Life at Eclipse

There has been a flurry of articles and blogging today regarding the process by which Sun will release Java into open source. I believe that this is a topic of specific interest to the Eclipse community, given our basis in open source and extensive use of Java.

Apparently yesterday Sun announced a roadmap for their eventual open sourcing of Java. They’ve set a goal of having both Standard Edition (SE) and Micro Edition (ME) open source by the end of this year. In my mind, it is really good news that they have committed to making ME open. AFAIK, that had previously been ambiguous.

One element that I find curious is that Sun has talked quite extensively about where they are looking for inspiration on community building. Some examples include this mindmap from Cote’, and this quote from Laurie Tolson, Sun’s vice president of developer products and programs: “On the governance side people want an easy way to contribute and interact. We are looking at the Apache, Linux, Solaris and other governance models. We are still in the early stage of gathering input”. Simon Phipps says “Expect a steady stream of news from now on, as well as an honest desire for dialogue with everyone.”

I hope Rich Green hasn’t forgotten my email address, as I have to admit that I find Eclipse’s conspicuous absence from their lists of examplars to be somewhat disheartening. Especially given what a gift Eclipse has been to growing the Java development community over the past five years. The fact is that an Eclipse Foundation-style model of independent, open governance with no special votes or vetoes for any particular corporation is exactly the model that Java needs to have if it is to be successfully re-invigorated. I really believe that Eclipse remains the very best current example of how a company can set free a community of innovative projects which supports a large and diverse commercial ecosystem. And Eclipse’s dynamic growth since the creation of the Foundation proves that the model works. If it works for Eclipse, why can’t it be a model for Sun and Java?

Back in May at JavaOne I participated in the JavaPro Java Technology Roundtable, where I made the following prognostication:

Mike’s fearless prediction: you’re going to use CDDL, and you’re going to use an OpenOffice, all-Sun governance model, and people will hate you for it. [The “you” in this sentence refers to Sun.]

If you take a look at the list mentioned by Tolson, it seems like other than Solaris the examples they’re looking at are unlikely to get them very far. Emulating Apache is a non-starter. First of all, the Apache style of governance is antithetical to any project under single company control. Secondly, emulating Apache would just be dumb. Don’t emulate it. If you like their approach, just open source Java at Apache. I’m sure that the code contribution would be welcomed by Harmony.

Emulating Linux also seems far-fetched, as it is ultimately a reflection of Linus Torvalds’ benevolent dictatorship. And regardless of any shared god complexes, Gosling is no Linus.

Which leaves us with the OpenSolaris governance model as the likeliest candidate. Note the majority votes held by Sun appointees. Note that it is an advisory board to what remains a Sun-defined and Sun-controlled Sun project, just like OpenOffice and NetBeans. (Let’s ignore the JCP for now.) Contrast that to the Eclipse Foundation Bylaws, which form a legally distinct entity controlled by an independent Board of Directors with a fiduciary responsibility to “…to cultivate both an open source community and an ecosystem of complementary products, capabilities, and services.“

I do totally agree with Simon that it is categorically unfair to “...characterise [the process of open sourcing Java] as an intransigent Sun fighting bitterly against an obviously right social movement…”. But at the same time I do think that it is fair to carefully scrutinize their decisions on the topics that really matter to all of us: licensing and governance.

I wish Sun well in their odyssey. I really hope that they can find a way to establish Java as truly free and independent. But so far I stand behind my prediction.

My lovely and much younger sister works as a paralegal in a Dallas law office when she’s not running triathalons. So one day last week I guess she was bored and typed our not-very-common surname into the US PTO database and found this piece of hilarity. I genuinely had no idea that this thing existed.

I only very vaguely remember reading the patent application. I do distinctly remember the “invention”, which consisted of one afternoon with a couple of guys (particularly Greg Melahn and Sam Ruby) brainstorming ideas on how we could get two very different systems to talk to one another. Not exactly earth-shattering design work.

My recollection is that this stuff never even got implemented, but I could be wrong about that, as I bailed from OTI/IBM in May, 1999.

An astute practitioner of the software art will quickly realize how lame this thing is. I certainly don’t mean to criticize anyone involved in this particular patent. We were all just doing our jobs within the system that exists. But the fact that the US PTO felt that it was worthy of patenting is all the reminder any of us should need as to just how broken the existing system really is.

And the funniest part of all this is……This patent is referenced as prior art by four other patents (6996826, 6883172, 6484311, and 6438744), all issued to (you guessed it!) Microsoft.

Anyone else out there with sinfully funny patents bearing their names that they want to ‘fess up to?