[cobalt-users] Where Is The Future?

[Greg Boehnlein <damin@xxxxxxxx>]

Hello all,
	I have been on this list almost since it's inception, and have
seen the Cobalt products evolve, mature and die with Sun's EOL of the 
product line. The release of the Cobalt ROM and GUI code will resurrect 
the older units as projects using them begin to gain traction, but the 
community will take a long time to reach critical mass. We are already 
starting to see a large number of divergent ideas emerge as the community 
ponders what it's next steps should be.
	I made a decision to migrate all of my commercial Web/FTP and 
Email hosting to a rack full of Cobalt equipment back when the Raq3 was 
introduced. I have been playing with the Cobalt stuff since the 
introduction of the original Qube on the MIPS, and I never regretted my 
decisions to support Cobalt. I was excited when Sun purchased the company, 
but quickly became jaded when I saw the product become an afterthought and 
eventually irrelevant. I'm sure this frustrated many of the employees, 
developers and clients as much as it frustrated me. But enough of the 
history.. I don't want this to become a Sun bashing letter.. There are 
enough other people doing that.
	I am writing this letter, personally, in an attempt to try and get 
a handle on all of the projects that I am aware of and to spark discussion 
from the Cobalt community on where we might go from here. In my opinion, 
the Cobalt products were successful because of a combination of great 
technology wrapped in an easy to use, inexpensive package. They were 
simple to setup, easy to manage and powerful enough to meet the needs of a 
wide variety of users.
	Those of us that are highly technical have 20/20 hindsight vision 
and can easily see the flaws in the Cobalt OS. There are plenty of 
opinions about what Cobalt did wrong with some of the decisions they made 
early in the process, but we can't change history.
	Which brings us to the present. We have several paths that we can 
take as a community now that the decision making process, code and 
development efforts are in our hands. I see the following paths available:

1. We can continue to support the Cobalt OS (RedHat based) through 
uprades, modifications and new releases to the existing code base. 
Examples of this would be the work that I have done to get Raq4 and Raq550 
systems to run under Vmware, User Mode Linux and on standard PC hardware, 
as well as my hacks on the Cobalt Installer.

2. We can graft the Cobalt GUI onto existing distributions, such as Fedore 
or Debian. Examples of this would be the Blue Quartz and OpenQube 
projects.

3. We can roll our own customized distributions learning from the Cobalt 
code, but writing a new Control Panel. Examples of this would be Ryan's 
Qbalt project.

4. We can build installers to drop standard distributions (Fedora, Debian, 
Suse etc..) onto the Raq hardware and add commercial control panels such 
as Plesk. I don't know of anyone that is doing this, although I did come 
accross a page that explained how to install the Mips version of Debian on 
a Raq2.

	Each of these paths has it's Pros and it's Cons. Each of these 
appeals to different people in different environments. For example, as a 
technical user I see the flaws in using Bind and Sendmail on the existing 
Cobalt distribution, but I also have a large base of installed users on 
this codebase and would like to see a migration path. As a businessman, I 
see the inherent value in morphing the Cobalt OSRCD and Distribution into 
something that can be installed easily on standard PC's. From a practical 
perspective, I think it will be a lot easier to fix what is wrong with the 
Cobalt stuff rather than scrap the years of development and engineering in 
favor of starting over.
	I know that everyone on this list has their own opinion and I'd 
like to hear it, and discuss it. If you have anything to add please speak 
up!

-- 
    Vice President of N2Net, a New Age Consulting Service, Inc. Company
         http://www.n2net.net Where everything clicks into place!
                             KP-216-121-ST