.e. drivers) will only work on GZ.

4: Enhance IPS to utilize the above two methods (the more choices we
have, the better).

5: Making improvements to IPS. IPS already has a large catalog of
software. It would be difficult to justify making a new pkg-system
from scratch, because repackaging everything would outweigh any
superiority of a new pkg system. Improvements include performance
improvements, bug fixes to IPS, better error messages, and bug fixes
to existing packages, that fail to install.

6: Modernize compiler support. Illumos still sucks at compiling modern
C++ code, even though some major commands/apps are in C++ (nmap,
Chrome, firefox). I am thinking of bringing Clang and GCC up to date.
This is mostly for userland, as the Illumos kernel already compiles
with an older version of GCC.

7: Port i3 and dwm, and add them to IPS-repos. Same for cmake
(surprising how many projects use it).

>>
>>
>>I'd like to reiterate what I've said many times in the past, when OpenSolaris was still a live distribution.
>>If illumos wants to become widespread, it needs desktop distribution, a usable one, so that people can use it,
>>play with it and learn it. This is how Linux crept into datacenters, because people like me installed it as their
>>hobby desktop, to look at X, to look at FVWM, to use pine and share modem connection at home. And then,
>>it was natural to start exim and apache on it. And then people like me became admins, CIOs and CTOs and
>>started to use what they knew and liked. And I'd love illumos to become the system people can install as their
>>hobby desktop.

Agreed. Hence why need to support the latest or near-latest popular
unix-apps. For example firefox is still at 3.X, and we don't have
chrome. Drivers are not getting made. My thinkpad has no wifi. NGZs
are not being utilized to their full potential. I am going to try to
fix as much of this as possible.

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



More information about the oi-dev mailing list