Microsoft Research Shows Off New Projects On College Recruiting Tour

In a recent college recruiting tour, Microsoft's Craig Mundie was able to showcase some of the experiments coming out of their Research division. Among some of the interesting projects were another pass at the Minority Report interface, eye-tracking, intelligent data sorting, a global carbon-climate model, and several other software and hardware experiments. A video and supporting slideshow are also available via Microsoft's press site. "Mundie also will discuss the kinds of computers students will soon be using – machines that will respond to gestures through new natural user interfaces; deploy the power of new microprocessors; migrate data to the cloud; and use live data to drive new simulations and visualizations. He’ll center on an environmental theme to show what it might be like to be a research scientist working on zero carbon energy in the future using new interactions with data and computers to increase insight."

Comments

Bar of soap mouse

[Enderandrew]

Microsoft's hardware research division ages back demoed a mouse that was akin to a bar of soap. You held it up in the air like a remote control, and rotated around in your hand. It could be operated from a couch or another location where didn't have a traditional hard surface. I thought it was a great idea.

Whatever happened to it?

Re:Bar of soap mouse

[MightyMartian]

No idea. Now shut up, I'm playing Mario Kart on my Wii.

Re:Bar of soap mouse

[Zerth]

No profit in it. Just get a sufficiently round mouse, stick it in a sock and sew it shut. Tadah.

The knockoffs would blow them away.

Re:Bar of soap mouse

[swanzilla]

Link to an Engadget article regarding the device in question...

I imagine it is caught in a development life cycle...or that Microsoft determined that most consumers are content using touchpads or game system controllers to interface from the couch.

Re:Bar of soap mouse

[recharged95]

There's an app for that on the iTunes appstore.

Re:Bar of soap mouse

[BJ_Covert_Action]

One guy dropped it in the shower and it all went downhill from there....

Re:Bar of soap mouse

[zippthorne]

They invented an extra bulky presentation mouse?

Re:Bar of soap mouse

[Enderandrew]

The Wii-mote is a pointer, and a very inaccurate one at that. It is a very different tool from a mouse.

Eye tracking?

[Johann Public]

Wonder if they're hiring people involved in academic research related to that and visual attention...

"Wild" body gestures eh?

[decipher_saint]

So sweaty monkey dancing is really some form of user input device.

I think I'll stick with my keyboard...

Re:"Wild" body gestures eh?

[jayme0227]

I love posts like this, and must admit that I've been guilty of the same thing.

"In it's current state, this technology is clearly outclassed by other technologies on the market. They're wasting their time. I'm not buying it."

Yeah, in it's current state, it looks retarded. But what will it develop into?

A lot of people thought the Eyetoy was stupid and motion control video games would only be a passing fad. Further development of motion sensing technology pointed to using a controller rather than a camera, which was then masterfully executed by Nintendo. Did the tech suck when it first came out? Absolutely. Was it worth forging ahead into the arena of motion control? Indeed.

Actually, now, there's a renewed interest in using cameras to control what's going on in the game.

In review: Yes, new technology is often outclassed by other alternatives before it matures. If you give it time, though, it can develop into something really cool.

Re:"Wild" body gestures eh?

[cbhacking]

Project Natal was actually part of the demonstration. Not only in its current form (no, Mundie did not get up and play a game of 3D-Breakout using his hands as paddles, but he showed a video of it from, IIRC, E3) but also as a PC input device; using gestures in open air to do things like manipulate a 3D model in a CAD program.

I wouldn't want that to become the *primary* input means, but it certainly makes sense as an option for certain types of input.

Re:"Wild" body gestures eh?

[Adm.Wiggin]

Like this?
http://link.brightcove.com/services/player/bcpid713271701?bclid=713073346&bctid=709364416

Bo-ring

[Lije Baley]

No intelligent robots.
No flying cars.
No ray guns.
*YAWN*
NEXT please.

In other news,

[melikamp]

Microsoft released a product that does not suck. Unfortunately, it is a vacuum cleaner.

Re:In other news,

[Itninja]

I don't get it. Vacuums don't suck. It's the outside pressure pushing dirt into it. So technically the MS Vacuum(tm) was a success.

Not necessary

[mraudigy]

Why does Microsoft need a tour showcasing their new research? I thought they already had that coverd with their employees LinkedIn profiles.

He presented at the U of Washington last night

[AVIDJockey]

The video can be found at UWTV.

Re:He presented at the U of Washington last night

[cbhacking]

Nice link, thanks.

It's worth pointing out that this was not in any way an actual recruiting tour. I'm sure it interested some people who may now go apply, but there was nobody from HR there, and nobody was taking resumes or discussing internships. Instead, Mundie was basically traveling through several leading universities to talk with administrators, faculty, students, and the general public. The big presentation (in the linked video) is the "general public" one; although much of the audience was connected with UW in some way, many were not. The meeting with students (of whom I was one) was much smaller (10 of us) and more of a roundtable discussion (mostly consisting of us asking Mundie questions and him answering them, with occasional small reversals of that).

He said the purpose of the tour was to see what students were thinking about, and where they saw technology going in the future.

I went to one of these talks

[AdmiralXyz]

And as a completely impressionable college student (who really needs an internship this summer), I have to say I was pretty impressed.

The two main points of his talk were a) we need to find a better paradigm for organizing the vast amounts of information we're being deluged with and b) we haven't found a good replacement yet for the mouse-and-keyboard model of human-computer interaction (he basically admitted that Microsoft's efforts at speech recognition so far have been a total failure), but we're going to keep trying.

After that he showed a couple of demonstrations. Of course, most of what he showed was probably hardcoded, but it was neat nonetheless. At one point, he just ran a search for "zero-carbon energy" and then used voice commands to sort the zillion results by type of origin (news vs. scientific papers vs. personal files). Later on he demoed a transparent display with gesture manipulation, and it was nice to see Microsoft isn't thinking about the flawed "Minority Report" model where you wave your hands in the air like an idiot, this was much more subdued: he used the surface of the desk he was behind to recognize a couple different gestures. Gorilla arm not included.

So yeah, I definitely recommend hitting up that video and slideshow link if you've got the time. Say what you want about Microsoft, but they aren't just resting on their laurels.

Well, not totally.

Re:I went to one of these talks

[h4rr4r]

The keyboard only model works great.

Of course with your sig I am sure the little unemplyed college student thinks he knows better.

Re:I went to one of these talks

[grcumb]

Arguing about vi versus Emacs is like arguing whether it's better to make fire by rubbing sticks or banging rocks.

Normally replying to a sig results in an automatic -1 Offtopic, but in this case, I think it makes a perfect point.

Here we are discussing the merits of new realms of human interface when we don't properly understand the interfaces we already have. Over the decades, people have tried again and again to reinvent the IDE, in many cases failing utterly, in others just trading one set of features for another.

If you're going to be working in human interface design, it won't help you to disparage the most mature and capable text editors in common use today. Yes, they may seem unappealing and even counter-intuitive, but that doesn't mean that they are not fit for a purpose. In nearly 20 years of computing, I've gone through the same process of chasing UI glitter. But I have learned through countless disappointments that GUIs are not ideal for character entry.

Don't get me wrong. I use Compiz. I've found the 3D rendering of the desktop cube makes it easier for me to visualise my workspace. I'd almost say I need it to work now. I prefer wobbly windows because they seem more organic, therefore part of the world around me. I like the smoother fades when windows open, because the experience is less jarring. I like it when things minimise into their icons, like genies going back into the bottle. All of these things provide a more consistent and visually 'natural' environment.

BUT... when I'm typing - and especially when I'm coding - my hands belong on the keyboard. Every time I lift them, I slow down and more to the point, I lose focus.

I write and code professionally, so I spend a lot of time thinking about my tools. The best word processor I've ever used was WordPerfect 5.1 for DOS. (That's likely because I was just a tad too young for WordStar.) I use vi and emacs for different purposes and, despite have tried Visual Studio and lord knows how many other more GUI-oriented editors, I keep coming back to them.

Again, don't get me wrong. I don't dislike VS. I actually wish that emacs would import the auto-completion functionality that VS has. But here's the thing. For all its bells and whistles, I'm just not as comfortable coding with anything other than emacs. Make as many EXTENDED/META/ALT/CONTROL/SHIFT jokes as you like; emacs is still my coding environment of choice. Because my hands never have to leave the keyboard, not to debug, compile, import other files, diff them or commit them to subversion. Yeah, it takes time to learn. Boy is it worth it.

Likewise vi. Just yesterday, I was working with another silverback programmer. We were rebuilding a website containing tens of thousands of documents in hundreds of different collections. Every single document had to be recreated from source materials, which implies dozens of filters, all context-sensitive. By importing a directory listing into vi and running a few regular expressions against it, we were able to create a shell script to automate the entire build in about a minute and a half. Show me something else that can do that.

I like new interfaces. My friends all make fun of me because of my attraction to shiny. But when it comes to interfaces, I'm not so blind as to throw away my tools just because they aren't pretty enough.

Good interface designers have learned this lesson, and focus on making things simple, but no simpler than they should be. I suggest you take that advice to heart as you back away toward the gate and get the heck off my lawn.

Finally!

[Locke2005]

machines that will respond to gestures through new natural user interfaces This is just what I've been waiting for -- a computer that actually understands when I'm flipping it off!

WOW! CUTTING EDGE STUFF FROM MICROSOFT!!

[DbZeroOne]

...a global carbon-climate model, and several other software and hardware experiments. WOW! GROUND BREAKING STUFF!! NOBODY has EVER thought to make a 'carbon-climate model'!! That's revolutionary!!!!!!! In typical MS fashion, they'll build a mediocre knock-off of what's now an old idea and act like they invented it.

Re:WOW! CUTTING EDGE STUFF FROM MICROSOFT!!

[Locke2005]

they'll build a mediocre knock-off of what's now an old idea and act like they invented it. A business model that has worked exceptionally well for both Microsoft and Disney (who is releasing yet another remake of "A Christmas Carol" -- talk about originality!)

All joking aside here...

[jimicus]

Exactly what truly innovative stuff from Microsoft Research has seen the light of day?

Yes, I know that such research is expected to produce remarkably few results in the real world. But the closest thing to innovation I've seen is the ribbon toolbar in Office 2007.

Re:All joking aside here...

[genghisjahn]

Try this link... http://lmgtfy.com/?q=microsoft+research+technology+transfer This will show you the search results of a website called "Google." I know, it's a funny sounding name. But really, it's quite useful. Next, click the TOP link. From there, select a year(like 2008) link to see what MS research has done lately.

Re:All joking aside here...

[grouchomarxist]

Back in the 90s Bill Gates made fun of how much Apple spent on research and how little came out of the labs.

Re:All joking aside here...

[jimicus]

Back in the '90s Apple nearly went out of business. Today they act as another Microsoft research facility.

"Recruiting tour"?!?

[Locke2005]

Excuse me, but isn't simultaneously spending time and money recruiting new employees and laying off over 5000 people just a tad bit schizophrenic?

Re:"Recruiting tour"?!?

[cbhacking]

First off, I wouldn't call it a recruiting tour. Nobody was taking resumes. Nobody from HR was present. There were no references to internships or other openings. Heck, I spoke to Mundie in person before the talk, attended the talk, and attended the reception afterwards, and nobody (Mundie or anybody else) made any reference at all to recruitment.

Second, the people who were laid off were not Microsoft's engineers and programmers, but instead were people in fields such as marketing and legal affairs. They've never stopped hiring new programmers, a fact which has been made clear on every occasion that the layoffs were discussed.

To the "OSS should stop copying..." crowd

[Quarters]

How often have we heard "OSS should stop copying and start innovating"? Anyone and everyone who ever said that, this abomination is your fault.

Reminds me of a quote by Michael Faraday

[file_reaper]

When the Prime Minister asked of a new discovery, 'What good is it?', Faraday replied, 'What good is a new-born baby?'...

One thing to mention here is

[melted]

One thing to mention here is if you join Microsoft itself (as opposed to its research arm - Microsoft Research), you won't ever get to work on any of the cool stuff. And if you join MSR, the shit you work on will never ship. So it's a lose-lose, no matter where you go, unless you're a researcher and publishing papers gives you a boner.

Why is Microsoft on a Recruiting Tour?

[bezenek]

Microsoft just added 500 more layoffs to their earlier 5,000. Why the heck are they spending money on a recruiting tour?

-Todd