Adam Rifkin writes:

The dominance of eBay in one sense violates the underlying philosophy of the web. That is, the web isn't all hosted in one place, so why should there be one dominant marketplace where buyers and sellers go to meet? Why should there be one dominant place where reputations and opinions of products get stored? Why should there be one dominant place where people I know and trust for referrals get stored? There shouldn't be.

Commerce over the web should be as decentralized as the web itself. In such a world, search engines become more important, and pubsub becomes more important (as buyers and sellers regularly want to publish and subscribe to messages saying "I want to buy this" or "I want to sell this"). There's a very exciting future out there, waiting to be invented...

WSJ writes:

The growing popularity of blogs and other online forums has prompted companies to pay more attention to what is being said about them on the Internet, and has given rise to a new kind of market research aimed at finding useful information in the sea of online chatter.

Jonathan Carson, chief executive of Internet-tracking firm BuzzMetrics in New York, says Web-monitoring is also useful as a market-research tool. A traditional researcher, he says, might gather a group of car fans to talk about their favorite models, or spend an afternoon driving around with a research subject. "We look at what they're already saying," he says.

BuzzMetrics has created a panel of what it calls "word-of-mouth influencers," a list of thousands of bloggers, message-board posters and other people BuzzMetrics has deemed influential in the online community, in part by examining traffic numbers. By studying their online interaction, BuzzMetrics says it can give companies important information about how they are perceived by customers.

In one recent project, an airline Mr. Carson declines to identify wanted to revamp its check-in process. Instead of talking to flyers directly, BuzzMetrics studied their conversations online, on Web sites such as FrequentFlier.com and FlyerTalk.com. The research revealed that many flyers had a positive opinion of the airline overall, but were confused about how and when to use self-service kiosks at airports, rather than waiting in line for an agent. BuzzMetrics helped the airline develop a new system that simplified the check-in process.

ZDNet writes:

Car hire firm Europcar International has migrated thousands of PCs across Europe from Windows fat clients to Linux thin clients, lowering both its hardware and maintenance costs.

The car hire firm has saved on hardware and maintenance costs by migrating its 1,500 stations to thin clients running Linux, but Stefan Ostrowski, the CIO of Europcar, said that while migrating to Linux thin clients has saved them money, but it would not have been as cost effective for them to migrate fat clients to Linux.

"The effort to install or maintain Windows and Linux is the same, though you might save a bit on licence costs," said Ostrowski. "You are not saving a lot by moving fat clients from Windows to Linux. But by converting fat clients to terminal servers we have reduced the total cost of ownership by 60 percent."

Europcar chose to run Linux rather than Windows on the terminals so that it could create a bespoke version of the operating system which included specific inventory, security and remote management tools needed by the company. It has based its bespoke version on Debian.

The main advantage for Europcar in migrating to Linux terminals has been the ability to centrally manage the terminals in its 1,500 rental stations, which are spread across Europe. Ostrowski said this has dramatically reduced the cost of maintaining the systems and in particular the cost of implementing updates.

Robert Scoble writes about his meeting with Doug Rowan, former CEO of Corbis. He has a list of some interesting photo apps.

The Economist writes:

Because VOIP service relies on software, rather than the traditional physical telephone infrastructure—voicemails, for instance, come into one's e-mail inbox and can be saved and forwarded—it upsets the entire telecoms industry, for two reasons.

First, while traditional telephony takes account of geography, distance, and time, says Michael Powell, America's telecoms regulator, “VOIP shatters all three”. In most cases it makes no difference to a VOIP caller where he is, how far away from the person he is calling, or how long they talk. VOIP phones can have traditional telephone numbers, yet still work no matter where they are, provided they are plugged in to a broadband internet connection. Lots of Indian mothers in Delhi have Vonage phones with the American area code 650 so that they can make cheap “local” calls to their sons in Silicon Valley.

Second, VOIP uncouples the two previously intertwined components of telephony: access to the network (via a wire running into your house, for example) and service (the ability to make and receive calls). Traditionally, both have been provided together. With VOIP you can buy broadband access from one firm and a telephony service from another—or even from a company in another country altogether.

Who will be the biggest losers? Not the fixed-line telcos, even though their revenues may fall by 25% by 2010 due to VOIP, according to Mr Mewawalla. The mobile operators are likely to be the big losers, with their revenues plunging by 80%. Together, VOIP and wireless broadband could fatally undermine their costly third-generation (3G) networks.

Computing in the developed markets over the past 25 years has been dominated by personal computers – running DOS first and Windows later on x86 architecture chips. Intel and Microsoft have dominated this world. The past few years have seen the emergence of PDAs and smartphones which deliver a small but increasingly growing subset of functionality of the computer. It is now not uncommon to see to see emails with a footnote attesting to the fact that the email was sent via a Blackberry.

For all their success, computers are outnumbered by cellphones. In fact, for many in the emerging markets, the cellphone is the first personal device that they get. It is not just upper-class consumers and business users who are seen using cellphones. From vegetable sellers to construction workers, from auto-rickshaw drivers to fishermen, from teenagers to grandparents, the cellphone is becoming increasingly ubiquitous in emerging markets.

Both devices – PCs and cellphones – have their advantages and disadvantages. The full-size input-output attachments of a computer are unmatched as compared to the miniaturised (in comparison) display and keyboard of a cellphone. The computer also has a vast library of applications developed over the years. Its versatility to work in the home, business and educational context is what has made it one of the greatest inventions of our generation.

Yet, the computer remains largely a developed market gizmo. In emerging markets, its dollar-denominated cost has limited its appeal to a fraction of the potential user base. In recent times, manageability challenges have increased due to the proliferation of viruses and spyware specifically targeting the Windows platform. So, even as the top of the pyramid in emerging markets can use the computers, the next 90% remains a non-consumption market.

By contrast, the cellphone user base has grown phenomenally. In India, 2 million cellphones are purchased each month, about six times the sales figure of computers. While the cellphone caters to a natural desire for communications, its ease of use has no doubt helped. Even as cellphones become smarter, they will be limited by three factors in comparison to computers: the size of the keyboard, the resolution and size of the display, and the control exercised by the operator for services delivered to the device.

To bring the next set of users into the world of digital services needs a computer which borrows some ideas from the cellphone – think of it as a thin client or a “network commPuter.” It is a device which is built assuming the existence of communications networks. It has the footprint of a computer with the affordability and manageability of a cellphone. The commPuter is a zero-management device which does almost no computing locally. It relies instead on a computing- and storage-centric grid for becoming useful. Think about it: the radio, television and the cellphone are useless without networks. It is time for the computer to also make the transition. This is the secret to building the Rs 5,000 ($100) computer – about half of that covers the cost of the device, and the other half is the cost of a refurbished monitor. (The TV lacks the display resolution – it is better to stick to a computer monitor.)

Tomorrow: Networks