blushimages.com

Those announcements probably didn’t feel Flash-y enough for the company, though, so Adobe simultaneously announced an embeddable player for virally marketing Photoshop Express posting your photos to home pages and blogs in glitzy slide shows. Given the relative simplicity of the application and broad appeal of photo sharing, this capability also sounds like a natural springboard for companies looking to dip their toes into Open Screen Project development–once Adobe releases the relevant API information, of course.

Update: Adobe has informed us that while the new Flickr connection isn’t live yet, it will be very soon. We will provide another update when we have confirmed that it is live.

When Adobe launched Photoshop Express at the end of March, it indicated that Flickr support would be next on the agenda, and today the company can cross that item off its to-do list. With the capability to round-trip photos into PSE for editing and back out to its site, Flickr joins Facebook, Photobucket, and Picasa in the ranks of Photoshop Express supporters. Additionally, users of Photoshop Express albums will now be able to create multiple versions of a given image, a much-requested feature, according to Adobe.

(Credit:
Dan Farber/CNET)

DeWolfe was fielding a few questions in an interview at the TechCrunch50 event in San Francisco with co-host Mike Arrington, who started off the interrogation by asking DeWolfe if he was dating Paris Hilton. The gentlemanly DeWolfe declined to answer the question.

Michael Arrington talks with MySpace's Chris DeWolfe at the TechCrunch50 conference in San Francisco.

He did talk about the new music service launching this month that will partner with the major labels, offer free streaming, and include some original content from its audience of 120 million members. DeWolfe said that MySpace is very focused on making the new service a success, with more than 70 people, including top MySpace management, working on the project. He was asked if Amazon was providing the downloading capabilities, but declined to answer. MySpace has a strong music foundation, and may be able to make some inroads into the Apple/iTunes territory.

MySpace plans to integrate Google Gears with its platform, according to the social network’s co-founder Chris DeWolfe. Users will be able to access their profiles offline using the Google Gears APIs, but the feature won’t be available for a few months.

Starting with a Chicago-area facility due to open later this year, Microsoft will use an approach in which servers arrive at the data center in a sealed container, already networked together and ready to go. The container itself is then hooked up to power, networking, and air conditioning.

A conceptual rendering of the planned data center due open next year in Dublin.

Josefsberg said his goal is to keep capacity a certain number of months ahead of where Microsoft’s utilization is running. To do that, he said, takes some serious planning. Business unit heads who used to have to just create a forecast for revenue and headcount, now need to be able to predict how much server capacity they will need, or at least give Josefsberg the data he needs to make such calculations.

“The trucks back ‘em in, rack ‘em, and stack ‘em,” Chief Software Architect Ray Ozzie told CNET News. And the containers remain sealed, Ozzie said. Once a certain number of servers in the container have failed, it will be pulled out and sent back to the manufacturer and a new container loaded in.

“One of the big drivers for us that I see is the move to IP-based delivery of rich video,” Josefsberg said.

No more off-the-shelf hardware
Gone are the days in which Microsoft settled for off-the-shelf hardware to fill its server farms. These days, Microsoft is looking for servers designed to its exact needs. It’s not just that Microsoft doesn’t want servers that have keyboard or USB ports–it wants motherboards that don’t even have the added wiring necessary to support those things that it will never use. Such moves eliminate cost, space, and power consumption.

Microsoft is seeing new demands, he said, such things like consumer video and photo services as well as its collection of hosted enterprise services under the Microsoft Online moniker.

Once upon a time, Microsoft used to fill its data centers one server at a time. Then it bought them by the rack. Now it’s preparing to load up servers by the shipping container.

Microsoft is close to announcing yet another data center, Josefsberg said. The software maker also has signed a memorandum of understanding to build a data center in Russia.

Only a couple of years ago, Microsoft was adding capacity one server at a time, adding individual servers to racks and taking a couple of hours to wire in each new server.

“When you think about large-scale data centers there are a number of limitations in the technology,” he said. “Some of the network protocols were designed years ago…Some are 30 years old.”

Josefsberg said the figures were accurate, but out of date, reflecting where things were at a year or 18 months ago.

And not all of his problems will be solved just because Microsoft can now get its servers by the containerful. Microsoft has sophisticated “heat maps” that plot the best locations for new data centers based on everything from government policy to water supply to power prices. But in other areas, such as networking technology, Microsoft is counting on the industry to make some quantum leaps.

(Credit:
Microsoft)

“That’s way too expensive, way too slow,” said Josefsberg.

He points to things like Microsoft’s work with the Olympics as indicative of the kinds of demands his data centers will see in years to come.

Correction, 9:31 a.m. PDT:
This story cited the wrong state for Microsoft’s Quincy data center. That facility is located in Quincy, Wash.

It’s just one way that Microsoft is trying to cope in a world where it adds roughly 10,000 servers a month.

“We are not physically building our servers, but there is very deep engagement (with the computer makers),” Josefsberg said.

Over the past 18 months, though, Microsoft has been on a buying–and building–spree. The company has opened a data center in Quincy, Wash., and will open the Chicago facility, as well as another in San Antonio, Texas, later this year. A facility is due to open in Dublin, Ireland next year.

Microsoft also used to lease much of its space, until it realized that data centers were going to be a very big part of its future as more and more software moved into the cloud. A couple of years back, though, it found itself running tight on capacity and bought two San Francisco Bay Area data centers in which it had been leasing space.

What the servers are serving up
So just what are all these data centers doing? Outsiders got a glimpse into this thanks to a slide included as part of a video that Microsoft put on its Web site touting its environmental efforts. The chart shows search accounting for the vast number of the servers–nearly 80,000 or so–with Hotmail and Messenger distant runners-up in terms of server usage.

“You contain your infrastructure but you also contain the heat that’s generated from the servers,” Arne Josefsberg, Microsoft’s general manager of infrastructure, said in an interview this week. “We are working incredibly hard to improve the energy efficiency of our data centers.”

“Search was a very large portion of our demand in fiscal year 2008,” he said. “Going into this year it is still a very large proportion. It is now not as dominant as it was last year.”

Even a 1 percent or 2 percent reduction in power consumption makes a big difference, Josefsberg said. As it is, Microsoft is trying to cram a whole lot of gear into a small space. While server racks at a Web hosting facility might have power densities of 70 watts to 100 watts per square foot, things are packed far more tightly in the containers, which might be consuming in the thousands of watts of power per square foot.

An aerial view of the San Antonio, Texas, data center under construction.

(Credit:
Microsoft)

The container approach is easiest to implement on the ground floor of a facility. In Chicago, for example, it will use containers on the first floor and more traditional racks on the second level. But Josefsberg said that, though it poses some logistical challenges, the company is also considering using multiple levels of containers at other sites, including at a Dublin, Ireland data center due to open next year.

Huang also attacked Intel’s marketing machine. “Just because they have this enormous marketing budget. Just because they have platforms everywhere in the world. It doesn’t make it right. To take on smaller companies. It’s just not right.”

(Credit:
Nvidia)

(Note: A contrarian take on the graphics market states that Nvidia remains the #1 graphics supplier because approximately 73 million Intel integrated graphics processors (IGP) are unused in systems due to “double-attach” with an Nvidia solution, according to Doug Freedman of American Technology Research. More here at ExtremeTech.)

Intel also has plans to bring out a graphics engine code-named Larrabee that uses “many cores” to take on high-end engineering and scientific applications. And presumably games too.

Nvidia CEO and co-founder Jen-Hsun Huang let rip with a diatribe against Intel at Nvidia’s financial analyst day on Thursday. Huang cited frustration with recent Intel comments stating that discrete graphics cards will become “unnecessary.”

“Claim after claim after claim. They’re just false. They cross the line of fair play,” he said. “Here’s another one. Nvidia’s gonna be dead. Because we’re (Intel) sticking the graphics in the CPU and (Nvidia) will have no place to stick it,” he said.

Because Intel, the world’s largest chipmaker, includes integrated-graphics silicon in most of its chipsets the company has become the world’s largest supplier of graphics chips. Its upcoming Nehalem processors will move the graphics from the chipset onto the same piece of silicon as the main processor. A design that is expected to result in vastly better performance.

Nvidia CEO Jen-Hsun Huang

Huang also mounted an aggressive defense of gaming on the PC–one of the main reasons many consumers opt for Nvidia graphics chips. He began by claiming that Intel graphics can’t run games. “We’re not the only ones saying this. This is Tim Sweeney. One of the most important game developers in the entire world. ‘Intel is incapable of running modern games. Intel’s integrated graphics just don’t work. I don’t think they will ever work.’ This wasn’t said in 1994. This was said on March 10, 2008,” Huang said.

“We don’t typically like to do this. It’s just that we’ve been taking it and taking it and taking it. Every single frickin’ day. Are you allowed to say that word? Every day all over the world. Enough is enough.”

Huang was especially upset about Intel’s claims of boosting integrated graphics performance in the future, saying Intel’s claims paled against what Nvidia will achieve by that time.

When asked to comment, Intel spokesman Dan Snyder said, “Are you surprised? Nvidia’s CEO has been very vocal about their feelings for several months now, so I don’t think any of this comes as a surprise.”

This image of Intel as an unstoppable graphics juggernaut is what Huang takes issue with. What set him off initially was a comment from an Intel graphics and gaming technologist who said that consumers “probably won’t need” discrete cards in the future. Nvidia’s primary business is designing and supplying graphics chips for discrete graphics cards that go into PCs.

“(It’s) one of the most important apps. I play games. A lot more people play games today than before. It’s a big industry. We happen to think games are important. Game developers are important. Game players are important. Online games, important. Retail games, important. First person shooters, important. Simulation games, important. I’m a perfectly grown adult. I’m not ashamed of them.”

The Supreme Court rejected that argument holding first that method claims were not immune from the doctrine of patent exhaustion. According to the Supreme Court, a method can be “embodied” in a product that is designed to practice that method. Thus, the authorized sale of the product embodying the method can exhaust patent rights covering the method. The Supreme Court’s full opinion is available here (PDF).

The U.S. Supreme Court handed a big victory to Quanta Computer on Monday when it held that the doctrine of patent exhaustion barred LG Electronics’ claims against it.

In doing so, the Supreme Court reversed the U.S. Court of Appeals for the Federal Circuit’s previous decision that patent exhaustion did not apply to method claims and extended that doctrine to licenses for products that “substantially embod[y] a patent.” This case is likely to substantially change the playing field for patentees seeking to monetize their patents in a vertical industry value chain.

The primary argument against exhaustion was that the Intel products themselves did not practice LG’s patents. The patents related to specific processes for managing memory that did not take place until the Intel chipsets were combined with the non-Intel memory and buses.

The Quanta case has big implications for patentees seeking to make money through vertical licensing schemes, that is, licensing the same patents to multiple parties at different levels in the supply chain. Indeed, the Quanta decision should substantially restrict that practice. Patentees will need to be more mindful about who they license to and the impact of each license on subsequent licensing activities.

The doctrine of patent exhaustion provides that the initial authorized sale of a patented item terminates patent rights to that particular item. What that typically means is that if Company A has a patent covering some gizmo and it sells that gizmo to Company B, Company A cannot sue anyone else for the subsequent use or sale of that particular gizmo. So if I buy the gizmo from Company B, Company A can’t sue me for using the gizmo because its patent rights in that gizmo were “exhausted” when it sold that gizmo to Company B.

Patent exhaustion also applies when the patentee licenses another company to make a product. Those are the facts that were at issue in the Quanta case. There, LG had licensed some of its patents related to memory management to Intel. But, according to its terms, the license did not extend to third parties that combined Intel’s licensed products with third-party components. The practical effect of this licensing scheme was that Intel was off the hook, but LG could still go after downstream users of Intel products for patent infringement.

Quanta was one of those downsteam users. It purchased licensed microprocessors and chipsets from Intel and used those parts with non-Intel memory and buses. LG sued Quanta and asserted that the combination of the Intel products with non-Intel memory and buses infringed LG’s patents. Quanta argued that those patents were exhausted as to the Intel products Quanta was using.

On a side note, the Quanta decision continues the Supreme Court’s recent trend of reversing the Federal Circuit in unanimous or near-unanimous decisions.

What Randy is arguing, however, is that there is a clear distinction between the service ecosystem approach of Amazon Web Services (which he calls an infrastructure Web service) and a more utilitarian infrastructure-focused cloud service such as the ones many of the hosting companies-turned-cloud providers have produced, including GoGrid, Flexiscale, and Rackspace CloudServers. He calls those companies providers of “cloud centers.”

Personally, I think Randy is on to something here, though the distinction may be somewhat short-lived. At the very least, my own experience has clearly been that so-called cloud centers do present much more focus on managing infrastructure components as if they were familiar elements of any data center. In fact, the NAS-like service GoGrid recently began offering, Cloud Storage, is much more like the network-attached storage that most system administrators would be used to managing than is AWS’ Elastic Block Storage, which allows only a single Elastic Compute Cloud instance to connect to each storage unit.

Cloud centers focus on making your cloud infrastructure look very much like infrastructure you already have or are already familiar with, while infrastructure Web services ask you to embrace a new paradigm…Besides the obvious advantage of ‘looking like’ your current data center, cloud centers allow for strategies like using the cloud for off-site disaster recovery.

To briefly recap the cloud market for context, commercial cloud computing has traditionally been seen as consisting of three distinct offerings:

At the very least, this is a very smart piece of marketing by GoGrid, given the current market conditions.

If there is a market for more “data center-like” infrastructure, I think Amazon could address it in a year or less
As developers begin to design for the cloud, taking advantage of existing service-oriented architectures to do so, those infrastructure Web services begin to be more and more important. If those services aren’t available from the Internet itself on an infrastructure-independent basis, cloud centers may find themselves offering similar services to AWS in just a few years.

However, I freely admit that this is a guess on my part, based on limited past markets with which to compare. It may very well be that the “cloud center” distinction creates a powerful market segment of its own, and that many system administrators, or those with unique architecture needs, will rush to embrace “data center-like” cloud offerings.

To Randy, the difference is striking, and in many cases, a cloud center has distinct advantages over an infrastructure Web service:

The difference is the degree to which managing the cloud service resembles managing a similar infrastructure in an enterprise data center. As Randy notes, the question is whether you want to take a developer-focused approach toward cloud capacity, or a system administration approach.

By separating out the purpose of AWS from the purpose of cloud center offerings, these companies can begin to compete with each other on a much more level playing field, and convince many enterprises that the effort involved in moving existing applications and architectures to Amazon may be much more than it would be to simply leverage a cloud center.

Will it work? I think that it remains to be seen. For a couple reasons, I think the distinction will be short-lived:

Why is the cloud center/infrastructure Web service distinction interesting? Well, it is only fair to give Amazon its due and acknowledge that it is by and far the largest and most successful IaaS offering in the market. Furthermore, its rapid pace of innovation makes it a difficult service to compete with head-to-head, as most other IaaS vendors would probably acknowledge.

Randy Bias, chief technology officer of ServePath cloud offering GoGrid, penned a post recently that raises an interesting distinction within the once uniformly defined infrastructure-as-a-service space.

It will be much easier to model a copy of your current data center to a cloud center than it would be to model a copy onto an infrastructure Web service.

software as a service (SaaS): Complete application systems delivered over the Internet on some form of “on-demand” billing system. Examples include Salesforce.com, WebEx, and Workday.
platform as a service (PaaS): Development platforms and middleware systems hosted by the vendor, allowing developers to simply code and deploy without directly interacting with underlying infrastructure. Examples include Google AppEngine, Microsoft Azure, and Force.com.
infrastructure as a service (IaaS): Raw infrastructure, such as servers and storage, is provided from the vendor premises directly as an on-demand service. Examples include Amazon Web Services, GoGrid, and Flexiscale.

What do you think? Is there room in the IaaS market for a distinction between infrastructure Web service providers and cloud center providers?

•  What should the revenue split be?

•  With the auto industry on the verge of collapse, there’s pressure to come up with another multibillion dollar bailout.

The idea was to jointly leverage resources and share risks in a bid to resolve common manufacturing problems. Seven years later, U.S. firms were again enjoying market share leads and by 1996 there was no longer any need for government matching funds.

Time to get out ahead of this crisis before the stuff really hits the fan. Don’t you think?

On the surface, none of this is encouraging news for the near-term prospects of start-ups, which are the lifeblood of the tech business. So it was that last month Ron Conway of Sequoia Capital counseled start-ups to hunker down and cut their burn rates until this squall lifts.

In the mid-1980s, the U.S. semiconductor business was in such a parlous state there were concerns the industry would not survive the competition with Japan. So it was that a public-private partnership formed. Some 14 U.S. chip companies and the U.S. government created a consortium called Sematech (SEmiconductor MAnufacturing TECHnology), in 1987.

And this was during the Reagan administration, hardly a proponent of free-spending, federal involvement in the private sector. But the leadership of the day acted in its enlightened self-interest and made an exception with Sematech. Obviously, the current conditions are not completely analogous.

•  And if there was agreement on some kind of tech start-up super-fund, how would you draw up the rules governing who receives funding and how much?

So you’ve got private sources of capital drying up just as the U.S. enters its deepest financial crisis since the Great Depression. I don’t pretend to have a crystal ball into how long any of this will last. But with a new administration taking power in January, now’s the time to get creative about figuring out possible answers.

With the elections behind us, Wall Street’s bears returned to form by dumping stocks in a big way on Wednesday and Thursday. Out in San Francisco, another tech conference devoted to all things Web 2.0-ish got under way amid dread about the future. And Cisco’s John Chambers cast a further pall when he warned that a sales slowdown has spread from the U.S. to Europe, Asia, and the emerging markets.

OK, it would be complicated but let’s not get lost in the weeds. The details are less important than the need to find a way to keep the funding spigot open. Maybe the private sector can’t do it by itself and maybe Uncle Sam’s not ready to go solo either. But don’t tell me that the best and the brightest from both sectors can’t figure out a way to figure out how to meet midway.

(Credit:
CNET News)

•  How long before a company would need to stand on its own? Etc.

•  The size of the budget deficit is huge and sure to grow.

Given current events, that’s sensible advice. But risk-taking is the sine qua non for innovation in the technology industry and this sort of thing suggests a lousy environment for innovation. In her recent book, start-up veteran Judy Estrin offers a parallel worry: the impact of reduced research spending on U.S. innovation. (The Organization for Economic Cooperation and Development ranked the United States 22nd in the percentage of GDP that nations spend on non-defense research.)

Listen now:

Dell plays defense with Mini 9 Netbook

Also, Spore is finally hitting North America, and GameSpot has an in-depth review.

A year after Asus kicked off the low-cost notebook craze with its Eee PC, Dell, the second-biggest PC maker in the world, is ready with an answer: the Inspiron Mini 9. CNET News reporter Erica Ogg is here to talk about Dell’s move into the world of Netbooks.

Intel ready to announce six-core chip

Sony recalls about 438,000 Vaio laptops

Apple patent application blends touch, voice, face

Amazon flicks on its streaming-video service

GameSpot review: EA’s balancing act with ‘Spore’

Microsoft and 12 others invest in Japanese TV

Yahoo’s stock hits new 52-week low

Download today’s podcast

Today’s stories:

LinkedIn, CNBC team up

(Credit:
PLX Devices)

The PLX Kiwi plugs into a car's diagnostic port to give a read-out of current and historical info.

The PLX Kiwi shows off feedback on how you’re doing in that regard and gives you a “kiwi score.” It’s also programmed with 20 increasingly difficult challenges to optimize your score.

Perhaps we’ll all be learning green driving techniques. The U.K. government earlier this week said that all drivers will need to pass “green-driving tests” in September, according to a report on Wednesday in Green Car Advisor.

Named after the green fruit, the PLX Kiwi from PLX Devices is a small device (a little more than 2 inches by 3 inches) that attaches to your dashboard or windshield.

Making changes to driving patterns can lower gas consumption by 20 percent with gas at $4 a gallon, according to the company. That savings can go up to 33 percent if your driving is mostly on highways.

Depending on the type of
car you drive, the savings can be several hundred dollars a year, between $560 and $800, according to the company. You can get a read-out of dollars saved on a given trip and a cumulative total.

Another in-car display designed for fuel efficiency is the ScanGuage II, which doesn’t have the same slick LED display as the PLX Kiwi or the scoring system.

(Credit:
PLX Devices)

For the eco-conscious driver, these foot pedal changes can add up: a sedan driver can reduce carbon dioxide emissions by 16,000 pounds. That’s the equivalent of changing out 500 incandescent light bulbs for more efficient compact fluorescents.

By plugging into a vehicle’s diagnostic port, it can display miles per gallon as well as trip information and even the cause of engine problems.

The PLX Kiwi in-car display gives you fuel-efficient driving feedback.

But really it’s designed to make your driving habits more energy-efficient.

Accelerating hard and jamming on the brakes to slow consumes more gas than a smoother, steadier way of driving.

The PLX Kiwi will be available later this month and costs $299, according to a company representative, while the ScanGuage costs $169.

Summer road trips are a bit less fun with gas over $4 a gallon and no peak in sight. But there are a couple gadgets to help make fuel go farther.

View results

Cauz also told the Herald that the big reason for the change is to improve Britannica’s search engine optimization and dethrone Wikipedia from its current dominance on Google.com and other search engines. These edits will also trickle down into the print edition, benefiting those who purchase its 30-plus volume set.

In an interview with the Sydney Morning Herald, Encyclopedia Britannica President Jorge Cauz said that the encyclopedia’s free, online version, Britannica.com, will soon be allowing user edits and additions to its pages. This system is not yet live, but according to the Herald should be within the next day or so.

Registered users will be able to make corrections, or add entirely new sections to encyclopedia pages–much like Wikipedia. The big difference, however, is that Britannica.com’s editing and approval system will be managed by its own editors and contracted staff instead of power users. According to the Herald, Cauz is promising a 20-minute turnover on these edits, but that number could go up dramatically if the company cannot anticipate a large influx of edits at once.

One that’s policed by users (Wikipedia-style)
One that’s policed by site staff (Britannica-style)

News.com Poll Encyclopedic reliability
Which Web encyclopedia content policing system do you prefer?