Marketing Hype or Opportunity?

Sun's 'Project Copy Linux' not a Linux copy: "Sun spent months and months hyping Solaris 10 ahead of its release in 2005. The OS included some ground-breaking features such as DTrace and ZFS that Sun continues to flaunt today. The company has been unusually silent about the tools Solaris 11 will bring, making us wonder if OpenSolaris hasn't been a distraction from the core OS. You can see some of the discussed Nevada features here." -- Ashlee Vance, The Register

Oh, I think we hyped S10 for way more than just "months and months." Wasn't it more like a year or something? Anyway, it was worth it. But Ashlee hits on a good point -- distraction.

Our pre-S10 hype was traditional marketing, basically. The source was closed, development was all inside the firewall, and there really wasn't a community. All information had to flow through filters, and the information that was release was carefully screened and targeted. Now, however, the source is open, there is a growing global community, and development is starting to move outside (and many projects are already outside). So, how do you market -- hype -- all that stuff since it's already open for all to see and even participate in? Also, OpenSolaris isn't a distraction from the core OS. Actually, OpenSolaris is the core os. What can be a distraction, though, is trying to organize all the noise and conversations and source and binaries and projects and people into a coherent package that can be delivered in the form of hype. That's marketing. And that's extremely difficult on a project the scale of OpenSolaris. Before, you couldn't see all of the mess (in the best sense of the term, of course) because it was all hidden inside and only dribbled out via information gatekeepers. But now, it's pretty much all just out there. And more is coming when core kernel development itself moves external (the main ON gate operations, I mean, which is huge). From a marketing perspective, I think we are just now beginning to get a handle on how challenging this is. And what a big opportunity it is as well.
Comments:

Ashlee does have one good point... we're rapidly approaching the three year mark for Solaris 10, the point where Sun has historically released a new version (I mean all the way back to Solaris 2.1). And yet, there's no talk about features for Solaris 11. Everything in OpenSolaris is getting backported into various Solaris 10 Update releases. At some point, purely from a support lifecycle perspective, you have to draw a line and declare the next version. Unless you want to start having your support matrix be based on 10u1 (10.1), 10u2 (10.2), etc.

Posted by spp on July 31, 2007 at 03:58 AM JST #

I'd say, for starters, that S11 should include: - ZFS boot - ZVOL swap/dump - Rock solid pNFS - dTrace GUI - Xen integration complete (DomU, Dom0) - Zones fully integrated with Crossbow - Brussels complete (Nice to have) - FUSE/sshfs integration complete - Caiman installer fully integrated w/ ZFS - Built with S.Studio 12 (optional)

Posted by Brian Gupta on July 31, 2007 at 01:53 PM JST #

I tend to agree with you Brian, those seem to be the biggest ones. I think the shell projects (KSH93 integration, particularly) might be fairly high priority, see Brussels and Studio 12 as requirements (be nice to have Studio 12 delivered with the OS, like back in the SunOS4 days). So, is there a Solaris 11 product manager who has a roadmap that can formalize this?

Posted by spp on July 31, 2007 at 04:44 PM JST #

Post a Comment:
Comments are closed for this entry.
About


Search

Archives
« April 2014
SunMonTueWedThuFriSat
  
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
   
       
Today
Bookmarks

No bookmarks in folder