VMware Unveils its Cloud OS

VMware Unveils its Cloud OS
(from DMXZone.com)

“VMware is attempting to bridge virtualized data centers-now known as “private clouds”-and growing cloud computing services from the likes of Amazon.com and others.”

VMware announced its cloud operating system-dubbed vSphere 4-with plans for general availability in the second quarter. VMware is attempting to bridge virtualized data centers-now known as “private clouds”-and growing cloud computing services from the likes of Amazon.com and others.

However, this bridging process is a work in progress due to the lack of standards. VMware’s big pitch is that vSphere can run your data center and allow you to bridge out when external resources are needed.

There’s a big gap between what most people talk about as cloud and what people are doing today in the enterprise. VMware’s plan is to get cloud providers to use its operating system and then seamlessly hook up to enterprises using vSphere 4.

It’s unclear what happens if a vSphere shop isn’t hooking up to another VMware powered cloud. ?he company is working behind the scenes on application swapping among clouds but didn’t have details or timelines for such standards. It is clear that VMware sees vSphere 4 as a way to thwartboth Microsoft’s cloud OS, Azure, and its virtualization effort, Hyper-V.

Gilmartin(director of product marketing) argued that Microsoft’s approach with Azure requires too many architecture changes for enterprises. He also noted that vSphere will support more operating systems.

Features in vSphere 4

* A 30 percent increase in application consolidation ratios.
* Up to 50 percent in storage savings by allowing virtual machines to only use storage as needed.
* Up to 20 percent additional power and cooling savings.
* vSphere 4 scales better with the ability to pool 32 physical servers with up to 2,048 processor cores, 1,280 virtual machines, 32 terabytes of RAM, 16 petabytes of storage, and 8,000 network ports.

Google voice search mobile applications

Analysts believe accurate voice search could boost Google’s business

The company’s VP Engineering made indicative comments at a discussion during the Web 2.0 Expo in San Francisco.

“We believe voice search is a new form of search and that it is core to our business,” said Vic Gundotra.

While voice search as a concept is not exactly a new form of search (see my 3 year old Sony Ericsson mobile) the accuracy and effectiveness of processing voice inputs may be at last achieving relevance as a software application. SearchEngineLand editor Greg Sterling agreed: “If done right, it could be a valuable strategic feature for Google.”

Mr Gundotra acknowledged to the audience that “voice recognition in the early days was a nice trick but not very usable”.

There were early complaints that Google’s offering could not understand a broad range of accents (other than ‘Californian’) and that results were often garbled.

“Look how far we have come. I get the advantage of looking at daily voice queries coming in and it’s amazing. It’s working. It’s reached a tipping point. It’s growing and growing very, very fast and we are thrilled about it,” said Mr Gundotra.

He declined to share figures about just how many queries the company deals with via voice search.

However, Mr Gundotra did say: “It’s one of those technologies we think gets better with usage.

“We launched it on the iPhone and have seen a 15% jump in accuracy because, as more people use it, we collect more data and our accuracy gets better.”

‘Queen’s English’

In 2002, Google Labs introduced a service that allowed users to search with a simple phone call. The company admitted it “wasn’t very useful because the results were displayed on your computer and Google discontinued it”.

Six years later, the search giant introduced an improved feature under the Google Mobile App for the iPhone.

Vic Gundotra says “voice search is core” to Google’s future mobile plans

It is also available on the Android based T-Mobile G1 and last month was introduced on the BlackBerry as a free download. The New York Times’s Gadgetwise blog rated the BlackBerry version the “App of the Week” earlier this week.

Early iterations that worked best with North American accents had problems understanding other accents, including British. BBC technology Correspondent Rory Cellan-Jones reported in November last year that his attempts to use it were “pure gibberish”.

For example, his query about the next train, West Ealing to Paddington “delivered some useful information about ‘neck strain’ - but no train times”.

Those problems have since been largely ironed out and Google said it was continuing to work on improving the accuracy of the service. This, Mr Sterling said, is crucial if the company wants it to give them the edge in the marketplace.

“My view is voice search could be a strategic differentiator if it works well. It depends on how much better Google’s system is compared to, say, Yahoo’s or Microsoft’s.

“If they come up with a really great version that is really accurate, it could retain users and likely increase search usage for Google,” said Mr Sterling.

“Stay tuned”

At Web 2.0, Mr Gundotra also talked about a web-based version of Google’s e-mail service, Gmail.

Google’s “technical prototype”, coming soon to the public

He demonstrated a “technical prototype” on the iPhone and the G1 and said “Stay tuned” for a release date.

Mr Gundotra said the prototype used HTML 5, an as-yet incomplete version of mark up language of the world wide web.

He revealed that Google would create a whole suite of offline apps using HTML 5 and that “we are going to be leaders in taking advantage of HTML 5″.

Mr Gundotra also said that engineers were working hard to bring the Chrome browser to the Mac and that while there was no date for delivery, “we are making progress to get it out as fast as we can”.

Thanks to Maggie Shiels - Technology reporter, BBC News, San Francisco- who, really, wrote about 90% of this article

BotJunkie » Artificial General Intelligence For Rent, $0.20 Per Minute

BotJunkie » Artificial General Intelligence For Rent, $0.20 Per Minute.

Coming soon to a telemarketing center near you!

This is something that future AGIs may be able to offer… Especially tantalizing is the possibility that you might be able to find this in computers and household electronics. The AGI behind SmartAction, though, isn’t quite so sophisticated. It’ll be able to do a couple things that current systems can’t do to make your automated call center experience a little more, um, pleasant… Read what improvements you can expect, after the jump.

A few of the system’s features are:

> When a spoken word or phrase is unclear or ambiguous, it chooses the hypothesis that is most likely or appropriate given the context of the conversation, just as people do. Example: It correctly understands red or read when the expected answer is either a color or an action.

> It expertly handles synonyms and disambiguates pronouns. Example: it recognizes the myriad ways how people say yes (yes, yeah, yep, sure, absolutely, for sure, jez, definitely, etc.) and no (no, nope, nah, no way José, etc.)

> It dynamically optimizes the speech recognition engine and provides personalized responses by continuously understanding the context of the call and remembering the caller’s preferences, previous calls and other relevant data. In subsequent calls, callers don’t have to answer the same questions again, greatly enhancing their experience. If a call is interrupted, it can call the customer back and pick up the conversation where it left off.

Related articles by Zemanta

• Voicent IVR Studio 1.0 (acakadut.com)
• Blogging at the speed of thought (u-g-h.com)
• NEC develops speech Interpretation software for cell phones (crunchgear.com)
• The Singularity Summit 2008 (laughingsquid.com)
• Kurzweil to found ‘Singularity Uni’ in Silicon Valley (theregister.co.uk)
• Will Singularity Result in Integration of the Human Nervous System and the Mobile Networks and Interfaces? So You Are Your IPhone. (futurelab.net)

Facebook: Real-time visualization of global social interactions

Project Palantir is a real-time visualization of global interactions between Facebook users. Created in Java using the Open Source Ajax framework, the visualization offers a beautiful perspective of how and where, and at what scale, Facebook serves as a platform for global connectivity.

Project Palantir was created by Jack Lindamood at the last Hackathon Project.

USB 3.0 and Uncompressed 1080p Transfer at 450MBps

According to Engadget, we’ve all been adequately teased with what USB 3.0 will be able to do. Recently, a number of companies took the chance in San Jose, California to demonstrate just how quick the protocol is. Most notable, Engadget writes, was the demo by Synopsys, which prototyped an HDTV video transmission system based on USB 3.0. Onlookers were ‘widemouthed’ when viewing an uncompressed 1080p feed at 30 frames-per-second being transferred at around 450MBps. The supposed performance of wireless HD would be a real competitor to USB 3.0 for this and other applications, but given that wireless HD technology is “on track for an August 2298 release,” there is plenty of time for USB 3.0 to make a real impact.

Source: Darren Murph, posted Nov 19th 2008 at 5:47PM on Engadget.com

Google patent for ocean-based sustainable data centers

Google filed a patent in February that documents a concept for designing data centers that could be powered and cooled by the ocean. The centers are shaped like a submerged pontoon, and could be located in large lakes or 3 to 7 miles off shore. The pontoons would require depths of around 50-70 meters. They could be a sustainable alternative to DC’s that are powered off the grid (or back-up batteries) as they could be powered by wind, wave and solar energy, and cooled by low temperature waters. There may be potential for OTEC power, an modular islands such as Energy Island may be suitable as housing or ‘docks’ to tether the data centers. Ocean based data centers may even be able to tap directly into submarine optical fiber cables.

The design has financial advantages beyond sustainable power: Offshore centers would presumably avoid property taxes, and building rights may be more easily obtained than those for existing data centers (which often occupy 2 or 3 football fields of land).

Larry Dignan (Editor in Chief of ZDNet) notes that wild-cards abound:

Jurisdiction issues will occur. Are states really going to allow Google or anyone else place these pontoons offshore without some tax hit? And the logistics of making this rollout happen are daunting. However, Google has the capital to make it work. More importantly, Google’s ocean data center model can scale. Once the first efforts are deployed best practices will emerge quickly.

Visit here to see the patent filed by Google.

Some considerations:

-Who will ensure the security of these floating data centers (from natural disasters, terrorism) and how?
-Could these structures have adverse effects on ocean currents, ocean temperatures or marine life?

New Chips Poised to Revolutionize Photography, Film

For the first time, professional-grade single-lens reflex cameras are gaining the ability to record high-definition video. That capability, photographers say, has the potential to transform both still photography and moviemaking

read more | digg story

Man Patents Entire Internet, Sues Digg, Google, Yahoo, more

Sheldon Goldberg, the infamous patent hoarder that made the EFF’s Top 10 Most Wanted (http://w2.eff.org/patent/), has made headlines again. This time, he’s claiming that 2 of his patents cover the basic network architecture that comprises the internet as we know it today. And he means business; he’s going after the biggest websites in the world…

Sheldon Goldberg, the infamous patent hoarder that made the EFF’s Top 10 Most Wanted (http://w2.eff.org/patent/), has made headlines again. This time, he’s claiming that 2 of his patents cover the basic network architecture that comprises the internet as we know it today. And he means business; he’s going after the biggest websites in the world, naming eBaum’s World, Google, Youtube, Yahoo, AOL, Digg, The New York Times, The Washington Post, CNET, The Weather Channel and others in a lawsuit filed originally last year. (http://dockets.justia.com/docket/court-txedce/case_no-2:2007cv00263/case_id-103830/)

The 2 patents in question are nos. 6,712,702 (Method and system for playing games on a network) and 6,183,366 (Network gaming system). Don’t let the gaming related patent names fool you, he’s spinning these patents to encompass all aspects of basic networking. Here are some of the ridiculous claims found in these patents:

‘702 Patent - Claim 53
“An apparatus for a service on a communications network”

‘366 Patent - Claim 1
“An apparatus for presenting one of products and services while providing an interactive informational service on a network”

To make matters worse, he filed the suit in the Eastern District of Texas, a division notorious for their favoritism toward patent hoarders:

http://www.techdirt.com/articles/20060203/0332207.shtml

http://www.technologyreview.com/InfoTech-Software/wtr_16280,300,p1.html?PM=GO&a=f&a=f&a=f&a=f&a=f)
We hope that the Judge in East Texas will realize that you can’t simply turn a terribly antiquated patent for blackjack over a network into a patent on networking as a whole.

read more | digg story

Cuil.com and the search for relevance…

The newest kid on the search engine block has a clean, sexy look and layout. Cuil.com (presumably pronounced ‘cool’ or perhaps ‘kyewl’?) proclaims itself to search more pages on the internet than anyone else. “Three times as many as google, and 10 times as many as Microsoft,” according to the about us page. Perhaps their software has the tools to quickly become this large, but my experience suggests they have not yet achieved this.

When I searched for ‘UltraFuture’ - the results were 3,760 pages for Google, 225 for Cuil. Google may have had a number of duplicates, but the first page of Google at least shows me the ultrafutureworld.com homepage (which, frankly, is very relevant to my search). Cuil did not show the homepage in any of my search results.

Self-interest and obvious bias aside, a search for BMW resulted in 57 Million for Cuil, 271 Million for Google.  Again no contest. However, the Cuil.com search did feature an interesting ‘Explore by Category’ feature that allows you to refine your terms to provide you optimized relevance before performing the actual search. Again, this feature looks great and appears to be useful.

Cuil suggests that its software searches by ‘content and relevance’, as opposed to ’superficial popularity metrics’.  Perhaps they simply haven’t been around long enough to collect all the pages they need to be as large as Google. They also claim to not collect any personal data, believing that the web should be searched rather than peoples habits.

Very nice looking site, with a nice interface and experience. While their scale is not yet quite what they suggest, privacy is one position that Cuil.com may be well-advantaged and well-advised to compete in. Let’s check back in a few months.

Google Announces Friend Connect. NOW 2.0 has arrived…

May 12, 2008
Related Article: Social:Learn - Education moves towards 2.0

Previewing Google Friend Connect: Website owners can make any site social

Easily insert social features to make “any app, any site, any friends” a reality

MOUNTAIN VIEW, Calif. (May 12, 2008) “ Tonight at Campfire One at the Googleplex (http://code.google.com/campfire/), Google Inc. (NASDAQ: GOOG) will announce a preview release of Google Friend Connect, a service that helps website owners grow traffic by enabling any site on the web to easily provide social features for its visitors.

Websites that are not social networks may still want to be social — and now they can be, easily. With Google Friend Connect (see http://www.google.com/friendconnect following this evening’s Campfire One), any website owner can add a snippet of code to his or her site and get social features up and running immediately without programming — picking and choosing from built-in functionality like user registration, invitations, members gallery, message posting, and reviews, as well as third-party applications built by the OpenSocial developer community.

Visitors to any site using Google Friend Connect will be able to see, invite, and interact with new friends, or, using secure authorization APIs, with existing friends from social sites on the web, including Facebook, Google Talk, hi5, orkut, Plaxo, and more.

To illustrate, independent musician Ingrid Michaelson has added music features from iLike with Google Friend Connect and is now able to run the iLike OpenSocial application on her official website (www.ingridmichaelson.com). As a result, starting tonight, fans who visit Ingrid’s site can connect with their friends without having to leave the site. Visitors will be able to see comments by friends from their social networks, add music to their profiles, see who is attending concerts, and enjoy other features of the iLike application, all at Ingrid’s website. With Google Friend Connect, people will be able to enjoy their favorite features with their friends on any website across the web.

“We want to bring ourselves to every eyeball, not bring every eyeball to us,” said Hadi Partovi, President of iLike. “Friend Connect is a significant opportunity for iLike, artists, and fans. The iLike Artist Dashboard will be the first content-management system that allows artists not only to post their songs, concerts, and videos to every leading social network from one dashboard, but also to simultaneously manage the content on their own websites.”

Google Friend Connect has been developed to lower two barriers to the spread of social features across the web. First, many website owners want to add features that enable their visitors to do things with their friends, but the technology and resource hurdles have been too high. Second, people are tiring of needing to create new logins and profiles and recreate their friends lists wherever they go on the web. Google Friend Connect offers a solution to both these issues.

“Google Friend Connect is about helping the ‘long tail’ of sites become more social,” said David Glazer, a director of engineering at Google. “Many sites aren’t explicitly social and don’t necessarily want to be social networks, but they still benefit from letting their visitors interact with each other. That used to be hard. Fortunately, there’s an emerging wave of social standards — OpenID, OAuth, OpenSocial, and the data access APIs published by Facebook, Google, MySpace, and others. Google Friend Connect builds on these standards to let people easily connect with their friends, wherever they are on the web, making ‘any app, any site, any friends’ a reality.”

For Site Owners: Traffic and User Engagement

Without requiring coding experience, Google Friend Connect gives site owners a way to attract and engage more people by giving visitors a way to connect with friends on their websites.

• Drive traffic: people who discover interesting sites can bring their friends with them, and can opt-in to publish their activities on those sites back into their social network, attracting even more visitors.
• Increase engagement: access to friends and OpenSocial applications provides more interesting content and richer social experiences.
• Less work: any site can have social components without hiring a programming team or becoming a social network.

Google Friend Connect is in a preview release, available tonight after Campfire One on a handful of whitelisted websites. All site owners interested in learning more about Google Friend Connect and signing up for the wait list can visit http://www.google.com/friendconnect/ starting tonight. In the weeks ahead we will be turning on more sites, adding more social applications, and integrating feedback from site owners and developers.

Google I/O
Learn more about Google Friend Connect, OpenSocial, and other social initiatives at Google I/O, a two-day developer gathering about building the next generation of web applications. It takes place May 28-29 at Moscone West, San Francisco. Register now for Google I/O at http://code.google.com/events/io/.

Conference Call Information
Google will host a conference call to discuss this announcement. The conference call will be held on Monday, May 12, 2008 at 9:30 a.m. Pacific Time (12:30 p.m. Eastern Time). To access the conference call, please dial +1 (800) 776-0087 within the United States and +1 (913) 312-1509 from international locations. Replays of both calls will be available until midnight Eastern Time, May 19, 2008 at +1 (888) 203-1112 domestically and +1 (719) 457-0820 internationally. The confirmation code for the replay of the call is 7571843.

About Google Inc.
Google’s innovative search technologies connect millions of people around the world with information every day. Founded in 1998 by Stanford Ph.D. students Larry Page and Sergey Brin, Google today is a top web property in all major global markets. Google’s targeted advertising program provides businesses of all sizes with measurable results, while enhancing the overall web experience for users. Google is headquartered in Silicon Valley with offices throughout the Americas, Europe and Asia. For more information, visit www.google.com.

Next Page »