« At OSCON | Main | links for 2006-07-26 »
20060725 Tuesday July 25, 2006

Respecting The Developers

Spontaneous Gathering

So why doesn't Sun just pick an OSI-approved license, apply it to the Java source code and be done with it? Why claim it's so hard to open source the Java platform? I've heard plenty of people asking that question (see, for example, a comment to Nat's blog accusing Sun of just making excuses to endlessly delay it happening). The number of comments on the Slashdot thread about the ComputerWorld interview I did show it's interesting to a lot of people.

There's some stuff you'll instantly guess - for example, the code has to be checked to make sure we actually have the right to put an open source license on it, and we need to make sure that compatibility is preserved as paramount. But there's a dimension that's being overlooked. In making Sun's implementation of Java SE open source, we are not starting from a position where there is no community. Every single line of code is already the responsibility of a programmer somewhere that works for Sun. Every module has a set of committers owning it. They are passionate and dedicated, they pour themselves into it just like any other successful project.

One of the biggest questions on the table right now is, how can we open the code, invite participation from everyone with the interest and the skills to do so, and still respect the work and ongoing contribution of the existing developers? Many of them have spent a decade or more making the Java source code excellent. We can't just dump the code on the street and invite applications for the role of owner for each module. That would be hugely disrespectful, dishonouring the developers who have devoted themselves to the code. And yet neither can we just assert that no help is needed.

Resolving this deep governance issue is the main challenge, and I know the Sun Java team intends to get it right, both respecting the work of the existing community and inviting the growth of a new community. It's a tough job.

Posted by webmink ( Jul 25 2006, 05:20:24 PM PDT ) Permalink Comments [2]
technorati del.icio.us digg slashdot
Comments:

If your Sun Engineers don't have the "we wrote it so it's ours" mentality, they'll find non-Sun Engineers to be quite accomodating, patient, and helpful. They might even get the occasional "Dang this piece of code right there is gorgeous", and that will mean a lot more coming from someone outside of Sun than it would coming from a peer reviewer.

Posted by Christopher Mahan on July 25, 2006 at 07:13 PM PDT #

Sun's internal Java programming community could learn a lot from Sun's internal Solaris programming community, who have been down this path.

Posted by rnc on July 26, 2006 at 02:31 AM PDT #

Post a Comment:

Comments are closed for this entry.