On Wed, Nov 21, 2007 at 10:01:53AM -0800, Gary Gendel wrote:

> Solaris 11 is the next commercial version of Solaris 10. It will come out of 
> the OpenSolaris work being done in Nevada and Indiana.
> 
> I believe the plan is to make the OpenSource hierarchy:
> 
> Nevada ON
> SXCE
> SXDE
> Indiana
> 
> SXDE is more stable than SXCE or ON.  Indiana should be stable once it gets 
> beyond the "preview" releases.

This doesn't really make any sense.  ON is a consolidation, not a
product at all.  It's not even possible to run ON by itself; although
it contains the kernel, bootloader, libc, and other core components,
it doesn't include libm.  You won't be able to do all that much
without that.  ON is one of several OpenSolaris consolidations.  These
consolidations together comprise OpenSolaris (a collection of code,
not a product) and are used by Sun and others to make products.  Among
those are SXCE, SXDE, and Indiana.  "Nevada ON" doesn't make any
sense; Nevada is simply the name of the current release; i.e., "ON
10", "ON Nevada", etc.  Nevada should have just been called 11 (SunOS
5.11) but the marketing people screwed things up as usual.  So there's
no "hierarchy" per se.  All of the existing OpenSolaris-derived
products *contain* ON, specifically the Nevada release (version) of
it.

Stability is based on changes from one release (or version, since the
word release implies that something is a formal product) to another,
so the stability of all the contents of the consolidations is the same
with respect to Solaris 10 and previous Solaris releases; Nevada is a
Minor Release.  The Architecture and Process Tools Group has more
information about what this means.  Note that Indiana contains stuff
that hasn't been through architecture review; although the stuff it
takes from the consolidations has, and will therefore follow Minor
Release rules, the other stuff it includes contains changes that may
not follow those rules.  Therefore it is unsafe to assume anything
about stability if you use Indiana.  SXCE and SXDE contain only bits
that have passed architecture review and been deemed suitable for
delivery in a Minor Release.  Therefore, relative to the previous
release (S10) they should all be equally stable.  From one build to
another, however, all bets are off.  Again, the ARC documentation
explains why this is so.

> Currently Indiana is trying incorporate features not found in the 
> ON/SXCE/SXDE products, such as a new packaging scheme and root being a 
> "role", not a user.  I believe that these features will be pushed back into 
> ON/SXCE/SXDE in the future.
> 
> So, pick up a copy of Indiana, SXDE, or SXCE and start playing.  All the dust 
> will settle by the end of Q1 2008.

Really?  I don't see any "dust", I just see a lot of people who are
confused.  Mostly that's Sun's fault; it's allowed some people to
corrupt the language of engineering through imprecise usage, and it's
done a terrible job of naming its products and explaining to customers
what they are and how they differ from one another and from
OpenSolaris.  This may be deliberate, or it may just be sloppy.  But
there also seems to be a marked unwillingness to learn how OpenSolaris
is constituted and how it differs from products built from it.  It's
not really that complicated.

-- 
Keith M Wesolowski              "Sir, we're surrounded!" 
FishWorks                       "Excellent; we can attack in any direction!" 
_______________________________________________
opensolaris-help mailing list
[email protected]

Reply via email to