Smart Innovation

WSJ has an interview with Nicholas Carr:

For a lot of companies, the people who are in charge of innovation, whether it’s the R&D folks or product developers or entrepreneurs in small companies, are the people who are the most passionate about the particular new technologies. They tend to be the early adopters, and that can give them a distorted view of the market. Because most normal people are actually quite conservative: They’ll adopt a new technology, but they tend to do it quite slowly. That opens up a big opportunity for companies that are smart in figuring out how to help normal, everyday customers create a bridge between an old, established technology or way of doing things and a new one.

New technologies tend to be difficult to use. They tend to be buggy and not work perfectly. They tend to be expensive. All of those things mean that they tend to be limited to a small, early-adopter customer base for quite a long time. If you can figure out a way to move with the market toward the new technology, I think you can do a lot better than jumping ahead.

Social Networking Sites

The Economist writes:

After plenty of initial scepticism, investors now accept that MySpace and its rivals can make a lot of money by selling advertising space. Advertisers were wary of putting ads on individual home pages, which often feature lewd and unpredictable content. But they cannot ignore the fact that millions of young people spend hours on these websites.

The question for MySpace, says Mr Werbach, is whether it can grow into a business on the scale of Yahoo!, eBay or Amazon. Pali Research, an independent research firm in New York, estimates that MySpace had revenues of $150m and made a loss of $50m in the year to June 2006, after making payments to its founders. Pali predicts $300m in revenues and profits of $200m in the year to June 2007, excluding revenue from its deal with Google. Like other social-networking sites, MySpace has not had to advertise itself, and its users provide most of its content, which should produce profit margins.

Microsoft’s Zune

Engadget has details about the announcement: “they’ve confirmed the basics we’ve known for a while, like WiFi, 30GB of HDD, built-in FM, a 3-inch screen and the basic music, pictures and video playback. They also finally let slip the screen res — an unsurprising QVGA — and some better news on the codec front: the Zune supports h.264, MP3, AAC and WMA. As for ballyhoo, wireless Zune-to-Zune sharing is where the real action is at, and it works pretty much like we’ve been hearing: you can share a full-length track with a friend, and they’ve got three times to listen to it over a three day period, after which they can flag the song for purchase on the Zune Marketplace — unless they’re an unlimited “Zune Pass” subscriber, of course. You can also share playlists and pictures with your buddies, along with what we suppose are “unprotected” homemade recordings.”

Video Advertising

AlwaysOn has a comment by Greg Stuart, president of the Interactive Advertising Bureau: “there’s absolutely no relationship between a click-through and a brand attitude change – it has nothing to do with it whatsoever. Consumers hear messages, they see messaging and they begin to shift their perceptions about that brand as a result of that, and that again leads to that change of behavior. So I would be very concerned if the broadband video industry started to measure click performance as some metric. It’s just not the right thing to do.”

Mobile Entertainment writes:

A day after his company bought a controlling stake in ring tone giant Jamba, News Corp.’s president and chief operating officer said the time for partnering with the mobile industry to distribute entertainment content is now.

Chernin said that today only 4 percent of the 219 million mobile subscribers in the United States watch mobile TV on their handsets. But if that figure increased to 20 percent and each viewer spent just $10 per month on mobile video, mobile TV would generate nearly $5 billion in revenue.

TECH TALK: The Now-New-Near Web: EventWeb (Part 4)

Ramesh Jains fourth post discusses Experiential Data.

The experiential attributes of an event are fundamentally different from informational attributes. Each experiential attribute represents a data (stream) that is experienced using a specific natural human sensor. Thus we may have visual data, audio, tactile, olfactory, and taste related data. Currently good sensing and reproduction techniques are available for visual (image and video) and audio data. Tactile is improving fast and others are slowly getting developed. Thus, experiential data will be defined as the data of a particular sensory type rather than an integer or real or character type as commonly used in informational data.

Experiential data is usually much larger in volume than other data types. For historical reasons, in computing most representations evolved to represent simple data like numbers and characters. A collection of numbers representing an image is thus represented using an array of data, an intensity value at different pixels forming an image. A video will be an array of such arrays. In databases, when designers faced such data, they usually lumped it all and called it a binary large object, or a blob. Search engines analyze a text file and identify words in it by analyzing arrangement of characters, but usually dont open an image to analyze it. In general, except few people specializing in particular experiential data analysis, people have avoided dealing with experiential data. Interestingly, slowly experiential data started becoming popular and now photos, video, and audio are becoming the central data elements in computing.

Another distinguishing feature of the experiential data is that it is always grounded in space and time. A sensor captures data at a point in space and in many cases over a time period. Thus the data captures a physical phenomenon at a given point in space and over a particular time interval.

The operators and methods to be applied to experiential data are significantly different than the methods used for processing alpha-numeric data that was commonly used in many traditional computing fields. Of course due to the nature of digital computers, the most basic operations must be reduced to the basic processing operation in computing. For human abstraction and use, however, these operations are fundamentally different. The computational techniques for experiential data are emerging and clearly are not as well developed as techniques for computing and managing alpha-numeric data.

Rapid progress in sensing, storage, processing and display (reproduction) technology is making experiential data rapidly popular. It is rapidly becoming not the secondary source of information, but a primary source of experiences and communication. Have you noticed that computer as well as mobile phone manufacturers usually advertise their devices based on their experiential characteristics? They tell you how good the camera or video processing capability of the device is. They know that experiential data appeals to humans much more than the abstract numbers.

Next Week: The Now-New-Near Web (continued)

Continue reading