When Oracle announced it was discontinuing the development of OpenSolaris, there was shock among the free Unix community. OpenSolaris was popular and had a very loyal user-base and good support from developers, internal and community. A fork of OpenSolaris was quickly announced. A fork of the kernel would become what is known as Illumos. And the operating system would become OpenIndiana, which would use the Illumos kernel.
Let’s fast forward to today and we have a complete Illumos based, OpenSolaris forked Unix distribution-OpenIndiana.
What is the problem?
The primary problem that lies within OpenIndiana’s development is the disunity and disconnect between the project’s leaders, developers and users. The philosophy of what OI should be, is there. And the very foundations for the development to produce the perfect template of the philosophy are there. Yet there seems to be an ongoing reluctance to implement a project hierarchy. The project’s founder, Garrett D’Amore, has done a fantastic job to hold the project together for as long as what he has done. But is it enough to take OI long in to the future?
The long-term survival plan for OI is quite simple; there is none. At least not one that is obvious to the community outside the OI Administrative circle. The topic frequents continued discussion on the OI-Developers mailing list. Feedback from fellow developers, contributors and OI users is consistent and parallel among most.
In addition to the complaints regarding lack of long-term planning of OI development, there is also bitter frustration for the lack of leadership of the project. OpenIndiana’s former project leader, Alasdair Lumsden, resigned at the beginning of the year. His reasons are outlined in our very own Exclusive interview here on Unixmen. The whole problem of lack of leadership and ability to construct and implement a working hierarchy is like an internal virus that won’t go away.
A too familiar site
In my previous years, I have been deeply involved with three major Linux projects, one of them a rather important project funded by Red Hat. I don’t wish to name the titles of the exact projects of my past, the first of the major projects I worked on suffered an identical internal crisis and inability to implement a functioning hierarchy.
The second project was plagued by the complete opposite problem of being operated by a dictator who would not compromise on any of his own ideas.
And the third was completely dysfunctional from the very beginning. Although I departed all three of those projects some time ago, the point is I have witnessed first-hand the same issues as what the OI project is currently experience. And if the problem is not resolved soon, then OI will die a slow, painful and lonely death. I don’t wish that upon OpenIndiana, simply because there’s too much potential and good code to allow that to happen.
Forking a fork
So what does the future hold for OpenIndiana? One thing is for certain, the project in its current form is not working and I am not too convinced that it can work without immediate action. Recently, project followers and contributors have taken things in to their own hands and forked the OI source code in to projects of their own. That will be something worth keeping a close watch on as something positive is sure to come out of newly forked code getting the attention it deserves. Hopefully we’ll see some much needed code improvements and updates to some of the included software in OI as outdated packages has long been another problem preventing the project moving forward.
OpenIndiana as a project only has itself to blame. The Illumos kernel project has well and truly become something respectful. OI was said to be the original project reference distribution for Illumos. Recent issues which I have outlined have changed that perspective and other distributions have pressed ahead with the backbone and leading support of Illumos. SmartOS and Illumian are the first to come to mind. Both of which have earned their respective places in the Unix sector.