Microsoft Research builds tool for multilanguage Wikipedia editing

Microsoft Research India has built a Web-based tool that lets people edit Wikipedia articles across multiple languages, allowing the quick filling-out of entries that may be longer in English than in another tongue.

The WikiBhasha tool, which is now in beta, is a MediaWiki extension, a Wikipedia user script and a bookmarklet, Microsoft said. Right now, it works with Internet Explorer 7 and IE 8 on Windows and Firefox 3.5+ on Windows and Linux Fedora.

The translation is done by Microsoft Translator, which supports 31 languages, but WikiBhasha only supports translation from English. Wikipedia text can be translated automatically, then tweaked.

A bunch of information is here. Below is an explanatory video featuring Aishwarya Ratan, a principle researcher at Microsoft Research India.

Microsoft Research shows off "Pen + Touch" input (VIDEO)

Sure, we've seen plenty of multi-touch implementations on the Microsoft Surface and other Microsoft projects (read Courier), but this one, brought to our attention by Long Zheng of iStartedSomething, breathes new life into the already creative world of multi-touch. While pen and touch each offer their own powerful capabilities, the combination of two presents endless possibilities. Microsoft is calling it "Manual Deskterity"--an obvious pun on "manual dexterity."

In the video below, you will see Microsoft demoing the use of fingers and a pen at the same time. Notice how the Surface is able to distinguish between the two. In the real world, doing both actions together is a no-brainer. Microsoft is aiming to recreate that natural interaction within its software. Such power will enable users to do what was never before possible in a digital world. This concept would work amazingly with Microsoft's upcoming Courier--an obvious choice for such an implementation.

 

Microsoft SenseCam Review: What It's Like to Record Your Whole Life

Imagine a format that lies somewhere between photos and video, and a device that takes that format automatically, without you having to click a button. Microsoft's SenseCam is a prototype that hangs around your neck, lifecasting everything you see.

Around three years ago, lifecasting was all the rage. We idolized iJustine, Justin.tv and the countless other people brave enough to film every action, every day. The trouble with their form of lifecasting is that it's done via a camcorder strapped to a hat, filming all your actions plus everyone else's. You could almost say the SenseCam is the lifecasting device for shy people who are merely interested in jogging their memory at a later date; people who want to tell a story without having to hear themselves.

So, What The Heck Is It?

Measuring the size of a square pack of cards, it hangs from a lanyard around the neck and films everything within your eyesight in 640 x 480 resolution photos, compressing them as JPEGs on the device's internal 1GB SD card. It can store over 30,000 images—which works out to around 100 hours' worth of lifecasting, based on approximately 300 photos taken each hour (which is the average number automatically shot), plus the time each photo was shot at.

Its 0.3MP VGA image sensor may not be as good as your cameraphone or even laptop webcam (though it does shoot in a wide-angle fish-eye style effect which I loved), but those devices require you to click a button every time you want a photo taken. The SenseCam takes photos passively, based on changes to the light, temperature or movement—or you can set it to take photos on a timer instead.

It contains several different sensors—light-intensity and light-color sensors, a passive infrared detector for measuring changes in body heat, a temperature sensor and an accelerometer for detecting movement. It's certainly interesting moving between rooms with different lighting conditions, and seeing how many more photos the SenseCam takes.

 

Every Step You Take, It'll Be Watching You

When connected to a computer it pops up as an external hard drive, with individual folders dedicated to each batch of 100 photos, or roughly 20 minutes' worth of memories. I used the SenseCam over two days, and by the end I had thousands of photos to sort through. Opening all the folders and previewing them on my Mac, I just ran through them quickly, so they turned into something akin to a flipbook. It was shocking seeing how often I open my Twitter tab when working, and how many times each hour I chew on my nails.

Cooking dinner provided the best results. The SenseCam detected the change in temperature and likely the light as well, so took as many as eight shots a minute. Chopping sundried tomatoes turned the shots into a movie when I ran through them quickly on the laptop later—and stirring pasta with a wooden spoon saw my hand move very slightly in each shot.

Sadly—and this is more of a reflection on my life than the SenseCam—none of the photos are really worth showing anyone. In fact, what you see below in the gallery are the only photos I deemed interesting enough. No-one, especially not me, wants to see hours' worth of photos of my laptop screen as I work, flipping tabs and checking email. An alarming amount of photos showed my BlackBerry in front of the camera, as I replied to emails when I was away from my laptop. A good number featured my cat in them.

Here Comes The Fun

But the potential here is huge. Whereas strapping a camcorder to a hat is deemed as intrusive, having a small box the size of a deck of cards strung from my neck on a lanyard is far from it. Filming makes both you, and the people around you, very aware of every action. Affectations are created that way; egos are born. Having a camera that you don't have to control means it's forgotten, so a truer representation of your life can be broadcast—should you choose to put the photos on Twitter, Facebook or Flickr.

As there's no plans for Microsoft to send the SenseCam down the production line (excluding the fact that they've licensed the technology to Vicon, who'll sell it to the medical industry), it's not too important hypothesizing on why you could ever want or need one.

I do wish however that I was wearing it several nights ago when my friend won tickets to the London premiere of Remember Me, and we were stood 5m away from Robert Pattinson (he of Twilight fame). The one shaky photo I managed on my BlackBerry, which while it has a better image sensor than the SenseCam's, was ruined thanks to my nerves and emotions running wild. The SenseCam, while triggered by changes in bodyheat or temperature, doesn't have stage fright when confronted with celebrities-you-really-shouldn't-fancy-but-actually-do.

There's a future here with the SenseCam, if Microsoft can find the right partner to license the technology to for personal use. They could even launch it successfully themselves. I wouldn't use it everyday, and certainly have no need for reviewing 100 hours of my life through the form of 30,000 photos, but it'd be great fun to wear while at a party—especially for those of us who often suffer from memory loss the next morning. Adding a 3G chip and GPS, so each photo could be sent to an online profile and tagged with your whereabouts would be future features I'd like to see...but then, who would be interested enough in viewing someone else's life from their perspective?

Innovative device with huge potential

Easy to use, easy to transfer to computers

Fish-eye effect is fun

It'll probably never see the light of day in Best Buy or on Amazon
Photos could be higher-res, admittedly

Microsoft researchers working on Mobile Surface projector

You know the downside of the Microsoft Surface table? It's giant, heavy and expensive. Wouldn't the average consumer be more apt to get excited about a smaller, lightweight, portable version?

Microsoft researchers are working on just that. It's called Mobile Surface, and it's a hand-held device with a camera-projector system. It projects a screen you can interact with onto any surface.

From the project's Microsoft Research page:

It is a novel interaction system for mobile computing. Our goal is to bring Microsoft Surface experience to mobile scenarios, and more importantly, to enable 3D interaction with mobile devices. We do research on how to transform any surface (e.g., a coffee table or a piece of paper) to Mobile Surface with a mobile device and a camera-projector system. Besides this, our work also includes how to get 3D object model in real-time, augmented reality and multiple-layer 3D information presentation.

Microsoft ostensibly is showing this off at TechFest 2010, which officially starts Tuesday. The main components are similar to the Bonfire system that Intel Labs Seattle researchers showed off in September.

Via ZDNet

Microsoft Research Project – Mental Tagging of Images

Microsoft Research is working on a project which will allow users, in future, to tag images mentally.

The team consisting of Desney Tan, Ashish Kapoor of Microsoft Research, Redmond and Pradeep Shenoy, Univ. of Washington, Seattle hope to explore in the realm of human-computer interfaces.

The work at Microsoft Research may provide a better means of tagging images.

One can just picture the “tagging farms” now – vast rows of people sitting at computer screens looking at images while they work on other jobs.

Details at Singularity Hub | Downlod pdf from Microsoft Research.

Microsoft Research finds new use for screensavers

When screensavers were first introduced in the 1980s, they were designed to prevent burn-in on CRT monitors. In the LCD era, they’ve largely been used as a distraction and minor security feature, but Microsoft now has a new potential use in mind: reminding you of stuff that you’ve forgotten you ever did.

An ongoing Microsoft research project to discover how to digitally archive every aspect of an individual life — capturing photos, videos, PC history and even TV viewing habits and archiving them automatically — had found that using a screensaver was one of the most effective ways to browse the massive volume of content that results.

“There’s so much stuff you can’t remember to look for a lot of it, and the screensaver brings it up,” Microsoft Research senior researcher James Gemmell told the Storage Visions 2010 conference in Las Vegas. “This could be the killer app of the lot.”

Read more at APCMag.

Microsoft's Research Lab Releases Chemistry Add-in for Word 2007

Chemistry Add-in for Word 2007 is a tool for the authoring and rendering of semantically rich chemistry information in Word 2007 documents. This project is a partnership between Microsoft's Research Lab's and Peter Murray-Rust at the University of Cambridge.

Download Chemistry Add-in for Word 2007

EntityCube from Microsoft Research

EntityCube is a research prototype and is a test bed for exploring object-level search technologies, which automatically summarizes the Web for entities (such as people, locations and organizations) with a substantial presence.

The Chinese-language version is called Renlifang.

The need for collecting and understanding Web information about a real-world entity (such as a person or a product) is currently fulfilled manually through search engines.

However, information about a single entity might appear in thousands of Web pages. Even if a search engine could find all the relevant Web pages about an entity, the user would need to sift through all these pages to get a complete view of the entity.

EntityCube generates summaries of Web entities from billions of public Web pages that contain information about people, locations, and organizations, and allows for exploration of their relationships. For example, users can use EntityCube to find an automatically generated biography page and social-network graph for a person, and use it to discover a relationship path between two people.

Please note that Microsoft is still working on improving the accuracy of the key machine learning problems including entity extraction, name disambiguation, entity ranking, and relationship extraction, as well as looking at a better way of incorporating user feedback.

Microsoft Research to double Cambridge space?

Microsoft Research Ltd is believed to be close to agreeing a deal with the University of Cambridge to virtually double its space on the institution’s West Cambridge site.

Microsoft Research's Cambridge lab, named after its first director, Prof Roger Needham

 

 

Microsoft needs extra space to facilitate its rapid growth from Cambridge and reports from the local property sector suggested it would move off the West Cambridge site to another local base if it failed to negotiate suitable terms with the university.

While neither the University nor Microsoft is prepared to comment, we understand an agreement is now imminent. Cambridge was Microsoft Corporation’s first research base outside of its Redmond headquarters.

It was set up in July 1997 with three researchers. Today over 100 researchers, mostly from Europe, are engaged in computer research at the lab.

Microsoft Research was one of the first residents at West Cambridge, the University's science and technology campus.