Joost, a much-vaunted online video startup, today announced that it will offer a white-label video hosting platform, thus entering a crowded market littered with the carcasses of other failed video hosts. The company is also losing its famous chief executive, Mike Volpi, whom it’s replacing with Matt Zelesko, the current vice president of engineering. And it plans to cut a portion of its workforce — between about 70 of its remaining 90 employees, according to Advertising Age. It also shut down its office in the Netherlands.

When I read about all the planned changes at the company earlier today, the first thought that crossed my mind was: Stick a fork in it; Joost is done. After all, this whole white-label video strategy is like a leaky lifeboat in the middle of the Pacific Ocean. The NewTeeVee crew sums up the situation very succinctly: “Becoming a white-label video provider was what a business did when all other strategies failed.”

As someone who has followed Joost from its very inception, when it was known as The Venice Project, I’m amazed at how badly it’s stumbled. It shouldn’t have.

It had everything going for it, including:

• Successful, Celebrity Founders: Niklas Zennstrom and Janus Friis started the company in 2006 after palming off Skype to eBay for billions of dollars.
• Proven Technology: Joltid formed the basis for music- and file-sharing service Kazaa and later Skype.
• Substantial Funding: It raised $45 million in funding from the who’s who of the tech world: Sequoia Capital, Index Ventures, Viacom, CBS and Chinese tycoon Li Ka-shing.
• Incredible Buzz: The company had incredible pre-launch buzz that helped it to convince thousands of users to download its P2P video client — something that doesn’t happen all that often on today’s web.
• Big, Famous Partners: It managed to gain early traction with content providers such as Viacom and CBS, which were also investors in the company.

So what went wrong? Quite a few things, actually. Other startups should learn from the mistakes of Joost and avoid repeating them, such as:

• Too Big, Too Fast: Joost hired too many people, too quickly. It never behaved like a startup but instead always felt like a grown-up company with too many bureaucratic layers.
• Too Geographically Spread Out: The company was based in multiple geographic locations — New York, London and The Netherlands — and as a result, each location became somewhat of a silo.
• Not Enough Focus: Remember what your mom used to say when you took too big of a bite? If you’re not careful, you’re going to choke. Startups are just like that. Unless you focus, you’re going to choke. Joost couldn’t focus on one single market — and startups need to focus on one market at a time in order to win.
• Too Much Hype Too Soon: Like many, we were one of the early fans of this startup. Its founder pedigree generated a lot of pre-release interest. Nearly 250,000 folks signed up for the beta version of the software. But when technology problems hit, the pre-release buzz turned into buzzkill.
• Slow to Fix Its Technology Problems : Joost’s P2P network had technical problems early on that resulted in user defection. The company didn’t move to address those concerns fast enough. These technology problems have continued to nag the company throughout its life, even when it switched to a browser-based focus.
• Client vs. Browser: The company took too long to realize that the client-based strategy was going to lose out to browser-based video services. Its legacy of building clients became its Achilles’ heel.
• Didn’t Press Its Early-Mover Advantage: Joost had correctly identified that it needed the blessing of the content owners, but it failed to move aggressively enough to convince them to work with its platform. The client and technology problems didn’t help matters, either.
• Big Media Dis-Connect: Its big media investors were never willing to give Joost a content edge over the competition, prompting users to tune it out in favor of other services.
• Too Many Internal Problems: The company had some serious management problems, some of which led to the firing of its CTO in January 2008.
• Hulu: It started with a simple, easy-to-use interface for its browser-based video service, offered higher-quality video and used content from its backers, NBC and Fox, to become a household name, which in turn allowed Hulu to convince other content owners to sign up for its platform. Now it owns 10 percent of online video traffic.
• Chasing Its Own Tail: Joost also made some basic mistakes, such as not having a good SEO strategy. It never quite figured out a social media strategy in order to garner viral growth, either. It was like a tech company from the 1990s — out of sync with today’s web environment.

The dark cloud of doom started to settle over the company last year, as the team at NewTeeVee noticed time and again. NewTeeVee writer Janko Roettgers offered a recipe to fix Joost last fall, but apparently it was too little, too late, even then. The company consistently failed to gain any traction, even after unveiling new APIs and a browser-based offering. In the end, however, it all boiled down to a lack of content.

Photo credit of Mike Volpi pic: Eirikso via Flickr.

NBC Universal General Counsel Rick Cotton, speaking at the Digital Media Conference in Washington, D.C., on Thursday, brushed off concerns that the deal between Comcast and Time Warner to test the feasibility of TV Everywhere was a first step toward bringing TV on the Internet under the control of Big Media.

“I know there’s been some static in the system that says this will somehow limit access to content online, but this is about increasing access,” Cotton said. “These are subscription networks now so it’s not really a big surprise that they would be subscription networks online as well.” He also shrugged off fears that the collaboration between programmers like Time Warner and ISPs like Comcast represented some sort of unholy cabal worthy of antitrust scrutiny from the government.

“They shouldn’t have called it TV Everywhere because what we’re really talking about is cable TV everywhere,” Cotton told me after his panel. “The idea is you would go to Fancast, or to Hulu, and there would be free content, just as there is now, and there would be subscription content, which you could access if you’re a subscriber.” It’s just like cable TV, he said. “There’s free, over-the-air content, but there’s also subscription content you can get by subscribing to cable.” The deals between subscription network owners and web video portals, he suggested, would be no different from the perfectly legal deals that currently exist between the networks and cable MSOs.

Maybe, but that’s only the simplest use case being discussed. What happens when we get to Time Warner CEO Jeff Bewkes’ nirvana where any multichannel video subscriber — whether to cable, satellite or telco TV — can access any of the content they subscribe to from anywhere on the web, whether directly through an ISP, through a web portal like Hulu or on mobile platforms? Apart from the sheer complexity of such a system, making it work will require a degree of information-sharing among nominal competitors that practically begs for antitrust scrutiny.

Let’s say I want to catch up with a missed episode of “In Treatment” by watching it online. I go to a web portal — we’ll call it YooHoo — and find the episode. Since I pay for HBO through my Comcast cable subscription I ought to be able to access the episode, presumably by typing in a password or some other authentication method. For that to work, however, YooHoo has to know something about my relationship with Comcast. Maybe Comcast owns YooHoo, in which case, no problem. But what if some other media company owns it? It now has access to information about my Comcast subscription.

Now let’s say I leave the house and want to access the same episode on my handheld device using Verizon Wireless’ V-Cast service. Now Verizon has to know something about my relationship, either with Comcast or with YooHoo — or both. Apart from whether that’s something Comcast would want to do — Verizon has recently begun soliciting me with offers for FiOS, after all — what we’re really talking about is multiple competing service providers sharing information with each other (or with the same third party) about prices, payments, customer service and behavior and all sorts of other data.

Such a system would just be ripe for abuse, even if intentions were pure going in. And it would only get worse as the number of ways to access the content increased. Of course, you could always break up the vertical monopolies among programmers, distributors and service providers and the problem would be solved, but somehow I don’t think that’s what Bewkes and Cotton are talking about.

Paul Sweeting is the author of The Media Wonk blog. He’s covered digital media and policy issues for over a decade.

If the growing number of games being played on it are any indication, then San Francisco-based micro-messaging service Twitter has the potential to become the next major casual gaming hub. The thought first came to me a few weeks ago, when I discovered Spymaster, a game that allows you to run your own spy ring. Every action in the game is tweeted to your followers. After an initial burst, the game activity has moderated somewhat, but in the meantime it got me thinking about Twitter-based games, of which there are many. Among them:

• Trivia on Twitter
• Terminator Salvation, aimed at promoting the latest “Terminator” movie.
• WhoseTweet shows you 20 random Twitter messages, one at a time, and you try to identify to which one of your friends the tweet belongs. When it’s over, you can compare your performance with your friends.
• Tweet Quiz makes you guess the hidden words using clues delivered by tweets (kind of like a crossword puzzle), and each word yields a different amount of points.
• Beat My Tweet consists of scrambled words for followers to unscramble. Its site has a leaderboard to track players.
• Twivia tweets trivia questions several times a day and followers respond with @ replies; the first to answer gets points, which are tracked on a leaderboard.
• Twitbrain tweets out a math problem, and people race to @ reply back the fastest; results are posted on the site’s leaderboard.

Most of the aforementioned games are pretty basic and text-based, but I think in time we will see the emergence of more complex and interesting games. Spymaster is going to be first of many. For context, let’s go back to 2007, when Facebook launched its application platform.

After initial experiments with pointless apps such as vampire bites and zombies, Facebook quickly became a casual gaming mecca, thanks to the success of games such as Scrabulous, Mafia Wars and scores of other multiplayer social games. Twitter could follow the same path as well.

After all, much like Facebook, Twitter offers the virality so vital to a social network, a name space and thumbnails, a communication channel for messages and notifications, and a simple API for developers. It also has momentum. But those are the basic building blocks of a good social gaming hub — they need to be aggregated to produce engaging experiences as well.

Just like Facebook has been pushing developers to use Facebook Connect for authentication, Twitter can help game (and app) developers build their own gaming destinations by leveraging Twitter’s distribution network.

Of course, all of this means nothing without good games that can hold people’s attention.

A lot of things make Twitter special. The 140-character restriction makes the writing more potent, because people are forced to get to the point instead of rambling on. Anyone can search for things that are happening “right now,” as opposed to waiting hours (if not days) for Google to update its links. And unlike Facebook, discussions are open to the public, which encourages greater participation.

But one feature has been grossly overlooked in terms of what helps Twitter stand out: the ability to publish headlines to the Internet using only text-enabled cell phones. How is that special, you ask?

Imagine how confined Twitter would be if it were web-only, requiring both a browser and Internet access, like most social media platforms (i.e. blogs, YouTube, and to a lesser extent, Facebook). It would still work. You’d still get an ego boost with each new follower. But it wouldn’t be as popular or used as often as it is today.

The makers of Twitter are seemingly aware of this. “Sending updates to Twitter while you’re away from your computer makes things much more interesting,” reads documentation on the devices section of Twitter profiles. “It’s all done through text messages (aka ‘SMS’), which you probably use all the time anyway, so there’s not much to learn.”

A couple of things to note from that. First, SMS is “more interesting” when coupled with Twitter because the publication of social media doesn’t have to wait for a browser, access to the Internet, or portable technology like an iPhone or BlackBerry. With Twitter, there’s no more “I’m blogging this when I get back to my desk.” You can report from the field as a story happens — so long as you have cell reception, which is better than the 76.2 percent of the world without Internet access.

Secondly, since you likely use your phone “all the time anyway,” the frequency in which you contribute to Twitter is much higher than other forms of social media, which again, typically require more advanced technology enabled with an Internet connection. I think it’s safe to say most people are a lot closer to their phone than their Internet connection. And without a smartphone and data plan, the two are mutually exclusive.

As an added bonus, Twitter “by txt” lessens the effort needed to participate in social media, since you’re already using your phone more. If there’s one reason that independent bloggers have begun abandoning their blogs, it’s because blogs require a lot of work. Since a Twitter account is much easier to maintain and update than a blog, its drop-out rate might be lower than other platforms in years to come.

Granted, other web services leverage text messaging to their advantage. I can find nearby movie times or restaurants by texting Google. And Yahoo can text me news feeds and sports scores. But Twitter is the only SMS service to enable social media for the general public, not just friends in your network (à la Facebook). No Internet required.

That’s how ordinary citizens with rudimentary technology can impact real-world events like Iranian elections. That’s how Twitter can change the way we live.

Last summer, when I got my first iPhone, I found myself spending an equal amount of time downloading and installing various applications — some paid, some free — and using the excellent Safari browser to surf the web. Over the past few months, I realized that I was barely using my browser anymore, that the applications had gotten so much better that I was content to let them speed me to my web destinations.

Sometime in the past few weeks, I had an even bigger realization: The browser is dead. And it’s because all those apps that now monopolize my time have taken their pick of browser parts from the bin and blossomed into a phenomenon all their own.

The App Age

Back in 1998, in a world not yet connected by Google, Facebook and Twitter, the gentle folk from Redmond launched Active Desktop with Internet Explorer 4. It allowed the user to add to the desktop HTML content that could be updated and synchronized without visiting the web site in a browser.

Fast-forward to today, and it appears the time for this idea has finally arrived, in the form of applications –- first on superphones like the iPhone, and as more apps untether themselves from the umbilical cord that was the browser, on netbooks and even laptops/desktop computers, too. With some 5.5 million apps being downloaded every day from Apple’s App Store, many iPhone users may have already forgotten that the rendering engine used underneath them all is a webkit, the same underlying layout/display engine used in Safari and Google Chrome as well as Android and Palm Pre webOS.

Going forward, this trend of apps jumping out of the browser and onto available screen real estate will only gain steam. After all, with active push notifications, clicking a stock price or weather forecast or gas price app on the live desktop is definitively a better user experience than that of a browser.

Google’s Chrome already enables shortcuts to be placed on the desktop, start menu or quick-launch bar in Windows OS machines. It strikes me that moving tabs toward the outsides of the browser display window is merely a first graphical step towards freeing the app entirely from the browser. Eventually the tabs may disappear altogether, with complex web applications, including those from Google, humming in their individual “containers” and reached via direct click, not via a browser tab.

This scenario is a fitting coda to the browser evolution that began in the early 1990s, when the browser was essentially the only purpose-built network-connected application. Today, browsers have lent their structure, chassis and struts to network-connected applications that devour user time and attention away from the browser itself.

The App Web vs. the Legacy Browser Web

A significant consequence of the potential emergence of the App Web vs. the Legacy Browser Web will be felt on the web’s monetization engine –- advertising. From Adsense to Adwords to DoubleClick, as well as various other forms of click-based monetization, all must evolve as the App Web begins to dominate users’ time. To the extent that this Legacy Web was a cognitive medium, the emerging App Web skews, TV-like, towards an emotional engagement with the user. In this scenario, deep, immersive video applications à la Boxee (note Boxee’s non-presence in a browser) and thousands of casual gaming applications dominate user time and engagement.

Just imagine, a whole set of applications that finally free the video screen from its HTML-caged rendering/display and enable the presentation of desirable content through an intuitive, immersive video screen/application. In this engaging, app-mediated format, the proven monetization engine of brand advertising that powers television and other entertainment media will finally be possible in the online world. And the analytics that can be gleaned through the App Web promise to be better than the 18,000-30,000 viewers that Nielsen uses to predict/analyze the TV habits of 200 million American viewers.

The dominance of click-powered direct response advertising online will then face a challenge from demand/creation-oriented brand advertising. And it will have an unlikely ally to thank -– the humble browser, its genius revealed through letting a thousand applications bloom outside its constrained windows and tabs.

We are witnessing nothing short of the birth of a new and mobile mass media channel, one that will will reach more than a billion people around the world over the next few years via smart/superphones alone. This billion+ audience will be supplemented by numerous network-connected mobile and not-so-mobile devices and screens with thousands of network-connected applications. Each application is like a TV channel –- available on your mobile device or netbook or laptop/desktop — while the browser is a mere TV guide channel -– useful, but no longer dominant.

Rohit Sharma works with early-stage entrepreneurs in the area of infrastructure for cloud computing, networking, and storage systems. He was most recently a partner at Mohr, Davidow Ventures.

As businesses struggle to remain viable, much less grow, cost management is an imperative. Massive data center consolidation, automation and virtualization can drastically reduce costs — reportedly up to a billion dollars annually, in at least one case. However, money isn’t everything: CIOs need to balance what I’ll call the six FACETS of IT: Flexibility, Availability, Cost, Experience, Timeliness and Security.

These goals are often in conflict with each other. For example, consolidation alone can negatively impact availability — by putting all your server and storage eggs into one data center basket. And creating engaging user experiences for interactive applications demands geographic dispersion, the antithesis of consolidation.

A hybrid approach that uses cloud services to complement consolidated enterprise data centers can holistically address these six FACETS:

Flexibility — CIOs constantly face the challenges of new technologies, shifting application mixes, and changing market conditions. Unlike fixed-capacity enterprise data centers, cloud services promise “near-infinite” scalability to meet variable or unpredictable demand. But flexibility is more than scalability, e.g., it might entail the ability to shift resources from voice to video, or change data retention policies. This is where virtualization, converged multiprotocol networks, and cloud platforms can help.

Availability -– Today’s global IT users — customers, employees and partners — demand instant gratification 24/7. High availability of applications and infrastructure benefits from reliable components and from rapid detection, diagnosis and repair processes, but fundamentally is enabled via increased redundancy, which drives up cost. However, the cost of outages usually substantially outweighs the cost of mitigation. Unfortunately, since availability can never exceed 100 percent, there are exponentially diminishing returns on redundancy investments. The good news: Flexible, on-demand cloud resources can be provisioned only in the event of a disaster, reducing the cost of enhanced availability.

Cost — Nick Carr has argued that there will be a “big switch” to pure reliance on pervasive clouds to service enterprise IT needs, in the same way that electric utilities — which are also clouds — meet enterprise power needs. McKinsey & Co. has argued the opposite, complaining that cloud services are too expensive. However, for most businesses, the truth is somewhere in between; whenever demand is variable or unpredictable, total cost can be minimized through a combination of enterprise data centers and cloud resources.

Experience — The user experience has become a key competitive battleground; think how much the success of Google search or Apple’s iPhone owe to usability and responsiveness. Unfortunately, global network latencies are greater than the threshold of human delay perception and reaction times (about 150 ms), so data center consolidation can aggravate response time issues as sites close to end-users are shuttered and interactive applications and functions — SaaS, AJAX, keystroke and mouse move processing via remote virtual desktops, interactive gaming — are moved to more remote sites. Therefore, consolidation of enterprise data centers must be complemented by dispersed cloud-based services such as content and application delivery if the user experience is to be positive.

Timeliness — Accelerating time-to-value to meet ever-shorter product lifecycle windows is more important than ever. Cloud-based applications, infrastructure and platforms-as-services can compress time by replacing ponderous requirements, development, testing and/or provisioning processes with template configuration and agile assembly of component services.

Security — Security, in all its dimensions — authorization, authentication, protection, privacy, compliance, etc. — is as critical, if not more so, than the other FACETS. Cloud services have had occasional issues, but the real question is not whether cloud services are perfectly secure, but whether they are more secure than their enterprise counterparts. Just like banks are more secure than homes — because they must be to remain in business — cloud services either are or will be more secure than enterprise data centers. Security can often be maximized either by leveraging pure cloud services, or by complementing enterprise IT with cloud-based security services such as network-based firewalls, anti-DDoS, cloud-based anti-spam and anti-virus, and web filtering.

The bottom line is that consolidation of enterprise resources must be complemented by scalable, dispersed, secure cloud services in all of the major global regions in which a company does business, with a core of dedicated mirrored resources surrounded by a cloud of content and application delivery and/or WAN acceleration to reduce latency for key applications. An optimal architecture is hybrid: dedicated and shared, fixed and flexible, consolidated and dispersed, and owned and rented. Maintaining this balance is the best way to address the many FACETS of IT.

Joe Weinman is Strategy and Business Development VP for AT&T Business Solutions.

The iPhone has disrupted many aspects of the mobile industry, but nowhere is its impact more evident than in rise of mobile data usage. The number of U.S. mobile Internet users has more than doubled since the device was first introduced to stand at 46 million at the end of 2008 from 22.4 million in July 2006, according to Park Associates. The market research firm predicts that there will be 60 million smartphones sold in 2013.

But more importantly, the iPhone has spurred us to use the mobile Internet more and more. A report from 3G Americas shows that in Europe, data usage on iPhone increased substantially when the 3G version of the device came to the market. (See graph.) It’s hard to discern exactly how much bandwidth is being used by the average iPhone user, but the usage patterns point to heavy data consumption.

With the release of the new iPhone 3.0, I think the networks are going to come under an even heavier load as we’ll see another big spike in mobile data usage (which explains why AT&T has been hustling to upgrade its wireless network.)

Chetan Sharma, a veteran mobile analyst who is part of the GigaOM Analyst Network and is the author of “4g: State of the Union,” thinks mobile data usage is going to jump at least 1.5-2 times with the new release.

The reasons are pretty simple: The 3.0 release improves the user experience, which would prompt people to use their device more. Multimedia-related technologies such as video streaming, MMS, built-in search and in-app sales are some of the key new features that are going to suck more bandwidth. (For complete feature breakdown of the iPhone 3.0, check out TheAppleBlog.)

And let’s not forget about push notifications: The more notifications we get, the more we’ll be using the iPhone. Sharma agrees, saying he thinks the upgrades will “spur new development, which means means more downloading.”

Listening to Jeff Bezos, founder and chief executive officer of Amazon, is like going to startup school where you learn that failure is part of entrepreneurial growth. Whenever I have talked to Bezos in the past, the things that have stuck in my head have been his willingness to be wrong and his unflinching abhorrence of the status quo. At the Wired Business Conference in New York City, Bezos reiterated some of those points in a conversation with writer Steven Levy.

Unfortunately, my month-old MacBook died in the middle of writing that post and I lost much of it; it wasn’t until later tonight that I managed to borrow a ThinkPad from a friend. Of course, by now many have already written about Bezos’s comments on the Kindle, Google’s book efforts, and e-readers, so I’m going to skip those and instead focus on some of the comments about innovation and entrepreneurship. Many of his tips have helped me think clearly about our little startup as we grow older. Hopefully you enjoy these nuggets of wisdom from a man I like to call a constant gardener of innovation.

Innovation and large companies

In describing change, Bezos said that rapid change is always scary for incumbents, but if you’re not an incumbent, you have nothing to lose. Large incumbents have to worry about legacy revenues. Innovation is hard for large companies because you need to be long-term oriented. And since the innovative projects are such a tiny part of a large company, there is tendency to be dismissive of the innovation.

“You need a culture that high-fives small and innovative ideas and senior executives [that] encourage ideas,” he said. In order for innovative ideas to bear fruit, companies need to be willing to “wait for 5-7 years, and most companies don’t take that time horizon.”

In tough times, focus on the customer

In the aftermath of the dot-com bust, Amazon was one of the many Internet companies that saw its stock price dive faster than a flying seagull in search of food. “The stock price deflated but the business continued to grow,” he recalled. As a result, Amazon refocused its energies on the customer.

“Focusing on the customer makes a company more resilient,” he added. “We had to tell employees that they shouldn’t feel smarter because the stock went up 35 percent, because then they would feel 35 percent dumber if the stock price went down.” His tip on managing during tough times such as those faced by Amazon during the bust was to communicate more with its employees. With too many external inputs, Bezos thinks it’s important for companies to be talking with its people more often, easing their concerns.

Be stubborn

The difference between founders and professional managers is that founders are stubborn about the vision of the business, and keep working the details. Professional managers, when things don’t work, want to change the vision. The trick to being an entrepreneur is to know when to be stubborn and when to be flexible: Be stubborn about the vision, but flexible about tactics, Bezos said. For instance, you can be flexible about reducing costs, but you don’t change your vision to reduce costs. Great point, because more often than not, founders get caught up in the tactics and change their direction as a result.

Prerequisites of innovation

Bezos listed a few prerequisites for innovation and inventing, but the biggest one is willingness to fail. You need to think for the long term and be misunderstood for a long period of time. “If you can’t do those things, then you need to limit yourself to sustainable innovation.” In other words, seek incremental change to grow your business.

In talking about the need for thinking long term, Bezos said you need conviction. I totally agree — if you take a short-term approach, then you are constantly stuck with trying to deal with minutiae. He noted how much hated the idea of “sticking to one’s knitting” and not taking chances. That said, Bezos was clear in pointing out that his company looks at everything from a customer’s standpoint. “We do make business decisions in a very deliberate way,” he said. “We work backwards from customer needs.”

Errors of omission vs. errors of commission

Many people make too much about the errors of commission. People overemphasize their failures when trying something new. Actually failure is not that expensive and it’s part of work. If something fails, then you’re going to shut it down and cut your losses, Bezos said. The focus, he said, should be on errors of omission. These are the chances not taken. He is not ashamed of his failures — A9 search and Auctions are two examples he cites often. Both markets were big enough for Amazon to take a flyer.

(I am paraphrasing from my notes and as a result there might be some mangled quotes in there somewhere. Once I revive my computer,  I will recheck against the audio files I have sitting on my laptop.)

IBM’s first true cloud computing products, announced today, consists of workload-specific clouds that can be run by an enterprise on special-purpose IBM gear, Big Blue building that same cloud on its special-purpose gear running inside a firewall, or running the workload on IBM’s hosted cloud. The offering seems like a crippled compromise between the scalability and flexibility that true computing clouds offers and what enterprises seem to be demanding when it comes to controlling their own infrastructure. I spoke today with the chief technology officer of IBM’s cloud computing division, Kristof Kloeckner, to learn more. Below is an edited account of our talk.

GigaOM: Let’s start with the hardware underlying IBM’s CloudBurst offering. How does this compare with what Cisco is doing or other cloud hardware out there?

Kloeckner:  This first instance for test and development workloads is built on Intel-based blades, but we anticipate other workloads might run on different platforms. We are actually working with the mainframe team for particular workloads. We have a prototype running that has p-series and z elements for SAP workloads.

GigaOM: So in IBM’s view the workloads dictate the hardware, rather than the idea of commodity servers being used to build out a general purpose cloud?

Kloeckner: We make the hardware selections based on the workloads you want to run, and we optimize the workload for you. But because it is in the cloud, in terms of what do you see as a client as to how each different cloud behaves, it’s all entirely consistent.

GigaOM: Why focus on workload-specific clouds?

Kloeckner: One should really instantiate clouds with the workloads that you run on them in mind. Depending on what the delivery needs are you might have an analytics cloud separate from your collaboration cloud, and you might also decide you want to keep the test and development cloud in-house, and then expand into the public cloud for collaboration services.

GigaOM: Why focus on test and development clouds for your first products?

Kloeckner: When we looked at development and test it’s considered so crucial for accelerating the business value of IT, and we think that making dev and test more efficient and accelerating the process through automation was extremely attractive. About 30-50 percent of our client’s resources are devoted to dev and test. It’s also part of the infrastructure that’s not well managed. For example, after new apps are tested, in some cases the department doesn’t want to give up access to those resources because it may take a long time to get them back. Making test and dev dynamic can be instantly attractive.

GigaOM: Is it also a focus because other enterprises are already using public clouds like Amazon’s EC2 for those workloads?

Kloeckner: The general practice of dev and test is to have it in-house. There is no massive trend by organizations to bring that out into the public cloud infrastructure. We see individual organizations try it out, but enterprise development is mainly in-house today. And this is the first of a whole series of offerings to come. We’re going to look at analytics and business apps in the future, but we started with dev and test.

GigaOM: If the vision of workload-specific clouds proliferates, how do enterprises work across different clouds? Does IBM have a solution for that?

Kloeckner: We demonstrated some early solutions with Juniper’s switching technology back in February and use our job scheduling software to schedule across domains. We have our efforts on the Open Cloud Manifesto, and have had public demonstrations extending our service management software so it can manage workloads in a variety of clouds. We do not have a packaged solution yet, but we can work with clients to extend across multiples clouds.

Happy Birthday to us — we are three years old!

It seems like only yesterday that I was sitting at the Starbucks near my old office with my friends Nitin Borwankar and Dave McClure, along with ace web designer Mike Rundle (via phone), plotting the design of a new GigaOM. I was making plans to leave the job I loved at Business 2.0 in order to turn my blog into a company. But while I wanted to launch this new company on Independence Day, July 4th, my plans leaked, and three years ago today, GigaOM went from being my blog to being my startup.

I had never founded a startup before, so I didn’t know what was in store for me. There was a lot of optimism, hope and excitement: I was and still am trying to figure out the future of media. And while the destination was (and still is) unknown, the promise I made to myself was that not only would I strive to build a company that would be my pride and joy, but I would be sure to enjoy myself along the way. And that is only possible if you surround yourself with a great team, made up not only of professionals, but of great people with great values. That principle has been the guiding force for our company. Since those early days, while much has changed, our ethos and vision remain the same.

We are a much bigger group. We are highly distributed. We have seen some friends leave and new friends join. We’ve survived some troubling times. Most importantly, we have grown both as a company and as a collective group — we have done it together, as a team. And I have learned the most important lesson that an entrepreneur can: Empower people, and in turn they will reward you with great things. So while letting go has been hard, it’s also been the best part of these past three years.

Back when I was in college, as a chemistry student, I came to appreciate the atom more than anything else in the world. Atoms, when bonded with other atoms, form molecules. Some of these bonds are magical and the molecules can, in turn, be magical things. Startups are essentially the same. Atoms are the individual people; the company is the molecule. And that’s what makes our journey into the future — some of it predictable and a little bit of it unknown — so enjoyable. Three years into the opportunity of a lifetime, I still have a spring in my step as I leave my apartment, for I can’t wait to be with each one of my team members.

But most importantly, I can’t wait to be with our community. This week, for the first time in a long time, I took a break from all the brouhaha of Silicon Valley to spend a couple of peaceful days in lovely Wyoming. The majestic beauty of the mountains, however, couldn’t stop me from thinking about you, our readers. There are many more of you today then there were three years ago. I don’t know all of you by name, but everything you say, I listen to, and every time you teach, I learn. And I know I speak for our entire team when I say — thank you, readers, for rewarding us with your time and attention.

As a business, we are executing our plan. GigaOM has grown to become a network of awesome blogs: WebWorkerDaily, NewTeeVee, OStatic, Earth2Tech, jkOnTheRun and TheAppleBlog. We now host four industry-leading events every year: Structure, NewTeeVeeLive, Mobilize and GreenNet. We recently launched a subscription-only research community, GigaOM Pro. We have syndication partnerships with The New York Times; CNNMoney and Fortune; BusinessWeek; and Salon. And we have some more new tricks up our sleeve, too.

As a long-time blogger, I have seen a lot of changes take place in the media marketplace — the emergence of Twitter, for example. Twitter is the ultimate expression of the continuous democratization of news. With it, both the distribution and discovery of content on the Internet have changed. But as I’ve written in the past, there is also a desperate need for context. It’s made me think and rethink the value of not just our blog but blogging in general. Over the past three years, we have evolved from a news-chasing entity into a technology insights company. From our blog posts to our research, we work hard (sometimes we stumble) to step away from providing commodity information to instead provide analysis and context to the relevant news of the day. The launch of GigaOM Pro is part of this evolution into an insights-based company. Over the next few months, you are going to see us make efforts to provide even deeper and at the same time, wider insights. So stay tuned.

Before I go, a big word of thanks to all the dear readers and friends who have encouraged us along on this breathtaking journey. I want to take a moment to thank my entire team for their hard work, dedication and love. To our investors — thanks, guys, for believing that a first-time entrepreneur can bring about change.

Now it’s time to get back to work.

Last September, on the eve of our first Mobilize conference, John SanGiovanni, co-founder and VP of product design at Zumobi, talked to us about the coming era of the superphone. Eschewing the smartphone moniker, SanGiovanni noted how this new class of handsets — led by none other than the iPhone — was starting to become part of our everyday lives.

With vastly better performance, desktop-grade web browsing, and high-resolution displays, a new category is born. I call them “superphones,” and they are achieving tremendous traction with consumers and professionals alike…the next wave of true superphones promises to back up a device’s good looks with deeper platform technologies and more robust back-end services.

Fast-forward to today, and we are standing on the cusp of what will be the summer of the superphone. (For what makes a superphone, please refer to the accompanying table.) Here are some of the devices you can expect to see over the coming months:

These launches come as growth in the overall mobile market is starting to slow — drastically. According to Brian Modoff, an analyst with Deutsche Bank Securities, the increased demand in the first quarter and start of the second lulled people into thinking that the mobile market was back on track. Retailers and carriers restocked. Oops!

In our handset model, we are currently modeling 6% sequential growth, and our sense is that the Street is expecting somewhere in the range of 5% to 7% growth. However, we are seeing increasingly seeing signs that this may be too much to expect. The first dissenting note came from Nokia who on their earnings call said they expect flat to 2% growth. Since then other contacts have told us that a flat or modestly up quarter may be what we should expect, as one of our contacts put it, “Our order improvement is done.”…As we have noted in past reports, there is a difference between not getting worse and actually getting better. We think the industry inventory levels may have reached their new steady state, better than 4Q08 but still well below historic levels. (From DBS research report, dated May 31, 2009.)

In the superphone business, there are none as dominant as Apple and RIM. Their dominance is one of the reasons why they’re able to make a lot more money than their rivals — despite selling many fewer handsets.

As the iPhone entered the market and RIM entered the consumer sphere the profits of the rest of the vendors has declined dramatically. Looked at another way, Apple and RIM each claim a greater share of profits than any vendor except for Nokia and Samsung. (From DBS research report, dated May 31, 2009.)

That said, there is a lot of interest in Palm’s Pre. Same goes for Nokia’s N97. Will that be enough? Or will the iPhone/BlackBerry juggernaut continue to suck the oxygen out of the handset business and ensure that Apple and RIM keep their stranglehold on the superphone business? My bet is on the latter!

PS: If nothing else, the SuperPhones has created a demand for mobile apps, The New York Times says. Increased M&A of these mobile apps is up next, the Times says.

Nothing quite works like the threat of the possibility of losing a money-making asset. With rumors that Apple is flirting with Verizon over a future partnership, AT&T has abruptly changed its future wireless broadband plans to include a sudden acceleration of its LTE deployment. LTE is an acronym for Long Term Evolution, the 4G wireless technology that is being favored by most carriers on a global level. Today, AT&T issued a press release touting even faster 3G mobile broadband speeds. Buried in the press release was this little nugget:”AT&T plans to begin LTE trials in 2010, with deployment beginning in 2011.”

Wow! Ma Bell has been insistent that its 3G network has a lot of headroom and it doesn’t need LTE right away. It had hemmed and hawed, saying it wouldn’t begin testing until 2010 or 2011, with full deployment coming after that. What has changed is that, as we pointed out last month, LTE might have been the real reason why Apple and Verizon were getting cozy. As part of today’s announcement, AT&T said:

• It will boost 3G speeds to HSPA 7.2, and the network upgrades will end by 2011.
• It plans to begin LTE trials in 2010, with deployment beginning in 2011.
• The upgraded network platform could allow for theoretical peak speeds of 7.2Mbps, though the speeds you are going to get are likely to be much lower.
• It will introduce multiple HSPA 7.2-compatible laptop cards and smartphones beginning later this year. You can count on one of them being Apple’s new iPhone.

Having been a harsh critic of AT&T, I liked that the company is finally addressing its network problems. Even CEO Randall Stephenson admitted onstage at the D Conference that there were problems and the company wasn’t quite ready. Of course, at the time, company executives said otherwise. In fact, most of its 3G-related advertising is misleading.

According to the press release:

In addition to the planned speed upgrade, AT&T is enhancing its mobile broadband coverage by nearly doubling the wireless spectrum dedicated to 3G in most metropolitan areas to deliver stronger in-building reception and more overall network capacity. Also, AT&T is adding thousands of new cell site backhaul connections to support the higher mobile broadband speeds enabled by HSPA 7.2 and LTE.

The company has so far been in complete denial about its network coverage. The limitations of its network were exposed during SXSW earlier this year. AT&T has refused to discuss what it was doing in order to fix the problem, instead putting its head in the sand. Today’s press release is the first step in admitting that it has had a problem. Here are some of the things AT&T is doing:

• Almost doubling the radio frequency capacity. It will be deploying more of its 3G network on the 850 MHz spectrum.
• More bandwidth to cell sites including adding fiber-optic connectivity and additional capacity to thousands of cell sites across the country this year.
• More cell sites. Deployment of about 2,100 new cell sites across the country.
• Many AT&T smartphones will be able to switch seamlessly between 3G and Wi-Fi connectivity.
• MicroCells. Customer trials leading toward general availability of AT&T 3G MicroCell offerings, which utilize femtocells to enhance in-building wireless coverage.
• AT&T plans to spend between $17 billion and $18 billion on its network in 2009.

Of course, we wonder what happens to some of those older phones that, like iPhone 1.0, work with the older EDGE network. We wanted to ask Ma Bell about the dropped call problems, but the executives are too busy with meetings or talking to reporters who don’t ask tough questions.

There is an error in the slide. HSPA+ should read 20+ Mbps. It currently reads 20+ kbps. The error is regretted.

Online product recommendation systems have been around for almost as long as e-commerce. They all share the common goal of recommending items a particular person is most likely to be interested in at a given time. In aiming for this common goal, however, recommendation systems take a wide variety of different approaches. These approaches can be grouped into four main categories. Some systems use only one of these; others use combinations. The key techniques are:

• Segmentation, which divides users into groups based on characteristics like age, gender, and geographic location;
• Collaboration, which starts with an individual and attempts to locate others like them;
• Personalization, which relies on a user’s prior actions to determine what they are likely to do next; and
• Similarity, which starts with products, rather than users, and models relationships between them to drive recommendations.

The first technique, segmentation, also known as demographic or psychographic profiling, assumes that members of each group are very similar and recommends similar products to all of them. Snow blowers are almost certainly more interesting to middle-class residents of Chicago suburbs than to retirees in South Florida. That being said, segmentation is a blunt instrument. Not everyone is typical of their age, gender, or ZIP code. Online, the challenges can be even greater, as many users arrive anonymously.

Collaboration, also known as collaborative filtering or “wisdom of the crowds,” attempts to improve upon segmentation by using observed behavior, rather than personal characteristics, to determine which users are similar. In this approach, Joe might be recommended a snow blower because he bought a leaf blower in the fall and a lawn mower the previous summer, and many of the people who bought those two products also bought snow blowers. The key drawback of this approach is that it often finds correlations that are not driven by any identifiable underlying factor, which can lead to curious and confusing recommendations.

Personalization, also known as one-to-one marketing or intent modeling, moves away from group behavior and focuses on individual behavior. If users spend time surfing the technical specs of three different snow blowers, there is a very good chance they are interested in purchasing a snow blower, regardless of what they have purchased before. But reliance on only individual behavior at the expense of the other techniques can be unnecessarily limiting. Once the user has put the snow blower into his cart, he might be more than willing to add the latest “James Bond” DVD because people who bought other DVDs the customer bought in the past are buying it in large numbers.

Finally, we have similarity, which identifies users who have interacted with similar products as good candidates for recommendations. This is particularly useful for new products. Say a band has just released an album. A retailer who hasn’t sold any copies of it yet can be reasonably confident that people who bought previous albums by that band are going to be interested. The weaknesses of similarity are that it assumes a site has reliable structured data on products, which not all sites have, and it assumes people are interested in groups of similar items more than they are interested in a diversity of items.

As you can see, each of the four techniques outlined has inherent strengths and weaknesses. The good news is that there is no reason whatsoever why choosing one precludes using any of the others. In fact, if a site is using just one recommendations technique it is, almost without question, leaving sales on the table. Sites like Sears and Burton, which use more than one recommendation technique, have been able to boost sitewide sales by up to 30 percent over the long term and drive significant increases in user loyalty and repeat purchase rates. The question every site has to answer is how to combine the techniques to create the best overall experience. After serving more than 20 billion product recommendations, we at richrelevance have found that the best approach is to provide a wide variety of relevant recommendation types, and to continually test their performance in real time in order to optimize our choice of techniques across the various page types and contexts on each individual site.

Darren Vengroff is chief scientist at richrelevance. Previously, Darren was CTO and co-founder of Pelago, principal engineer at Amazon.com, and a vice president at Goldman Sachs.