Java Following Solaris

Interesting to see that the opening of Java may follow a similar path as the opening of Solaris -- Sun to offer some open Java goodies while it figures out the rest. Ok, they are not that similar, but in this one respect there may be some similarities with a demonstration of sincerity. We did that with DTrace when we opened it in January of 2005 while our pilot program was still under way and before Solaris 10 shipped and before we launched OpenSolaris. I was not supportive of the strategy at the time because I felt it distracted us and diverted resources, but I was wrong. It worked out pretty well -- even though it did distract us and divert resources. :)

But Dana Gardner says, "Oh, and while the good will gestures of early OpenSolaris, such as DTrace, didn't exactly change the open source world given the lack of immediate practicality...."

Well, sure, it wasn't immediately practical in the sense Dana is (understandably) thinking of it. But it really wasn't intended for that purpose or to "change the open source world" in any way at all. Instead, it was intended to demonstrate our seriousness in opening the very best code we had, and it was directed at future OpenSolaris developers, Solaris customers, Solaris system administrators, and anyone involved with Solaris in universities and governments. Keep in mind that even at that late date many people still didn't believe that we were actually going to open all of our most advanced innovations in Solaris. The stuff we just spent hundreds of millions developing. In fact, I regularly had conversations with customers at that time who said, "No way. You'll probably keep the best stuff for yourself." Well, we did the opposite, and that was our intention all along. After the DTrace demonstration, though, I never had another conversation with a customer or developer who questioned what we were going to open. The strategy worked perfectly. It was really just another step in opening Solaris and building the OpenSolaris developer community.

Tags:
Comments:

The SPARC vs. x86 comment by Rosenberg was incorrect, as you pointed out. But the other examples are very valid, I think - Netbeans never recovered from Eclipse's head start, despite (finally) becoming a good IDE now. Nobody uses SunOne AS, JBoss owns the low-cost market - it works and I know it, why should I care about SunOne?

And while you verbosely dismiss the claim that Sun hyped the release of the Solaris source, you do not deny that it was a (desperate?) reaction to the success of Linux. The IBM comparison is telling, and supports Rosenberg's assertion: IBM chose to incorporate their AIX code into Linux instead of establishing yet another open source OS. I think that's the right way, bundling our forces.

Posted by Stephan Jaensch on June 21, 2006 at 05:57 PM JST #

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


Search

Archives
« July 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
31
  
       
Today
Bookmarks

No bookmarks in folder