DivaBlog

Cross posting this to http://blogs.sun.com/DaneseCooper and my new blog at http://danesecooper.blogs.com/divablog

After the flurry of stories on Friday and over the weekend (my favorites being Jim Grisanzio's lovely farewell, Matt Asay's surprising testimonial and the original blog written by David Berlind), it will come to no surprise that I'm leaving Sun. I'm off to see if I can have any influence at Intel, a company which has benefitted hugely from the increased popularity of Free and Open Source software around the world. I'm quite sad to leave the many good friends I've made over the years at Sun. Change it always hard.

I'm excited to be going to Intel, though! There will be fascinating challenges (like seeing if I can run my Intel life from a computer running open source software), new people to make friends with (84,000 of them) and hopefully good work to do for the open source movement.

Ciao for now, Sun! Although global, the internet makes this is a very small playground of an industry and I know I'll be seeing you continue to shine!

Cross posting this to http://blogs.sun.com/DaneseCooper and my new blog at http://danesecooper.blogs.com/divablog

On the last day at eTech we were treated to the incomparable Larry Lessig speaking about the Future of Culture. He'll be at OSBC in early April doing a whole hour, and I highly recommend listening to him.

Also yesterday was also the "LifeHacks" talk. Cory Doctorow took better notes than I did. These guys are interested in helping make people more productive, and the talk focused on all the distractions that can come with some of the new technologies. I found myself disagreeing with much of their advice, however (although they did have some cool sites to talk about like 43 Folders) . Call me a weirdo, but I actually like having more than one thing going in my brain at a time. I am often seen knitting at conferences because it helps me listen for instance. So I have little friction from "context switching" on the computer or in my life (although increasingly my husband would take issue with that characterization). Okay, maybe I do need to focus more...:-)

Last (but not least), my pal Chris DiBona was there to launch the new Google open source site ! Chris and others have been working really hard on this for awhile and its so good to see it up and running. Open Source is just bustin' out all over the place :-)

And before I wrap up my notes on eTech 2005, I must commend my very good friend Duncan Davidson for his outstanding photo coverage of the event. It was so much fun watching him work (and checking the photostream to see what he'd captured). Hope he does a repeat in future!

But the absolute high point of Duncan's participation at this eTech (to my mind anyway) was his question to Larry Lessig. Duncan has been studying up on Creative Common's Licensing and he wanted to know whether they would develop a license that only grants rights to creative content to licensees who modify that content...except Duncan had a more shall we say graphic way to express it (and Larry's first response, "...I don't think I've ever heard it said quite that way...and if people choose to, ahem, do that with your content...that would be an interesting way to enforce remixing, yes..."). Duncan's an original all right :-)

And as long as Creative Commons is coming up, congrats are in order to the very cool Paula LeDieu who has evidently snagged the Executive Director job there. She'll continue to work for the BBC part time on the Creative Archive project that she has been directing for the last couple of years, which is a very good thin for them as well. Go Paula!

Cross posting this to http://blogs.sun.com/DaneseCooper and my new blog at http://danesecooper.blogs.com/divablog

Floor space only at eTech the last couple of days...but its such a great conference that I don't mind sitting on the floor :-)

Last night was the Make! party at eTech. For those of you who haven't yet seen the fabulous Make! magazine, check it out.

So yesterday there were lots of opportunities to meet people who tinker. Like the fantastic Andrew "bunnie" Huang, and Natalie Jeremijenko from the Feral Robotic Dog pack.

The dog thing was perhaps my favorite thing from yesterday. Its not just a geek project, its an environmental activism project. As my only brother is an environmental attorney, I'm always interested in work that supports the Sisyphian task of fighting the good (environmental) fight. The modified robodogs that Natalie and her students build are fitted with a new nose that sniffs for environmental contaminants. They build a bunch of them and then release them into the "wild" at sites where people are concerned about pollution. The dogs roam around (on wheels...they are robots after all) and when they find pollution they stop and "do something cute". Some of them are networked as well and those send data back to collectors. How cool is THAT?

Cross posting this to http://blogs.sun.com/DaneseCooper and my new blog at http://danesecooper.blogs.com/divablog

Just listening to Cory Doctorow speaking at eTech.

"...We eTechers are the rich hummus on the floor of the technology rain forest...we're the tiny frogs in the bromeliads."

You gotta love those images!

Best sessions from yesterday (for me)...first was George Dyson on "Von Neumann's Universe" telling us about the earliest days of computing at Princeton Institute, when Von Neumann and Kurt Godel were leading literally pouring the foundations for what we call the tech industry today. I'm a sucker for historical pictures, and this talk is chocked full of pics of people from that time (including a very cute picture of George and Esther as tiny children) and wonderful pages from journals and logs which were consigned to the archive where George works now. If you get a chance to hear him, by all means GO!

My other favorite session from yesterday was "How Sex Laws Incite Technological Change" by Annalee Newirtz from the EFF. Really fun and thought provoking session. I was reminded of my time at Apple working on QuickTime Conferencing. We were sure that videoconferencing via IP would really take off as soon as the porn industry got onboard. I loved the conclusion of this talk, where Annalee exhorted all of us to begin to support annonymizing technologies such as Tor (which she thinks will be increasingly useful for safely obtaining legal sex content in appropriately private ways, but which is even more important for people working to overcome opressive information control regimes). If everyone starts running and using annonymizers, then it will be impossible or at least much more difficult to kill freedom of information on the web. So you can do well (for the free world) by doing (yourself) good (as in Good Vibrations)!

I'll try to post more about todays sessions at some point.

I'm sitting in Larry Rosen's morning session at OSDL Summit today. Larry is making the point that license proliferation is bad, bad, bad...and mistakenly using the CDDL as an example of part of the problem. I have to say this irks me just a bit. CDDL was written to solve a problem which has caused much of the actual license proliferation, namely, that in order to use what I'll call the "Mozilla model" (meaning the hybrid of BSD and GPL under which Netscape chose to license Mozilla code 6 years ago)...you are actually forced to create a new license. MPL language is very specific. It not only names Netscape and Mozilla as Original Contributors...it stipulates Santa Clara, California as the venue and legal jurisdiction for any legal actions associated with the MPL.

Most of the "vanity" licenses seen at OSI are somehow derivatives of the MPL coming from large or small vendors. For PR reasons they want their name (or their project name) on the license and they have to change the name of the MPL anyway so why not submit a new one? They think as long as they are modifying MPL, why not make one or two additional changes? Add to this the fact that lawyers write licenses, and lawyers are sort of like coders...except their "code" is license text. The temptation to bit-twiddle as they edit to "make it their own" is enormous. As David Berlind has noted in his blog

"The result has been a balkanization of the open source software world into provinces whose borders are represented by license incompatibilities that prevent the unfettered intermingling and sharing of code that's theoretically one of the cornerstones of open source religion..."

OSI has long maintained that proliferation licenses are their own reward...In other words that companies who choose this sort of licensing won't ultimately get the benefit of the "open source effect" and will hopefully learn the lesson that sharing is better. This is a long process, not an "at the speed of business" sort of process and that has been frustrating for those watching the open source revolution from their leather office chairs.

But back to CDDL. If you consider it on its own (as a license, without all the FUD & speculation around Sun's intention in applying it to OpenSolaris) you'll see that its really meant to be reused. Its a "template" license, without pre-set jurisdiction, without brand-name issues. Its definitely not just another license. I would like to see OSI add a requirement going forward that new proposed licenses be required to prove how and why their license is different from all the predecessors.

Most, but not every minute of my last few months have been filled with OpenSolaris. I also attended a meeting in November where the GNUClasspath folks were trying to understand how to gain access to the J2SE TCKs so they can check their work for compatibility. This is very exciting. Now today a friend has pointed out "this". At the same time, our friends in Brazil have finally achieved full funding for the "javali" project.

AFAIKS, this is all goodness. Compatible Open Source J2SE was a foregone conclusion back in March of 2002 when the rules of the JCP changed to support compatible open source re-implementation of JCP specifications. As happened last year with J2EE, there will probably be multiple compatible open source re-implementations. The thing to watch for with IBM or anyone starting such a project is transparent open source licensing and process and full support for compatibility.

My friends have been asking for months why I've not been blogging. The answer is simple, I signed on to the OpenSolaris team awhile ago and its been taking up nearly all of my time. Also, most of what I've been working on has been "not for comment"...until today!

I'm listening to Scott McNealy just now talking about OpenSolaris release under the CDDL. What's more, he's announcing that the Sun Board of Directors has agreed to indemnify the OpenSolaris community for all 1600 patents associated with Solaris! Wow! I never thought we'd get them to go so far. This is better than IBM's grant: all the patents affected are core software patents, not a collection of patents that won't affect the community. And for OpenSolaris we are governed by the same CDDL license as everyone else, including the patent peace provision. Go and read it. This is a level playing field.

So, go check it all out, and you'll know what I've been up to for the last several weeks. More on why I think this is a good use of my time in my next blog. For now I'm going back to listening to the annoucement call....

Okay, so some of you have noticed that I've been missing in action for awhile.

My Mom, who makes great cookies, recently hurt her shoulder. Specifically, she had a torn rotator cuff (ouch!). Hurt mommy = no cookies :-(...

There was an open surgery (not arthroscopic...this was full-on scapel surgery and a 4-week immobilization of the arm), and I spent most last month in Los Angeles at Mom's helping her out.

Happily, she's much better now and I'm once again working to make the world a better place for open source. Expect to see more postings from me.

One cool thing I got to do while at Mom's was to visit the Renoir to Matisse exhibit at the LA County Museum of Art (the art museum of my childhood. Awesome show! Nice to see some alternative canvases

Spent most of this week in Europe, first at the opening day of the UNCTAD F/OSS Expert Meeting and then yesterday and today at the second OpenOffice.org Conference, in Berlin.

A common theme of both meetings has been concern about the effect of software patents on the F/OSS community. For those of you unfamiliar, the European Union has for months and months and months been considering whether they should start granting software patents. It seems the EU patent office was chartered not to grant software idea patents, but as applications have continued to pour some have been granted anyway, and there is mounting pressure coming from the US to formalize what is happening. There are even associations that provide detailed analysis of every twist and turn of the issue.

The hacker community (and I use that term to indicate white-hat F/OSS programmers, not black-hat people who break into systems illegally) are agreed that software patents are a big problem. They don't really have access to the infrastructure to file a lot of patents and even if they did they don't have the resources to defend any patents they might be granted against the corporate might of the principal software patent holders. They maintain that many software patents granted today are spurious because they grant monopolies on obvious solutions to software problems. It is increasingly difficult for a small player to write software that doesn't touch prior art, and that means software can only come from big companies who can afford to fight or buy rights to patents.

On the pro-patent side are people who point out that there are so many software patents already (particularly in the US) that it is impractical to try to toss out the whole patent system. They believe that the original purpose of patents (which are supposed to incent and reward original R&D by granting limited monopoly) is still a need of the technology community. These people believe that the system is broken not inherently but in its current implementation. Patent examiners are not doing a good enough job of winnowing out the obvious solutions from non-obvious innovations, and the barrier to entry for individuals is too high. Also many have noticed technology is moving too fast now for legal remedies to be much remedy (because by the time a judgement comes down protecting a given technology, the decision is often irrelevant in the marketplace).

At the UNCTAD meeting I heard several proposals, including creation of a patent collective from a South African (which I've heard proposed before and which actually might work). At OpenOffice.orgCon I heard that projects such as Apache have gotten much more saavy about patents and have begun to accumulate rights that cover their communities. How this debate works out is going to influence more than just the European Union, though, as many developing nations simply adopt much of EU policy as their own.

What do YOU think should happen?

You may have read some of the articles, blogs, and newsgroup postings about excerpts from Sun's settlement with Microsoft that appear in our SEC filing. The discussion centered around the protection we secured for the StarOffice brand, and implications for the OpenOffice.org community.

Well from my perspective Sun loves OpenOffice.org and has been a good steward. When we open sourced it nearly 4 years ago (can it really have been that long?) I was personally concerned that Sun would lose interest in the project before it had a chance to really take off, but I was happily mistaken. OpenOffice.org is in some respects our best open source community, because it shows that code isn't the only valid contribution and that people want to increase their freedom by breaking away from vendor lock-in to open standards, even at the desktop productivity level. After 4 years we still develop one of the most popular F/OSS software projects in the clear, even though it takes more time to do it that way. We want people to use it. We're trying to build a community here.

In many ways I think a strong community is the best hedge against legal troubles, too. All of F/OSS is still virtually untested legally, so participation is about revolution, not business as usual. What really keeps big companies from suing F/OSS projects is the negative PR impact. And notice that OpenOffice.org is widely adopted in Europe, where Microsoft has also been having other troubles of late. A strong community behind OpenOffice.org makes it harder for to pick on, period.

The funny thing about open source community is that one party cannot unilaterally push their agenda forward - even if they originally donated the code. Five years ago we hired an engineer to work on the Tomcat team and since we'd hired him to write code, of course we gave him commit privileges the day he started work. The community was outraged. They pointed out in no uncertain terms that it wasn't fair for a new engineer to automatically get commit privileges just because he worked for Sun! Ever since then every Sun F/OSS project requires new engineers from Sun to go through the same process of building reputation as any other community member, because like Tomcat they are now shared resources.

Doing things in F/OSS ways is a sea change for proprietary software companies. For years Mozilla.org was legally still part of Netscape. They actually wanted to separate it but they couldn't figure out who would pay (and provide health insurance) for the employees if they did. And most of the engineering was coming from inside Netscape, despite their many efforts to attract outside committers. But the Mozilla community continued to press on through negative press predicting their demise and supposedly scandalous stories that most of the committers worked for Netscape. They ultimately even survived losing their corporate steward, and lost some developers in the process. But true F/OSS people are stubborn and work continued on Mozilla until finally they are getting results! And I'm not the only one who thinks so.

I believe we would have loved to protect OpenOffice.org from future lawsuits by Microsoft in our settlement with them but F/OSS software is by definition a shared resource and we have no explicit control over all the elements of that community. The language of the settlement takes OpenOffice.org as an example and makes it clear that Microsoft reserves the right to bring suit against any F/OSS project against which it has a claim (whether or not the project is stewarded by Sun). This isn't really news, is it? Nothing has really changed, except that in our settlement with Microsoft we managed to get some protection for our brands.

There are many individuals, companies, and government agencies who benefit from the use and the growth of OpenOffice.org. Many of them also give back to the project by validating builds, serving on one or more of the groups that have formed to advise the project, or through free-form evangelism (telling their neighbor or their church or school or even their employer about the benefits of using OpenOffice.org). We are always looking for more people (and companies) to join the community. That's why we sponsor a yearly conference, and why, when Microsoft applied to have a booth there this year, we were happy to oblige. I'm still holding out a hope that they will see the light and join the standardization process for OpenOffice.org's XML file formats.

Think about what M.K. Gandhi said about the stages of change...first they ignore you, then they laugh at you, then they fight you, then you win. I'll be at the OpenOffice.org conference next week, working for community. Hope to see you there!

Okay, so I didn't write much more about EuroFoo last month...it was a great time and like many I learnt lock picking...but it didn't have the same sparkle as the "original" Foo (I guess we have to now call it CaliFoo? Since its in California?). Last night people here at CaliFoo were up late...4am...talking and talking and talking. And yet this morning everybody is up and sharing again!

Definitely my favorite conference these days ;-)

More later.

Zak Greant (Community Advocate) has left MySQL. His resignation letter (well, blog really) says he's left not because of concerns about the rightness of MySQL's licensing and community policies, but rather because it was the right thing for him to do...(my paraphrasing here) "leave before burnout sets in". All-in-all, Zaks' blog is pretty classy, and anyone who has done a stint working on community within a corporation can tell you about burnout.

But notice he asks the MySQL community to let the company know if his leaving is a concern...if it signals a "MyPocalpyse".

This blog makes me think of the not-so-classy, but famous resignation and postmortem posted by Jamie Zawinski upon his decision to leave Mozilla back in 1999. Most of the present Mozilla crowd will tell you they didn't thank Jamie for his letter (since it declared the project "dead")...but I might argue that his final act of publicly calling the direction of the project into question may have helped Mozilla to regroup and eventually prevail. Its a powerful thing when someone well thought of falls on their sword and says why.

Certainly Jamie's letter caught the attention of folks at Sun, who still reference it from time to time when we are discussing the importance of voice and community in our projects. If a given community it mostly happy and strong, then the occasional spat or disgruntled letter goes largely unheeded (as with the resignation of Jon Stevens from Jakarta), but if a community is teetering...a well-placed public letter can really galvanize change.

Just talking to my pal Doc Searls, and he's pointed out to me that Microsoft's Longhorn is in trouble. I especially like the penultimate paragraph, where Doc says:

In his opening remarks, Tim O'Reilly credited the need for the EuroFoo meeting to George W. Bush!

No kidding! George W. Bush? How's that?

It seems that so many F/OSS people from Europe have sworn off coming to the US (because its such a total hassle now due to Homeland Security paranoia promulgated by the Presidential Administration of George W. Bush) that O'Reilly had to bring the meeting to Europe...

Every cloud has a silver lining :-)