[OpenIndiana-discuss] Making OI friendly to new users

[Andreas Wacknitz]

Am 12.08.25 um 23:48 schrieb Atiq Rahman:
> Hello,
> I’ve been following OI for some time, and I’m concerned about a pattern
> that has been slowing growth: new users frequently encounter installation
> or boot issues and leave frustrated. I see this repeatedly in the forums,
> often without resolution. This is a major barrier not only to adoption but
> to community contribution. If we want more contributors, we first need more
> users and for that, the first experience must be smooth.
>
> At present, there are several critical blockers:
>
> *1. EFI Installation & Custom Partitioning*:
>    There is major confusion around custom partitioning. Most advice
> discourages it in favor of using the whole disk which is unrealistic.
> Nobody is going to buy a separate external drive for every OS they test. At
> present, folks are able to install many distros using the internal disk.
> Better EFI support in OI should allow safe custom partitioning without
> restrictions.
You are right. But that needs someone who is able and willing to enhance 
OI here.
At the moment there doesn't seem to be anyone.
For me it's sufficient to use whole disks for OI (or even whole 
computers) because
you can get these for little money in the used markets. My personal 
recommendation
is to look for an older workstation (eg. Dell Precision 3630) with an 
NVIDIA Quadro graphics card.

>
> *2. Missing Core Drivers:*
>
>     - Many common Ethernet/LAN drivers are absent.
>     - Common wireless drivers are also missing.
>     - Graphics drivers are broken or nonexistent for much of today’s and
>     yesterday's hardware.
If someone is able to write such drivers from scratch (or is able to 
port existing drivers from other operating systems)
it is very likely that this person wouldn't do that work for free but 
expects (rightfully) to get paid for it.
Most likely such a person would end as a kernel developer (not 
necessarily for illumos based distributions but more likeky for any of 
the major operating systems) as such expertise is rare.
As OpenIndiana is not backed up by any company it's our best bet to 
create a foundation that is able to raise some money
for driver development or any other missing parts that nobody cares to 
write for free.

The best supported graphics hardware on OI is from NVIDIA because they 
release drivers for Solaris 10 and thus most likely work also on 
OpenIndiana. My workstations all have such graphics cards. I have 
successfully used Quadro 600, K600, K620, P600, P620, RTX 2000, RTX 
4000, and RTX 5000 (and probably more I don't remember anymore).

>
> These are basic requirements for installation and first boot. Without them,
> users can’t even get online to troubleshoot or fetch updates.
>
> These issues don’t just frustrate individual users, they shrink our
> potential contributor base before it even forms. Every failed installation
> attempt is a lost opportunity for feedback, testing, and future development
> help.
My experience is that someone who is really interested in OpenIndiana 
finds a way to install it and contribute to it.
It's just a very small number (single digit number depending on how you 
count them).
 From time to time someone new pops up and contributes. As this happens 
rarely the number of active maintainers don't really change because we 
also loose people because their priorities and interests changed.
Most actual contributors have a limited area of interest they care for. 
So we have many areas where we are far behind everybody else.
> I am deeply frustrated. At times, it feels like the stagnation in the OI
> Leadership (or Association whoever leading it) is slowly killing the
> project instead of helping it thrive.
>
> I believe OI can still grow, but it will require prioritizing these
> entry-level fixes. Making installation seamless for new users should be
> seen as an investment in the project’s long-term health. I’d be happy to
> help test or provide feedback on any efforts aimed at improving these.
I am using OI as my secondary operating system because I haven't found 
anything that better suits my interest or needs.
Whatever I have tried in the past has its own pros and cons and OI is 
the best compromise for me. I have "ported" quite some software that I 
am (or was) interested in like ccl (Clozure Common Lisp) and swi-prolog. 
I even think that swi-prolog is better supported on OI than on FreeBSD 
(try to run swipl-win or download external libraries ;-) ).

Regards
Andreas