On 8/18/2010 12:24 PM, Linder, Doug wrote:
On Fri, Aug 13 at 19:06, Frank Cusack wrote:
OpenSolaris is for enthusiasts and great great folks like Nexenta.
Solaris lags so far behind it's not really an upgrade path.
It's often hard for OSS-minded people to believe, but there are an awful lot of places
that actively DO NOT want the latest and greatest, and for good reason. They let the
pioneers get the arrows in the back. Their main concern is stability over all else.
Gee-whiz new features might seem great to someone who's used top patching their Fedora
installation with 32 patches every morning, but for critical high-availability stuff that
absolutely, positive, can NEVER go down, staying comfortably in the middle ground is the
ideal strategy. Sun's own white papers on patching advise that the best practice for
patching is "do it when there's a specific reason".
Solaris isn't "so far behind". It's right exactly where the market wants it.
There are plenty of bleeding-edge operating systems out there for those who prefer to
live on the edge. As a Solaris sysadmin, would I like to use all the nifty geegaws on my
production systems that I use on my desktop? Sure, in a perfect world I'd be able to do
that. But that's not the reality, and I'm not risking the business or my job on anything
less than ten thousand percent tested for years before adopting it.
"Newer" != "better".
----------
Well,
Most of the systems people like me that I know also value stability and
conformity to expectations (i.e. standards) over new features.
That said, stability vs new features has NOTHING to do with the OSS
development model. It has everything to do with the RELEASE model.
Also, to answer Frank's statement: yes, there *is* an upgrade path from
Solaris 10 to OpenSolaris. There will likely be a *better* one when what
was OpenSolaris is productized and turned into Solaris Express, soon to
be Solaris Next (11).
Take a look at Fedora vs RedHat Enterprise. This is the closest Linux
analogy we've come up with for showing the (former) difference between
OpenSolaris and Solaris 10/11/etc.
While there were certainly a few folks who ran OpenSolaris in production
(who absolutely needed the new features and couldn't wait until they
made it to Solaris 10), I'm going to say that 99.999% of people ran
Solaris 10, for exactly the reasons you indicated above.
All that said, using the OSS model for actual *development* of an
Operating System is considerably superior to using a closed model. For
reasons I outlined previously in a post to opensolaris-discuss.
--
Erik Trimble
Java System Support
Mailstop: usca22-123
Phone: x17195
Santa Clara, CA
_______________________________________________
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
