The Real Ecology of Computing

Last Friday was Java day for Sun - Happy 10th Birthday to the Java platform! And to the whole team, those at Sun, those no longer, and those with our partners and customers: Thank You. You've changed the world. And it keeps on changing. The market's tipped, there's no going back. The world is participating in a fundamental (and fundamentally entertaining) way.

Friday was also Earth Day. And reading some of the Earth Day PR coming from the big PC companies got me to thinking about how wasteful our industry is. Most of the PC companies made announcements last week that made it easier to throw computers away. And not like I'm going to go inhabit a tree anytime soon, but surely what we should be doing is figuring out ways to stop throwing PC's away. And stop the waste more broadly, vs. making the waste more palatable. We are among the most wasteful industry on the planet. Want proof?

The average utilization in a datacenter is 15% (what we see is tyipcally lower, but we like to be polite). That implies some 80% of the capital that goes to purchasing computers, 80% of the real estate, and best of all, 80% of the electricity to power and cool the unused systems - is flat out wasted. Greenpeace, where are you? Want to know how to solve at least the tech industry's contribution to the power crisis? Some ideas:

Display over IP. DOIP ("Do IP") is to the PC industry, what VOIP (voice over IP, simplistically, using the internet to make phone calls) is to the telecommunications industry. Phone calls are near to free at this point, and the business model is undergoing radical change. It's inevitable that pervasive and sufficient bandwidth will allow most of what happens on a client to migrate to the network. Why upgrade your PC if you can rely on plentiful bandwidth to have someone centrally deliver it as a service? You don't upgrade your TV set, BBC and News Corp do it for you every evening with fresh content. And you don't buy a new TV to watch it. The same should apply to your PC. DOIP is to a PC as XMRadio is to a CD player.

Sun's entry into the DOIP race is called a SunRay. Its primary value? It's a PC you never, ever upgrade. Whose intelligence is located in someone else's datacenter. With a beautiful monitor secured with a SIM card (just like the one in your cell phone). No, it's not perfect for all applications - SunRay's only work where Google works (ie, where there's a reasonable network connection). But I wouldn't bet against ubiquitous network coverage. Even on an airplane.

Now why use a SunRay (or other DOIP device)? A Sun Ray uses 15 Watts. Fifteen. Compared to 120+ Watts for your basic PC. Multiply saving a minimum 100 watts per office (plus the savings of not having to cool the offices from the heat dissipated by the PC), and at 15 cents per kilowatt hour, that's minimum $1.5 million dollars a year in savings for 10,000 employees. Free money, and you're saving the planet. And real estate. And noise (the SunRay's completely silent.) And a side benefit? You can't steal the data on them - they're completely stateless.

Next on the list?

Logical Partitioning, also known as LPAR's. I ask CIO's all the time, "how well utilized are your mainframes vs. other systems?" The answer's always the same, "they're so expensive, we try to get every last ounce out of them." Why are they more highly utilized? LPAR's, the virtualization technology that allows a mainframe to behave like lots of littler mainframes. Which is why we added containers to Solaris 10 and our storage and networking platforms. Like logical partitioning on a mainframe, containers drive utilization up by enabling a big (or small) IT asset to be divided up into lots of independently secured assets. As utilization goes up, relative power and space (and absolute dollar) consumption declines. Which is to say, if you're really serious about greening the planet, use technology to solve the utilization problem and stop wasting so much energy. And space. And money. Did I mention Solaris is free for download and use? On any computer?

(And for the analysts who quip, "but won't higher utilization mean people will buy fewer computers?" Funny, when we build 'em twice as fast, people don't buy half as many... same applies to utilization. Two times the utilization is two times as useful.)

And finally, chip multi-threading. Check out Dell's latest 6850 gas guzzler with 1470W (!) power supplies - wow. That'll dim the lights. Surely someone ought to be working on way more performance, with way less power required. I know our customers would love that. Here's a quote from an account summary I was reading over the weekend:

[Customer Name Withheld]'s team met with lead developers from various business units about two weeks ago. BU's have been warned that CPU speeds will no longer increase and will likely be lower in the future due to thermal limitations. Therefore, [Exec Name Withheld] has put educational and code review processes in place to assist developers in writing threaded code.

Seems like a massively threaded, lower power systems lineup would be just what the doctor ordered. The next wave of computing won't be just about fast math. It'll be all about the economies of scale, whether motivated by financial or environmental responsibility. Seems like a safe bet, even when it's not Earth Day.

Maybe it's time we called Greenpeace.

Posted on 07:43PM Apr 24, 2005 | Comments[1]

Share this post  | | | | | |

Choosing to Participate

Of the speeches I've given over the past six months, the one responsible for driving more discussion in our customer base, on the 'net, through blogs and media coverage (or Slashdot postings and Groklaw debate), is this one:

The Participation Age. (And don't be alarmed, the folks at IT Conversations got the title wrong - it shows up as "Building Billion Dollar Businesses").

Why did it generate all the coverage? Because competition - if only in the marketplace for ideas - is way more interesting than having everyone stand around and agree with one another. Or worse yet, sue one another.

The coverage was far more inflammatory than the speech. (Then again, my wife tells me I appear taller on TV.) But it's driving education and awareness around open source licenses - and our decision to use a license for the distribution of OpenSolaris into which any code can be combined (GPL code, too, despite the pundits) - without obligation or constraint. All licensees are shielded by the entirety of Sun's software patent portfolio, one of the industry's most robust.

Like I say in the speech, choosing to participate is a good thing.

Posted on 07:49PM Apr 17, 2005 |

Share this post  | | | | | |

A License for the Participation Age

UPDATE: Presentation from OSBC now available here.

As I mentioned earlier in the week, I delivered one of the keynote speeches at the Open Source Business Conference on Tuesday. I'll provide a link to the presentation as soon as it's available. The premise of the speech was that all traditional software will ultimately go to free - not because it's lessening in value, but because an open network is eliminating the price of distribution as a barrier to entry or opportunity. Free and simple access implies a far freer market, in all senses of the word.

Since the speech, I've been deluged with input from around the world. The messages generally fall into one of a few areas. First, overwhelming support for Sun's commitment to free and open source software and the Participation Age; second, questions from those wanting to compare Sun's CDDL license with the Mozilla and GPL licenses; and third, sharp (frankly, disturbing) rebuke from those that believe the GPL should be the only software license allowed (and that dissenters should be pilloried).

So I thought I'd provide some clarifying thoughts on the evolution of Sun's CDDL, and our primary motivations. (As a preface, I'd encourage anyone with the interest to read the license. We designed it to be readable by non-lawyers, to do away with the legalese and opacity that normally surrounds software licenses.)

To date, the evolution of a considerable volume of free software has occurred without much discussion around the differing rights and obligations of the licenses (and licensees) involved. As more of the world participates in an open network, more, and more valuable IP is being created - across the globe. Which means the licenses under which IP is distributed are only increasing in importance.

With that in mind, Sun worked hard, with a global community of contributors, to build a stable license for the contribution and evolution of not only Sun's intellectual property, but all intellectual property, to the F/OSS world. At its heart are a few assumptions.

First, we assume all developers (whether nations, universities, manufacturers or software developers) want choice: a developer using OpenSolaris under the CDDL can choose to weave their own IP into an OpenSolaris or derivative distribution without any obligation to deliver their privately developed intellectual property back to Sun or their competitors. Licensees can stand on our shoulders, build derivative products based on OpenSolaris, without our hands in their pockets or business plans.

What does this mean practically? Under the CDDL, you are free to choose what to reveal, what to withhold, and how to price your products. The CDDL encourages self-determination by giving developers the basic building blocks of the entire OpenSolaris operating system without any obligation to disgorge their private property or predetermine its price. You can withhold from Sun, from the OpenSolaris community, from the world, anything you build. Or you can choose to share. It's entirely your call.

Unlike the GPL, this choice permits participants to take control of their future - to build upon the assets created by Sun and others without any future obligation. The rigidity of the GPL is what renders it incompatible with the CDDL and the Mozilla Public License - both the CDDL and MPL gladly accept the intermingling of any other source code (including GPL). And why didn't we use the MPL? For the most part, we did - we simply removed those portions that conveyed to the Mozilla Foundation the ability to revoke the license, and force any disputes to be heard in Santa Clara, California (see Section 11., which obviously doesn't work for a global market).

A second assumption was that Sun's considerable patent portfolio has enormous defensive value for the future of free and open source software - and rather than being silent on whether patents are good or bad (my answer, it depends), CDDL licensees simply inherit the full protection of Sun's portfolio. They gain that benefit through a blanket patent license, covering every software patent Sun has issued. We didn't pick an arbitrary subset of our patents to generate press. Instead, we granted a license to the entirety of our software portfolio. We attempted to encourage participation. (Separately, do I think software patents need reform? Yes. But in the interim, we put our money, and our patents, where our mouths are - and saved the community from having to sacrifice their future to support our views.)

Finally, we explicitly designed the CDDL to be a license companies and individuals could reuse. We worked hard to architect a license that wasn't just good for Sun, but was good for any entity - whether commercial or governmental - seeking to accelerate participation, without foisting our or their social or economic models on potential recipients. The CDDL is a reusable license large companies and IP holders can use to make contributions to small companies and nations, conveying the protection from multi-billion dollar patent portfolios without dictating business terms. The license preserves the rights of a licensee to run their economy and business as they see fit - without colonialistic interference or obligation.

Does this mean we don't believe in the GPL? Absolutely Not. We're one of the most prolific contributors of code under a GPL license (remember, we delivered Project Looking Glass and OpenOffice under it). But a license governing the development of a productivity application used to create documents deserves fundamentally different scrutiny than a license for infrastructure to be integrated into nascent industries or economies. And understanding the difference is crucial to preserving choice and opportunity. Just as there's no one desktop for all the world to use, there is no one operating system - and critically, there's no one license for all open source software.

Do I think there's hypocrisy on the part of large US companies encouraging usage of the GPL as a license in the developing world? Absent education and awareness related to the risks of unintentionally incorporating GPL code, yes I think it's hypocritical. Who are the largest economic beneficiaries of the GPL to date? No one disputes it, dominantly American companies.

But we all benefit from a network economy in which all nations and business can participate on an equal footing. From Sun's inception, we've always believed a rising tide lifts all boats. Just as it did with TCP/IP (the basic building block of network computing). And if there's a lesson in that history, it's that choice and competition matter. In technologies, licenses and most of all - to my friends on the extremes of the GPL community - the free exchange of ideas.

Posted on 09:35PM Apr 10, 2005 |

Share this post  | | | | | |

The Participation Age

I'm keynoting at the Open Source Business Conference tomorrow. Tune in, and you'll hear me talk about Sun's objectives with free and open source software: to lower the cost of computing, lower barriers to entry, and consistent with our history, fuel the communities that give rise to the next era of network computing. What's the next era? The Participation Age, in which an open and competitive network fuels growing opportunities for everyone - not simply to draw data or shift work around the world, but to participate, to create value and independence. If the Information Age was passive, the Participation Age is active.

You'll also hear me tell a couple stories that highlight a growing threat to the evolution of a participative network - duplicity working against the objectives of those most in need of assistance.

The first story relates to integrity in the world of open source. A few months ago, I was in a park over a weekend, when an individual, the CEO of an open source company, walked up and introduced himself. We got to talking, he was obviously a very smart guy. We were talking for awhile, when I began to relay my frustration surrounding Sun's (now fading) perception as 'proprietary' in the marketplace, especially among the open source community.

I told the individual that we'd been earnest in our contributions to the community, from NFS to OpenOffice, Java to OpenSolaris (Sun is by far the single largest contributor to the open source marketplace, having recently dwarfed even UC Berkeley's contributions). And we're sincere in our desire to preserve and promote standards and competition - from railroads to electricity, standards are critical to progress.

That said, we believe there is real value in intellectual property, real (if maximally defensive) value even in software patents (notwithstanding the USPTO's atrocious track record). But nonetheless, we'd been criticized for our unwillingness to allow Java to fork, for continuing our investment in the open sourcing of Solaris (while our peers dumped their OS's for Linux), and for resisting the position that software patents should be banned entirely. And most of all, we'd been roundly criticized for suggesting open source was not, and cannot be considered the equivalent of an open standard.

His advice? "Just lie. It's what a bunch of us do to keep the slashdotters at bay." If you've been reading my blog for any length of time, you know that's not my style - and if you've been watching the headlines recently, you've seen the impact duplicity has on executives and the companies they manage.

So why are we open sourcing Solaris? To drive innovation, and eliminate the barriers to participating for those currently unable to afford the benefits of an open network (that's why the binaries are free, not just the source code). Why did we choose to author our own license, the Common Development and Distribution License (CDDL), vs. use an existing one? First, because we felt the existing licenses had serious flaws - the Mozilla Public License, for example, restricts from the issuer any power to change the license, and predetermines all disputes must be heard in Santa Clara, California (not good if you're a Bolivian developer). Alternatively, the GPL expressly limits choice by disallowing the inclusion of non-GPL code into GPL projects - and exports a form of IP colonialism to nations seeking to create their own means of production.

What tack did we take? We created an open, reusable license, and contributed it back to the community. And with it, we remain consistent in our belief that intellectual property licenses exist to enforce IP models, not the political theories of a paternalistic nation or individual.

And having just spent some time with a breadth of network equipment OEM's at the GSM World Congress; and a series of representatives from developing nations at a recent customer event - I can assure you they are both suspicious of richly valued companies, with enormous patent portfolios and legal teams, evangelizing the benefits of the GPL and the elimination of software patents. They're beginning to see it as a means of forcing them to disgorge their intellectual property, and convey it to those same richly valued companies. Free software they understand: they also understand exploitation.

The second story relates to the economic motivations behind some in the world of open source. One of the most motivated folks I met recently was a venture capitalist, among the most successful in the history of the industry. He's backed a variety of household names in the IT world. And he'd begun to turn his attention to the frenzy around open source.

Over the course of our conversation, he started telling me about his efforts to encourage his portfolio companies to lobby governments to bring software patents to an end. What? Until then, my view on the elimination of software patents was that the vanguard of that position were those without the ability or wherewithal to fight against established patent aggressors. Those who could honestly look at the confusion the US has created around the proliferation of spurious patents, who sought to help others defend against potential inequity - while they built their own value.

But I'm confident an accomplished Silicon Valley VC wasn't the sympathetic constituency the European Union had in mind when it recently considered the reformation - and elimination - of software patents. Asking fledgling nations without software patent portfolios to forego the creation of defensible IP - while the wealthiest nation on earth keeps its powder dry - doesn't seem equitable or desirable. At best, the view that patents should be eliminated for everyone but the US is misguided - at worst, it's a truly cynical attempt to magnify inequities rather than destroy them.

What's my view on patents?

Remember the $92M we paid to Kodak - to resolve a legal dispute brought about by Kodak's acquisition of a patent whose exclusive value was in litigation? What is indemnity worth? Clearly something, minimally the $92M Sun paid to protect its customers and the Java community - the whole global Java community, not just those we're fond of. Companies should stand behind their products.

And having said it before, let me say it again. I believe in IP. I believe in its value, both economic and social. I believe it should be protected, as any other property, as a means of fostering independence, investment and autonomy. And not just in wealthy nations - but in those struggling to build wealth or pay down debt. I believe the creation, protection and evolution of intellectual property can accelerate everyone's ability to participate in an open network.

And that, surely, should be everyone's common goal with free and open source software. It's not about bringing the competition down, it's about driving global participation up.

Posted on 11:11PM Apr 04, 2005 | Comments[1]

Share this post  | | | | | |

The Downside of Being an Officer of a Public Corporation

...is that it's very difficult to write a good April Fools blog without feeling the need for serious engagement from the corporate legal team.

And much though I love Sun's lawyers, creative cooperation doesn't seem in keeping with either April Fools day, or our collective fiduciary obligations.

But it's not for a lack of creative ideas. You'll have to trust me on that...

Posted on 04:42AM Apr 01, 2005 |

Share this post  | | | | | |