Gary Zellerbach's Weblog: Focus on eMarketing and Electronic Software Distribution

While my job responsibilities remain essentially the same for now, our team was recently "re-org'd" into the Cloud Computing Marketing group at Sun. As such, I quickly learned that to those outside of our industry, they usually have no idea what I'm talking about when I say I'm in the "Cloud Computing" group. "Cloud" is of course the buzz word at the moment, but it's still not a term that has permeated much outside the tech realm. My colleague, Neave Connolly, was facing the same challenges describing "Cloud" to her friends and family, so she undertook some research to look for a good, simple definition we could use for the non-techies in our lives (yes, this means you, Mom ;).

Neave found some great information so I thought I would summarize and share.

First, though, I often lead with the "utilities" analogy, as that usually helps. (Remember when Cloud Computing was pretty much "Utility Computing" a few years ago?) It goes like this: When electricity was first invented, companies would have to build their own power plants and hire their own electrical engineers to manage them. After time, electricity became a utility, and companies were no longer burdened with providing it themselves. Computing is (in theory, at least) following a similar pattern. Why should every company have to build their own data center and hire their own system administrators? Why not buy computing as a utility delivered over the Internet, just as electricity flows through wires into our factories, offices, and homes?

And that brings us to the Cloud Computing articles for those who'd like to understand a bit more...

• The first reference is a good, simple, high-level overview and definition of Cloud Computing from TechTarget.
• Next up, Gartner Highlights Five Attributes of Cloud Computing.
• And finally, there is a great deal of Government interest in CC for the efficiencies it can offer, and the National Institute of Standards and Technology (NIST) has weighed in with their Working Definition of Cloud Computing.

All the articles basically define and describe these key attributes of CC:

1. CC is sold on demand. In many cases, you might only need a credit card to buy computing power and services.
2. It is "scalable" and "elastic," meaning you can use as much or little capacity as you need, there is no upper limit on what you can use, and you pay only for what you use (just like electricity).
3. It is provided as a service, fully managed by the provider. The customer need not know "what's behind the curtain," just that they are getting a service that meets their needs. (Sticking with the analogy, do you know what power plant provides your electricity? And even if you did, do you know how many generators are in it, what their capacity is, etc.? Do you care?)
4. The "Cloud" is shared. Multiple users may be serviced from the same infrastructure and systems (virtualization being one key, but that's another tricky definition I don't want to get into right now!). 
5. It's Internet based. All services are managed and provided over the Internet using Internet standards and protocols (though as a still-evolving technology, there are bound to be new CC-specific standards emerging over time).

There are many interesting debates ongoing about what is or isn't CC, how does or doesn't it differ from "older" technologies like Grid and Utility Computing, where's the line between "public," "private," and "hybrid" clouds, etc. It just points to the youthfulness of this technology, while the huge amount of interest, attention, and debate (within our industry, at least) point to the almost unlimited potential this "paradigm shift" in computing holds. Will company data centers go the way of company owned power plants? We'll find out in time, but meanwhile, you hopefully have a better idea of what is meant by "Cloud Computing"!

I first wrote about our forays into "1-click" downloading a while ago (could it really be two years already?), and as that post noted, there were significant limitations in our first 1-click widget. We had to take on some higher priority projects that delayed our ability to address the limitations for a while, but as you'll see, we're now solidly back on track. 

While I don't work full-time on downloads any more, I am still the Product Manager of the Sun Download Manager (SDM) and of course remain very well-connected with the Download Team. As such, I was asked by Alfred Chen, the current Download Team Lead, if we could try out the latest version of the "advanced download widget" (ADW) on the SDM download page. SDM remains one of our most popular downloads so would get plenty of traffic for "real world" experience and also met the criteria for the new ADW. I was glad to say "Yes," and we went live with it several weeks ago. 

One of the main limitations of the original 1-click release was that it worked for one file and one file only. The new ADW now accommodates multiple platform and language options so is much more versatile.  It even "auto-detects" your platform and language and will pre-select the appropriate option when available. Once platform (and if available, language) are selected, it automatically pushes the correct binary for 1-click downloading. Here's a screen shot of the new widget:

You can try it out for yourself by downloading SDM.

There's more work to do, however -- ADW still does not support optional or required registration, so that's the next set of enhancements. They are underway and if all goes well will be released this summer. I have previously touted the virtues of optional registration, and there are several places on Sun now where we're using this with very good results. For example, on the Glassfish Download Page, there is an optional registration "pop-in" widget that comes up when you click the Download button. This illustrates a new twist on "optional download registration," as it includes the added incentive of receiving an informative white paper for registering. Being optional, you can just close this out and the download will start automatically (there's that 1-click functionality again!), but many customers are taking advantage of the white paper offer. So they get a white paper and the Glassfish download, and we get valuable customer information. 

While this new model is working great, the Glassfish implementation is frankly a "one-off". What the next version of ADW will provide is a standard way of doing this across Sun sites with minimal effort. I'm really looking forward to this new functionality and think it'll be a great win for both Sun and our customers.  It continues the march towards our goal of seamlessly integrating the download experience directly into the web pages you're on, rather than forcing customers to "move" between the web site and the "download system." 

In my last post, I displayed an image of a new display widget for personalized content on sun.com web pages. I noted it was a draft, and with good reason -- nobody was ready to sign off on the design at that point! The main issue was that it took up a considerable amount of the most valuable "above the fold" space on sun.com pages, and the content owners weren't enthralled with having their content pushed down the page. Our excellent design team mulled this over quite a bit and came back with what we feel is a great compromise design. (I am a big proponent of "compromising" and try to do so whenever reasonable, so that all parties feel relatively satisfied with an outcome. That's probably one of the reasons I'm about to celebrate my 24th wedding anniversary! But I digress...)

Here's a wire frame of the new solution. ("Wire frame" refers to more of an "outline" picture of the solution, not the final coded and realized design. Ignore the numbers -- they cross-reference content in a design specification. While we have the final design complete and coded, I'm not at liberty to preview it publicly just yet. Suffice it to say this wire frame has been translated into a very cool new widget!)

So here we see the recommendations as more of an overlay rather than a full page-wide component. This offers the immediate advantage of making it readily clear there is content behind the widget (and in fact you can still see the whole left hand top of the page), and no content had to be pushed down the page. It's also attention grabbing, and we do want customers to notice the personalized recommendations we have for them. We keep the fashionable "carousel" functionality so that you can click to see more recommendations, and we make it readily apparent how to close the recommendations to view the content underneath. Once closed, it'll remain in a closed state until the user opens it again or until the cookie that tracks this preference expires or is removed.

I'd be remiss in not acknowledging and thanking the design team that worked so hard to find not just a workable solution, but what I think is an excellent solution as well: Margaret Brown, Chris Haaga, Sara Shuman, and Andrew Payne.

Now that I've whet your appetites, you're probably wondering when this will be live. Sorry, I can't publish dates for functionality that we're still building, but suffice it to say we're making great progress.

I wrote an intro previously on SMILE, the Sun Machine Learning Engine, our new personalization system that utilizes predictive analytics. It went live in January, 2009, and we've been carefully measuring results since. So far, we are seeing higher click-through rates for SMILE served ads versus default ads, which is what we hoped would happen. But overall click through rates on these ads, whether personally targeted via SMILE or simply default, are very low, so we need more bang for the buck. We believe that the small ads served by SMILE in the right-hand navigation are easily overlooked -- many users simply ignore banner ads and/or the right hand column, concentrating on the central content of interest.

So our next step is to provide personalized recommendations in the body of the page. Here is a draft mock-up of a "carousel" type approach we hope to use:

There are numerous challenges with this approach, as the page real-estate "above the fold" is highly valued, but that's where we need the recommendations to be if we really want them to be noticed and of value. A likely approach will be to make the carousel collapsible, so users can close it up if it's in the way, and then open it to look for personally recommended content. We'll also provide a new "All Recommendations" page that is solely dedicated to showing all recommendations for visitors.

We did a usability study recently to test these concepts and were pleased with the results. Customers told us:

• They like what we're planning. We have literally millions of web pages, so if we can use our analytics to help customers quickly find the products, services, and content of most interest and value to them, we are saving them time and making them more efficient. Definitely a "win win" for all of us.
• They would likely not notice the recommendations if they're not centrally located high on the page.
• They don't have serious privacy concerns. Because we are an enterprise business site, they expect us to use analytics to improve the experience, and they'd appreciate the benefit. They would look at things differently if we were their financial institution, for example.
• If the recommendations are not accurate or useful, they will quickly learn to ignore them. So the onus is squarely on us to ensure our system is up to the task of making accurate, valuable recommendations that meet our customers' needs.

This is an ambitious undertaking involving a whole bunch of teams across Sun, from Design to Publishing to Engineering to Analytics, so it's at least a few months away.  I'll keep you informed of progress and of course announce when you can check out your own personal recommendations!

I've actually been on my new job for a number of months now, so it's time I start writing about it! After 10+ years working in the ESD (electronic software distribution) space here at Sun, it was time for a change, and so I took on a new challenge in the eMarketing realm. I am overall Program Manager on the sun.com web team for an initiative we call "Contacts to Revenue" (C2R). The essence of C2R is that we have zillions of web visitors, and especially downloaders, and we need to do a better job of identifying contacts and hopefully "nurturing" contacts into customers. There are several primary initiatives now underway.

First, we're creating new content we feel is of value to our customers and asking them to login to receive it. (If you don't already have a Sun Online Account, then there is a brief registration form that must be completed.) We feel this is a fair value exchange -- you tell us who you are, and we provide valuable information in return (examples include white papers, blue prints, webinars, and downloads). Here's an example:  Deploying Hybrid Storage Pools with Sun Flash Technology and the Solaris ZFS File System.

Second, I want to discuss Project SMILE -- The Sun Machine Learning Engine. We're building a new analytics engine to help provide more relevant content to web visitors. I'm the Project Manager for SMILE and pleased to say we went live with "phase 1" in mid-January. For the initial release, our goals were modest, focused primarily on improving the relevancy of small advertisements displayed on our main web site. We used a good ad banner server program previously, but it lacked sophisticated segmentation and targeting capabilities, so we built our own. If you are a return visitor to sun.com, we may serve you small ads that are more targeted to areas in which you've indicated an interest. All the ads are unobtrusive, appear in the right hand column, and are of this format:

 

For SMILE Phase 2, we're working on the ability to more proactively serve "Recommended Links" and other content of value based on web visitor's interests in our products and services.

We realize that asking for a login to obtain content and using machine learning and analytics to serve customers may be of concern to some. However, the bottom line is that we're in business to make a profit for our shareholders. Times are tough (duh), and it's imperative we do a better job of converting our web visitors to customers in order to grow our customer base and increase revenue. It's not a one way street -- we will continue to provide exceptionally valuable and unique content in return as well as striving to make relevant content more apparent and easier to locate on the web. 

In conjunction with these changes, we have recently updated our Privacy Policy to more accurately reflect how we'll be using customer data in support of these programs (among other updates). Please read it if you wish to be fully informed on this subject. (We also make it possible to opt-out of web personalization if you wish.) In any case, as this blog post and the new privacy policy illustrate, we're openly communicating about these new programs and changes. Our intent is to better serve our customers and our shareholders.

(This post covers the last day of the three day conference. Also see: Day 1, Day 2.)

The final day kicked off with a "web meets food" theme as Michael Pollan (noted author focusing on "the places where the human and natural worlds intersect") took the stage. His was not a cheery message for first thing in the morning. He noted that our current methods of food production are large contributors to green house gases, with a heavy reliance on fossil fuels. He feels we can and must turn agriculture into part of the solution, not the problem, and that we'll all be much better off and healthier if we "eat sunshine, not oil." A big part of the problem is we're locked into a cycle whereby the more we process food, the more profitable it is, but the less nutritious. 

He suggested ways the web can improve the situation. First, make the food system more transparent. If we really knew how food was made and its impact on the environment, it would be a major catalyst for change. Secondly, the Web is the perfect place to form communities of like minded people who can focus of enabling sustainable alternatives as well as cultivating and identifying local food sources. Just the cost and impact of transporting food is huge, that's why farmer's markets may well be a model of the future (back to the past?). Is it smarter to bake butter cookies in Denmark and ship them to New York, or to ship the recipe to New York and bake them locally?

Back to more mundane matters (like making money), Richard Rosenblatt of Demand Media Inc. gave a very impressive talk and demo of his company and its new Pluck on Demand product. They make it incredibly easy to add contextual content and social media to any web site, in a self-service model. (I saw a guy in the row in front of me on his laptop doing this real-time to his web site once he saw the demo!) Richard explained some of the driving concepts behind his vision and services:

• In a search driven world, audiences fragment but then reform around passionate verticals. Provide the tools to engage them quickly in your communities once they find you.
• Social media will be the key success factor for every web site.
• Look to grow organically via community, not just search.
• Create content at scale, leverage the community content model. 
• Support it with algorithmic, ROI driven content placement.
• The right content + social media = user engagement and revenue.

Reinforcing that "cloud computing is all the rage," we had our second panel on the subject, featuring Padmasree Warrior (CTO, Cisco) and Shane Robison (CTO, HP) in a conversation with Tim O'Reilly. They started by discussing the changing role of the CTO, which is more focused now on strategy and the future of business technology. (That was contrasted with the CIO role which they described as more operational, often implementing what the CTO recommends.) And then onto the benefits of the cloud...

• It's the next evolution of computing.
• Applications are abstracted from physical location.
• We haven't hit the tipping point yet. We still need "hybrid clouds" that incorporate standards and can interoperate with "private clouds." Think "inter-clouds" as opposed to inter-net.
• HP wants to provide "infrastructure as a service."
• Clouds can help with Green IT. Think of instrumenting the environment through the cloud.

For a conference that started the day after the presidential election, we finally got down to some politics. Enter author and mega-blogger Arianna Huffington, San Francisco Mayor Gavin Newsom, and high-profile political consultant Joe Trippi. It was a lively and spirited discussion, with everyone in a pretty good mood as you might imagine. A few take-aways:

• 1960/JFK -- the first "television" presidency. 2008/BHO -- the first "Internet" presidency.
• Huffington stated something Al Gore would also say later in the day: Obama would not have won without the Internet (for fund raising, communication, and community building).
• There was a concept that the 'net "surfaces the truth," and perhaps Right and Left are outdated terms. What's important is where the truth lies (and it's not necessarily in the middle). 
• This could ultimately affect the future of our two party system.
• You can "fake it" on a 30 second TV ad. It's harder to do 24x7 on the Internet.

I thought Elon Musk provided the ultimate example of living your dreams. He said that during his college years he felt there were three important problem areas for focus: the Internet (he was a founder of PayPal), clean energy (see Tesla and SolarCity), and space (SpaceX). He had an unassuming manner about him, but you can see he's really putting his time and money to work on these problems. Compare that to making your fortune, then kicking back on your own tropical island...

"The Platform" had some very interesting dynamics when you consider the fierce competitors sharing the stage: Moderator Max Levchin (Slide), Vic Gundotra (Google), David Treadwell (Microsoft Live), Amit Kapur (MySpace), and Elliot Schrage (Facebook).  Fortunately, peacefulness reigned, though there were a few testy moments (mostly Microsoft v. Google, as you can imagine). The main theme was how to evolve the platforms and support the developers:

• Enable developers to build viable biz models while increasing customer engagement. ("People like to share information.")
• Join people, places, devices, & applications.
• The Platform must benefit all three to survive: provider, developer, and customer.
• How do you deal with fast evolution vs backwards compatibility (especially for your 3rd party partners)?
• Build strategic alliances
• Be transparent and open to feedback
• Provide tools, resources, and support
• Rely on the evolution of the Internet to push forward the platforms built on it.

Bob Sutor, IBM VP of Open Source, said the world is at an "inflection point" where we must start using IT power to start solving world problems. Make big bets, public/private partnerships, and open standards are key.

I was fascinated with Shai Agassi's talk about building an "electric car network" (see BetterPlace). This is truly a big bet! Imagine an electric car that you pay for by the mile, and you don't own the battery. Need a charge? Pull into a "filling" station -- they remove the dead battery, put in a charged one, and off you go in less time than it takes to fill your tank. How fast this can happen is a "matter of policy" (and some deep pockets I presume), though he said they are actively working in Israel to release the first network.

"Track Me" was a panel on geo-location and some of the cool things happening now and possibly in the future. The panel was composed of moderator Brady Forrest (O'Reilly Media), Greg Skibiski (Sense Networks), Ted Morgan (Skyhook Wireless), April Allderdice (MicroEnergy Credits), and Rich Miner (Google).

There was a fascinating demo of the kinds of things Sense Networks can calculate and present, based on "analyzing massive amounts of real-time location data to understand and predict aggregate consumer trends." You'd be surprised by the correlation between the time and location of taxi cab trips in San Francisco and the rise and fall of the stock market! They have a lot of data and are still trying to figure out the meaning of some startling results and trends, plus ways to "segment" based on location.

Miner from Google talked about building location into the platform, then letting developers build upon it. Privacy was an obvious concern that came up. People need to own their data and have the ability to control, review and delete their location data. What is the geo-location equivalent of a cookie?

Rebecca MacKinnon (Global Voices)  and Isaac Mao (currently a research fellow at Harvard Law) talked about overcoming censorship on the Internet (mostly in China), and its impact on closed governments. McKinnon talked about the difficult place Internet companies find themselves, wedged between governments and individuals. Mao introduced his concept of "sharism".

And with that, it was time for Al Gore, the exciting grand finale. I really enjoyed seeing him in person, and he held his own quite well, showing an impressive capability to talk tech and appeal to the audience. As mentioned, he reiterated the message that Obama would not have won without the Internet. He talked about Current TV and their efforts to "democratize the TV medium." He used some powerful analogies:

• The invention of the printing press almost 600 years ago "freed" information for the first time from the elite and created a new information ecosystem. Television "refeudalized" the control of information into a few elite hands. Now the Internet will democratize information again.
• We are moving to a time when Web 2.0 will be taken for granted, like a "fish in water." He challenged us to move to "World 2.0."
• Finally, a great analogy that really resonated with me (showing my age, I guess):
  JFK took office in 1960 and said we would put a man on the moon in 10 years. It seemed incredibly ambitious and perhaps unattainable, but it was done in a little over 8 years. He challenged Obama and all of us to take on a similarly ambitious goal: Make 100% of all electricity in the US from renewable energy within 10 years. If we could go to the moon back in the 60's, we can do this in the 2000s!

The conference came to an end with a final reception (i.e. yet more drinks and hors d'oeuvres), and everyone I spoke to thought it was an excellent event. I agree. If you've read this far, I thank you and hope you found this overview provided a decent taste of an unusual and exceptional conference.

(This is a continuation of my prior post, "Overview of  the Web 2.0 Summit 2008.")

Day Two started with Paul Otellini, President and CEO of Intel. He focused on two big areas of opportunity that Intel sees. The first was "enterprise" (or professional) social networking. He showed a thought-provoking demo of the unified enterprise community network of the future -- tightly integrated communication tools, professional learning management, robust search and RSS feeds. The demo focused on a fictional Intel new hire in Asia and how such an enterprise web community would smooth her integration into the company and also accelerate her learning and effectiveness. Then he said something like, "Of course, these tools don't exist yet..." 

The second big opportunity was the "personal Internet," where the machine brings relevant and timely info to you rather than you having to seek it. This led to another demo, this time of the smart mobile device of the future. Features included machine translation (written and verbal) in the palm of your hand, instant commerce, lots of geo-awareness, etc. It demonstrates Intel's focus on bringing the power of the web to mobile devices.

Next up was Kevin Kelly, a long time reporter on the culture of technology who helped launch Wired Magazine. This is more or less what his first slide looked like:

Care to guess the meaning?

Turns out it had been exactly 6,527 days since Tim Berners-Lee transmitted the first web page. The Internet (r)evolution since then has been amazing, but Kelly really said we need to hasten the next phase, the so-called Semantic Web. It's about awareness and linking of data. First we linked computers, then pages, and next we need to link data. 

So what will the next 6,500 days bring? Kelly feels it'll go way beyond "the web, only better." He spoke of one large machine of interconnectedness, the Web Operating System (I couldn't help thinking of Sun's seminal mantra, The Network is the Computer). The web will "own" all data -- if it's not part of it, it doesn't exist. All devices are connected, processed, and structured. One media platform for all. Everything is always on. "Believe in the impossible."

This could be an awesome thing, though it does have that "big brother" ring to it as well. Let's hope the "security guys" get it right.

The plethora of iPhone enthusiasts in the audience were entertained by a talk with Ralph De La Vega, CEO of AT&T Mobility and Consumer Markets. Let's just say he's pretty happy with how things are going. 

The web can and will play a huge role in the future of health care, as discussed in the "Future of Health" panel. Lots of talk of genetic testing and online communities. Check out 23andMe, then order your own genetic profile for a mere 400 bucks.

Appearing in full battle fatigue, Army CIO Jeffrey Sorenson showed the military side of social media. The Army lags behind the speed of commercial innovation and is just rolling out Web 2.0 now. Security concerns are paramount of course. But we saw interesting demos of "Battle Command Knowledge System" and "Command Post of the Future" -- web based systems that will greatly enhance our military capabilities and speed/accuracy of communication.

The "Future of Media" panel provided an engaging discussion featuring Evan Williams (Twitter), Joel Hyatt (Current TV), and Ken Auletta (The New Yorker). I remember coming out of my first Web 2.0 Expo in 2005 thinking I had to start a blog (which I did). This time, it was all about Twitter, so I'm going to give it another try too! (I'm admittedly not 100% convinced, but time will tell.) They talked about Twitter changing how people connect --> will it actually change our culture over time? 

Today we are enabling "conversations" in every possible format, though the questions remain on how to monetize these new interactions. (They didn't have the answer, sorry.) 

I found Hyatt's thoughts on Current TV provided a refreshing perspective:

• Many people innovate, then look for a business model. They did the opposite.
• They're developing "user generated advertising" (he showed a clip of a Toyota commercial made with a hand held video camera). Their study shows these ads are "preferred 9 to 1."
• The networks are focused on moving TV to the Internet. Current is focused on moving the Internet experience to TV. 

Next speaker, Saul Griffith, a rocket scientist type for sure, reminded us that we all contribute to the energy crisis. We all have to help solve it. He's built a very clever web site that lets you easily determine how much energy you use in a day. He suspects we'll all be surprised and ultimately better conservationists after finding that out.

I have a 17 years old son, so couldn't help noticing the physical similarities to Mark Zuckerberg, founder of Facebook. Yes, he's young, but he was very composed and well-spoken. (BTW, it's been a big week for 24 year olds around here!) He took some heat for the closed nature of the Facebook platform, but that's changing, and he said there is a natural progression from closed to open systems over time. (That's a key part of Sun's open source vision, so let's hope he's right.) He spoke of a long term goal to enable people to share information any way they want on any device. How to make money? Think of how to best entice customers to share information, then deliver "engagement ads" that are highly effective. 

A very high powered (pun intended) panel was up next on Cloud Computing -- Paul Maritz, CEO of VMware, Marc Benioff, CEO of salesforce.com, Kevin Lynch, CTO of Adobe, and David Gerard, Google. You know this is one of the hottest topics of '08 and extremely well covered, so I don't think there was a lot of news coming out of this. Some interesting concepts were mentioned though:

• Information will outlive devices and applications; the cloud becomes the "information bank."
• Providers have to protect, enable access, and add value to information stored with them.
• We're hitting the "fourth generation" of software, getting back the richness of client-server applications (which we lost with Web v1.0) but keeping the power of the Internet.
• They need to provide the tools to enable the movement of apps to the cloud.
• We need to break down the disconnect between rich customer experiences and pedestrian enterprise experiences.
• What if you had a rich Flash interface to SAP? (Adobe wants to be the "interface to the cloud.")
• It doesn't have to be a low margin business. Differentiate by providing unique value and/or a large ecosystem.
• Clouds are still proprietary, though. More value will be created as they interoperate and provide new services. (We're seeing this now with Salesforce's integration with Google Apps, for ex.) 
• Standardization is needed, as well as new data models that break down data silos.
• Eventually, consider we'll have "personal clouds" as opposed to multiple friend lists on separate web sites.

Shifting back to the future of advertising, there was a talk with Jack Klues of Vivaki -- they control billions and billions of dollars in ad budgets for some of the world's biggest brands. His message for weathering the economic storm:

• Maintaining and protecting brand is at least as important during a downturn as any other time.
• You can increase share as media buys become cheaper.
• Don't lose sight of what people care about; keep on message and brand.
• Continue to look at how to optimally blur the lines between advertising and content (ethically, of course...)

A "Launch Pad" event was next. Five start-ups gave their elevator pitch to a panel of VCs. The audience voted on our favorite by texting votes which were tabulated and projected real time. I agreed with the winning choice, GoodGuide. Great concept and valuable info -- take a look. (No real losers here, all were of interest but not sure they'll all survive: Predictify, Sungevity, Everyscape, Qik.)

The day closed with a Music discussion with Chris DeWolfe, co-founder of MySpace, and Edgar Bronfman, CEO of Warner Music. It was a bit too much focused on the launch of MySpace Music I thought. They said the music industry will be "very different in 5 years," but not necessarily how. And with that, another long but stimulating day ended.

Continue to read about Day 3...

I felt fortunate to attend my third Web 2.0 Summit, November 5-7, 2008, in San Francisco (thank you, Andreas!). I really thought it was an excellent event and fairly different from the prior ones I attended (here are my comments on the 2005 Conference and 2006 as well.) What was most different was the strong and well-coordinated thematic focus on "World Meets Web" -- it truly permeated and unified the sessions and discussions. The concept behind this is that we've come a long way with the web and it has become a major influencer on the global society. How do we use this new found medium for the betterment of the planet and to deal with all the issues we face? It couldn't have been a more timely theme,  nor punctuated better than Al Gore's closing comments, as he noted Barack Obama could not have been elected without the web! Talk about web meets world....

The format was generally the same as previously. Wednesday morning provided the only workshops, where you have the choice of attending several tracks that run simultaneously. (This is where you typically wish you could be in two or more places at once so you don't miss anything.) Starting Wednesday afternoon and through the conference closing late Friday, we all assembled in a large conference room for the engaging interviews with true industry leaders as well as the “high order bits” (short, 10-20 minute talks by various domain experts).

There's lots of info online of course about the Summit, and in the interest of time and efficiency, I'm not going to bother deep linking to everything. Here are the bios for all the speakers, and many of the presentations (and video) are online too. My main goal is to share some of the insights I captured from the various speakers and events.

For the workshops, I chose topics related to presentation and advertising, as this relates closest to what I'm working on now (developing new web personalization and ad serving capabilities for Sun).

The first talk was a bit of a strange start, as a little ways into it I realized it had nothing to do with the web (at least directly). Go figure. Nancy Duarte's topic was "Telling Meaningful Visual Stories" and was really about creating compelling presentations. It was targeted partly at least to the many entrepreneurs that attend, with lots of guidance on how to make compelling pitches to VCs. But I think there are lessons to be learned for presenting info on the web as well. Some key takeaways:

• To connect with the customer, become a story teller. Share experiences, be vulnerable perhaps, believe that your own story can touch customers. (This is something that cannot be outsourced!)
• Consider your audience first and how to reach them, then develop the content. A presentation is not a “document”.
• Break out of the mold – start with a blank page, not a template. Change your environment, get a different perspective, collaborate and brainstorm on ideas. You want to inform, inspire, persuade.
• Avoid complicated architectural diagrams, as they don't resonate. "Tell" your architecture as a story.
  
• Use metaphors. Include conflict to grab attention. Produce ideas and stories, not slides. Be transparent, be different. Your stories can change and influence!

The next workshop I attended was "Consumer Brands Tackle Marketing 2.0," with some high profile marketers from leading companies. Some nuggets from that panel:

• Technology changes, people don't.
• Social media offers new opportunities for gaining customer insight, if you know what to do with it. Consider "perpetual marketing," as the engagement doesn't end. We're still evolving how to use social media consistently in marketing programs.
• The audience defines the message. Good creative still wins, technology can't overcome crappy messaging and content.
• Traditional media has the reach but many in the audience won't be interested. For example, if you advertise Pampers on national TV, what percentage of the audience cares? This is really where we can differentiate advertising on the web, by targeting interested and relevant audiences much more closely.

As I mentioned, we're working on more personalization initiatives here at Sun, and one outcome is that we're bumping up against our own privacy policy. There's a fine line there about what we can and cannot do. I asked the panel about how they deal with it, and mostly got shrugs! David Knox from Procter & Gamble did say they had just changed their privacy policy to enable global opt-in/opt-out on their site, instead of by section as it was previously. Their approach was to email every single customer and inform them of the new policy.

So it seems everyone is trying to deal with this, and there is no perfect answer. I think the best thing is to be transparent with what we're doing. We need to revisit Sun's privacy policy to see how it can evolve to match the evolution of our web and personalization capabilities. When we do make a change, we'll need to communicate it appropriately.

The last workshop I attended had the not-so-helpful title of "Please - Not Another Network Panel." Sponsored by Yahoo, it actually focused on web advertising and ad networks. I'm pretty new to this area so didn't follow all the subtleties of the conversation about how these mega-networks work, but I managed to learn more about that side of the industry.

And with that, it was lunch time, then we headed in for the start of the mega-conference. The focus on web meets world was clear from the start, as the first interview was with Larry Brilliant, head of google.org, Google's philanthropic arm. Google committed from the start to putting 1% of revenue, profit, and resources into philanthropy. We'll see this theme repeated throughout the conference – the web has made you millions (or billions), so how will you use that to make the world a better place?

Google is focused on three key initiatives:

• Inform & Empower -- mostly a third world focus on empowering communities to get better information and accountability from their governments.
• Predict & Prevent – Early disease identification and prevention. This touched on another recurring theme, “instrumenting the world.” We have great new sensor capabilities and can tie the sensors together via the web for almost real-time, unprecedented insights into the global ecosystem.
• Renewable Energy -- specifically, we need a form of renewable energy that is cheaper than coal. At that point, there is no financial barrier to adoption.

Bottom line: "It's time to use technology to make the world a better place."

Mary Meeker gave her usual dizzying blitz of a zillion financial slides in about 15 minutes. It's pretty hard to digest that quickly but my takeaway was that, while the economy may be tanking, there is still plenty of growth and opportunity for Web innovators.

Rajesh Jain from Netcore Solutions (see his blog at emergic.org) explained how there are about 10 times as many people with cell phones in India as have Internet access. As a result, SMS is huge, offering ease of use, low cost, and easy value added services. There's a huge direct to consumer reach. He used the term “invertising” for invited ads in SMS messaging.

Up next was super-VC John Doerr, who helped fund Sun (amongst many other companies) years ago. He talked about Obama being the first President who will have a CTO, and suggested (Sun co-founder) Bill Joy would be a perfect fit. He felt energy research is our #1 priority and expressed a big concern that we're not graduating enough scientists and engineers. He also felt it was absurd that we train foreign students, then kick them out of the country once they're trained. I loved the way he summed it up:  "We should staple a green card to their diplomas!" Makes sense to me.

Doerr offered up survival tips for start-ups during the economic meltdown (which he described as more than bad mortgages, it's a "crisis of confidence"). I think much of this is great advice regardless of how big your company is:

• Hunker down, take on the long view, focus on your core
• Act now, and with haste (before funding gets cut)
• Protect the core of your business (don't lay off the key people!)
• Have 18 months worth of cash in the bank
• Defer facilities investments
• Re-evaluate R&D priorities
• Renegotiate all your contracts
• Everyone in the company needs to SELL
• Give equity instead of cash
• Secure the cash (i.e treasuries, somewhere safe)
• React quickly to leading indicators and KPIs
• Over communicate, don't sugar coat bad news

He was quite bullish on the iPhone and said they're seriously investing in it as a platform. He likes cooliris.com. He likes companies focused on the conversion of IT and "green technology."

Next up, Tony Hsieh, the personable CEO of Zappos.com. I think the Zappos story is well-publicized so won't go into details, other than to say it's quite impressive what they've done. This web-based "shoe store" is now a $1B business and expanding quickly beyond footwear. A few key points:

• The company is "powered by service".
• The goal is to WOW each & every customer.
• Business is built by repeat customers and word of mouth.
• It's good to have people call your customer support line. There is no better branding opportunity then speaking with your customers.
• Company culture = company brand.

Larry Lessig, the outspoken Stanford Law professor, talked about the current failures of our government. He says it's come to the point where the Number One job of our elected representatives is to get "tenure" (i.e. raise enough money to get elected over and over again). Money destroys trust. We must change congress to fix our economy.

Day one ended with Yahoo CEO Jerry Yang on the hot seat, grilled somewhat mercilessly (but fairly, I think) by conference promoter John Battelle. Jerry took the hard questions graciously and stood his ground. He explained Yahoo's strength as a consumer brand that enables people to get what they want from the 'net. It's the starting point.

He talked about recent efforts to open up the Yahoo platform with the Yahoo! Open Strategy, enabling third party apps to integrate with Yahoo services. There is opportunity for all in the creativity that will come out of opening up the platform.

What I hadn't realized was the large press presence and how they would pick up on what was being said. While discussing Microsoft, Yang said he thought Microsoft should buy Yahoo. I was surprised when this casual statement during a conversation turned into national news.

As you can see, it was quite a stimulating Day 1!

As an aside...
I was amazed at the proliferation of iPhones in the audience (and Blackberries -- the lady sitting next to me had one of each on her lap) and somewhat dismayed at how many people were looking down constantly with an eerie blueish light on their faces from the phone displays. It didn't seem to me they were paying attention to the conference, and I wonder about this newest contributor to attention deficit disorder. But perhaps they're just better at multitasking then I am.

Continue to read about Day 2...

Here's a "best practice" that is a source of never-ending debate: Should we require login/registration in order to download our products? There is no right answer, but there are plenty of pros and cons.

Sometimes, we have no choice but to require login in order to comply with government export regulations. (For example, if a product contains some types of strong encryption, we must keep a record of who has obtained it.) An obvious reason in favor is that knowing who downloads our software is of great value in follow up marketing initiatives as well as demographic analysis of our customer base. It's not all a one-sided equation though. For our customers who login first, we offer a useful download history function, and we also automatically provision our My Sun Connection portal with a list of products of interest to you. (This automatic portal provisioning is set up for many, though not all, of our downloads.) It's very handy, as you can click on product names and quickly see a valuable list of related resources and information. (I encourage you to check out the portal if you're not familiar with it. Once there, be sure to click the "My Products" tab to see how we aggregate product information from many resources in one place!)

The other side of the coin is that requiring login or registration before downloading definitely hurts the conversion rate. Using our web analytics, you can see users drop off with every extra click, and the largest abandonment is at the login/registration step. In many cases, the value of getting our software into the hands of our (often first-time) customers as quickly and with as few roadblocks as possible trumps the value of collecting user information, so we do not require registration. 

With such conflicting agendas, you must be wondering by now what I'm going to recommend as the best practice? Well, that's easy, don't use either -- use optional registration! This feature provides a middle ground that works for most products -- just give customers the choice. If they want to login to gain the benefits offered, we welcome and appreciate it and believe it's worthwhile. And if the customer just wants the software as quickly as possible, there's no need to register, just click Continue. Here's what you'll see when you request an "opt reg" product on Sun Download Center:

For those who do choose to login first, we find a better quality of data since they are doing so of their own volition, and it allows us to build a stronger relationship with the customer. Of course, we have to face the facts -- a majority of customers will skip the login step and click Continue. For that reason, we highly recommend building optional product registration into the product's installer as well. Once you have committed to installing the product, there's an even stronger opportunity to build the relationship and help support the software on your system.

Fortunately we built our download system to support all three options (required, optional, none) easily, so we can meet whatever the product team wants with the flip of a bit in the product database. However, our recommendation remains to use optional registration whenever possible.

This is the third in a series of download best practice posts.

• Download Best Practices, Part 1
• Download Best Practices, Part 2

After more than 10 years of focus on download systems here at Sun, I was ready for a change of pace (as you might imagine). Once we got our new download system out the door, my manager was much more willing to consider a change for me, and then a great opportunity arose. I'm now managing a new project that is developing a machine learning engine for better targeting our offers to our customers. While the obvious intent is to benefit Sun by increasing the uptake on our products and services, we feel it benefits customers as well by providing more relevance to the ads and content we provide you on our web sites.

It's pretty much a whole new world for me, but I'm enjoying the challenge and learning about a whole new area of analytics. It would be a shame to ignore the experience I've acquired in the ESD realm, so I'm still participating in download strategy discussions and helping out where I can. I may well blog further on the subject as time allows.

But in any case, I am happy to see two new blogs about downloads and ESD that have sprung up recently. As I wrote in my very first blog posting three years ago, "I haven't found many resources for us 'ESD Professionals.' I'm not aware of trade journals nor web sites dedicated to ESD. I always enjoy meeting and speaking with my peers at Sun and other companies but wish there was more interaction, more sharing of best practices, and the myriad other benefits that come from knowledge sharing in a particular field. This blog is intended to be a small outreach..."

Well, I can't say I've seen an explosion in content, but I heartily welcome two new contributors to the ESD blogosphere!

• The overall Project Manager of the new download system was Alfred Chen, and he's the guy in charge now that I've stepped into a new role. Alfred is leading our Download Business and Strategy Teams, and we are very lucky to have him in those roles. Working with Alfred very closely for many years, I can only say he's a tireless contributor with exceptional knowledge, skill, and dedication. (And he used to be the manager of our download engineering team, so he's really one of those great persons with deep IT and Business experience.) So, to get to the point, Alfred's started blogging lately (about a variety of subjects) -- check it out.
  
• OMS Safe Harbor is one of the few companies out there with a major focus on electronic software delivery, and they've built up an impressive set of capabilities, products, and services. They recently let me know about their new ESD blog, and I've already read a number of interesting articles there. Great to see this outreach from OMS.

I hope you'll find these new bloggers of interest!

Mozilla and Firefox have garnered a lot of well-deserved publicity for their new Guinness record for most downloads in a day. This raises the question (again) of what constitutes a "download" and how is it measured? (I first touched on this subject a couple of years ago, writing "When is a download a download?".) I was glad to see in the above referenced BBC article, "The official figure was confirmed after logs from download servers were audited and checked to ensure duplicate and unfinished downloads were not counted." So that implies "attempted" downloads weren't counted, only completed. That's good, as I'm sure some stats we see reported don't make such distinctions. 

So with Guiness in the mix now, and with all the focus on downloads in the free & open source software world, I propose it's time for an agreed upon standard on how the number of downloads is reported. (To be honest, I'm uncertain if this could ever be a "formal" standard and who would govern it -- please let me know if you have any suggestions.) For starters, here's what I propose:

• Simple case: One Product = one file
• This case means that the entire product is contained in a single downloadable file. If it is downloaded successfully, the customer can install and use the product.
• The suggested measurement is the one generally used -- the last byte of the file is delivered to the customer.
• Note this may not be 100% accurate. Threaded download managers can conceivably download the last byte of a file and then fail an earlier byte range before the entire download completes. But this is an edge case. It would also be extremely resource intensive to have to match up all the byte range requests in a threaded download scenario to "prove" that one user completed the download. Thus, I think the "de facto" standard of last byte delivered is acceptable.

• Complex case: One Product = multiple files
• First, make the distinction between required and optional files:
• Required files are required to install the product. There may be one or more required files as part of the download. 
• For example, on Sun Download Center, there is an option to download Solaris OS as 5 CDs. If you do not download all five, you cannot install the product. Thus, all five are required files, and each one must be downloaded to install Solaris.
• Optional files are not required to install or run the product. Optional files may consist of "plug-ins,"  "add-ons," or "modules" that add functionality to the base program. Also included would be documentation, checksum files, additional language packages, and the like.
• The measurement of a completed download in this case is that all required files are downloaded by the same customer. (This can be a challenge to measure, but it can be done in most cases with a good degree of accuracy.)
• Optional files do not count and should never be reported as "product" downloads.

• In either case, "Attempted Downloads" should never be reported as a product download. 
• For a single file product, an attempted download is when the user starts the download but does not receive the last byte of the file.
• For complex products, an attempt is when any individual required file is not completed and/or the customer does not successfully download all required files.

I'm going to list three more of the download best practices that we encourage all our product teams to use as they develop products for electronic software distribution (ESD). This is the second part of a continuing series I'll publish from time to time. (You can see Part 1 here.) Some of these suggestions may seem like "no-brainers" (at least once you understand the basics of ESD), but they need to be stated and documented nonetheless and are certainly helpful to teams that are just starting out to prepare downloadable software.

1) Compress the files as much as possible. Obviously smaller files are cheaper/faster for everyone to download, and faster delivery means less time for things to fail and higher completion rates. Generally, we recommend using zip compression, as it's pretty much a de facto standard at this point (and is supported on Solaris OS as well). There are many free and open source distributions product teams can use, such as Info-ZIP, bzip2, and 7-Zip. You should experiment to see which one provides the best compression (in one test here, 7-Zip outperformed the others by about 20%, but your results may vary). There are also proprietary compression programs ("NOSSO," for example) that can greatly outperform "vanilla" zip compression, but they come with a price tag. If your download size really really matters, they may be worth investigating.

2) Use standard MIME types for file name extensions, for example "filename.zip" and "filename.exe." Web servers, browsers, download managers, and many other pieces of the Internet infrastructure rely on standard MIME types to ensure download transactions work correctly and consistently. So, don't release files without a MIME type extension (for ex., use "README.txt," not "README"), and don't make up file extensions or use non-standard ones, such as "filename.aa," "filename.bb," etc.

3) For large files, offer alternatives. Not everyone can successfully download a large file, especially ones larger than 2 GB. Other choices should be provided, such as a segmented ("chunked") version of the large file, CD images instead of a single DVD, and/or the option to order free or low cost CDs or DVDs. See my earlier post on "Breaking the Large File Barrier" for more details.