Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror

"Interface-Free" Touch Screen at TED 194

Posted by kdawson
from the audience-goes-wild dept.
Down8 writes, "Jeff Han, an NYU researcher, has recently shown off his 'interface free' touch screen technology at the TEDTalks in Monterey. Some sweet innovation that I hope makes it to the mainstream soon." The photo manipulation interface is reminiscent of "Minority Report."
This discussion has been archived. No new comments can be posted.

"Interface-Free" Touch Screen at TED

Comments Filter:
  • Interface-free? (Score:5, Insightful)

    by WilliamSChips (793741) <full@infinity.gmail@com> on Sunday October 29, 2006 @11:11PM (#16638351) Journal
    How do you not have an interface?
    • by jayratch (568850)
      I guess it means you don't need to specifically design a touch interface- he's got software to automagically adapt to touch based?
    • by hords (619030)
      Sure looked like there was an interface on the top of the Google Earth-like software. It is a very cool technology though.
    • Re: (Score:2, Funny)

      by kfg (145172)
      How do you not have an interface?

      If it's useless; although you can still load it up with chrome and tailfins if you'd like.

      KFG
      • by x2A (858210)
        Did he keep pressing some button under the screen to switch from one demo piece to another? I guess with "no interface" you have no alt+tab...

    • Re: (Score:2, Funny)

      by solitas (916005)
      He doesn't have an interface because he says he doesn't have an interface. That makes all the difference in his world.
    • Re: (Score:3, Interesting)

      by fsterman (519061)

      Exactly, you can't have an interface free interface, we are interfacing with the world. Want some really mind blowing interface design work check out Jeff Raskin's The Humane Interface [amazon.com] Go back to the fundamentals of how humans interact with the world, find where we retain the most information, are the fastest to react, what gives us higher error rates, etc and redisign computer interfaces. Imagine an OS without applications or files. That's what he outlines. This is just another input device.

      Even if

      • Re:Interface-free? (Score:5, Insightful)

        by Atlantis-Rising (857278) on Monday October 30, 2006 @01:38AM (#16639107) Homepage
        I took a gander at that book, and right away as I skimmed the amazon page, I noticed problems. He may be a wonderful cognitive psychologist, but he's no technocrat.

        The whole "Why shouldn't my computer take three nanoseconds to turn on, read my mind, and then never ever have errors!!!?!?one1" thing is a very amateur approach to the problem, if you ask me. Sure, it would be nice, but I'm absolutely sure it's technically impossible.

        To be more specific:
        "There has never been any technical reason for a computer to take more than a few seconds to begin operation when it is turned on."
        I can name half a dozen; power consumption for suspend to RAM, system process cleanup for suspend to disk, disk space storage for suspend to disk, driver software that doesn't gracefully handle failing down to a hibernate state, plug-and-play hardware detection on bootup... not to mention the whole raft of problems that occur when users never shut down and clog their system up by never ending processes.

        The problem with the view he espouses is that it practically requires a suspend-state, when users aren't good with suspend states. It wasn't until Windows XP and the relatively modern (last three or four years) (okay don't flame me I'm sure SOMEWHERE there was a build that had really optimal suspend, but I couldn't find it) linux systems that suspend really started working, and even so, your device drivers really depend on when you can suspend the system and how it restores.

        For example, when I tested Vista on my laptop, the base sound driver would for some reason kill the audio after restore from suspend. It just wouldn't make any noise until it rebooted. When I upgraded the driver, it went away.

        It is, in fact, only recently that we have had flashmem and the concept of keeping your 'bootfiles' on a seperate flash partition to read from for a quick boot has been a realistic and close to mainstream idea for the desktop.

        The same thing comes up here.
        "Why should you have to double-click anything? What does Ctrl+D mean one thing in one program and a completely different thing in another? And what's the point of the Yes/No confirmation if the user is in the habit of clicking Yes without thinking about it?"
        All of those things make sense in the context they are being used in, and they're relatively intuitive. After all, it's not the programmers fault the user is an idiot, especially with something as simple as a yes/no dialog box, as long as the dialog box is written in language comprehensible for the designed userbase.

        • I think books like that are good for programmers, even if they're idealistic. These kinds of books tell us (programmers) what to aim for.
        • by MrNougat (927651)

          "And what's the point of the Yes/No confirmation if the user is in the habit of clicking Yes without thinking about it?"

          It's CYA. If there was no yes/no dialog after something like "Do you really want to format your hard drive?" you can bet that the number of end users accidentally formatting their hard drive would skyrocket, and that there would be some attempt at a class action lawsuit against the developer of the (not free-as-in-beer) OS for malpractice or some such thing. If you have seen this dialog

        • I'd like to remind you that many of the things we take for granted today were considered impossible ideas only a few years ago. As Larry Ellison mentioned in one of his presentations (in 1995) "..in the last 20 years, computing power has increased a million-fold..." Of course, a lot of his predictions were based on the questionable assumption that computing power would continue to increase at the same rate. I think he may have under-estimated it. (At this point, that is probably as likely as over-estimating
        • by kabocox (199019)
          The whole "Why shouldn't my computer take three nanoseconds to turn on, read my mind, and then never ever have errors!!!?!?one1" thing is a very amateur approach to the problem, if you ask me. Sure, it would be nice, but I'm absolutely sure it's technically impossible.

          I think that we could do or try for 1 - 10 second boot times, but it would involve "cheating." It would have to be a suspend state with something like a flash boot drive just so computer instant ons to it's previous state. You need to give the
          • I think that we could do or try for 1 - 10 second boot times, but it would involve "cheating." It would have to be a suspend state with something like a flash boot drive just so computer instant ons to it's previous state.

            I've wondered about this, specifically how to isolate "stale" state (things like network buffers that refer to sessions that are no longer active), and if it's possible to lazy-initialize/restart hardware. You could certainly bring the system back up using a memory snapshot, but you'll al

          • With regard to the errors, I was not thinking of coding bugs. Coding bugs, while probably inevitable, should be minimized to every extent possible, I agree.

            But most 'bugs' are not coding bugs. They're user-error bugs. Spyware, for example, is not a coding bug, (most of the time)- it's a user bug. Remember that old programing adage, GIGO? Garbage In, Garbage Out?

            The reason why computers have such a bad reputation is because there's so much user-garbage going in that we can't help but have garbage coming out.
        • by vertinox (846076)
          The whole "Why shouldn't my computer take three nanoseconds to turn on, read my mind, and then never ever have errors!!!?!?one1" thing is a very amateur approach to the problem, if you ask me. Sure, it would be nice, but I'm absolutely sure it's technically impossible.

          Sometimes, you need people who know nothing of limitations of technology to design them. Of course you end up often re-inventing the wheel and wasting a lot of VC cash if done poorly... BUT!!!

          If done correctly and with the correct amount of "N
        • by ahg (134088)
          "There has never been any technical reason for a computer to take more than a few seconds to begin operation when it is turned on."

          If I recall correctly, my Apple II turned on in about a second and had a command prompt where I could write such useful programs as:

          10 PRINT "HELLO WORLD"
          20 GOTO 10
          RUN

          :-)

          You couldn't save anything to disk or do more useful functions, but the computer started and had a usable shell. Today, you turn on a PC without a usable disk and you get some beeps... I think so
        • by GauteL (29207)
          "There has never been any technical reason for a computer to take more than a few seconds to begin operation when it is turned on."
          I can name half a dozen; power consumption for suspend to RAM, system process cleanup for suspend to disk, disk space storage for suspend to disk, driver software that doesn't gracefully handle failing down to a hibernate state, plug-and-play hardware detection on bootup... not to mention the whole raft of problems that occur when users never shut down and clog their system up b
    • "How do you not have an interface?"

      Though I get your point, the implication is that it's gesture based instead of requiring on-screen input. It's a misappropriate use of the term, but the idea is effectively communicated.
    • by Fordiman (689627)
      It should say 'transparent interface'.

      'cos, seriously... that was REALLY cool.
    • by treeves (963993)
      You did notice how 'interface-free' was in single quotation marks [blogspot.com], didn't you? That's all you need to know.
      It's also why I'll never eat at a place that has a sign that says "Fresh" Seafood, or the like.
  • This was shown on CommandN and commented on and blogged to death since FEBRUARY! GOSH!
  • by skogs (628589) on Sunday October 29, 2006 @11:28PM (#16638411) Journal
    This is an exciting setup...and I agree with his assertion that the OLPC (one laptop per child) is sort of like introducing millions of children to our inane weaknesses instead of our strengths. Really, I know that something like this wouldn't completely remove the need for a keyboard and such for many years, but it is a striking evolutionary step forward.

    Just think how easy all those dramatic situations would have been in the 24th century if the Starship enterprise had some of these!

  • by Anonymous Coward on Sunday October 29, 2006 @11:30PM (#16638425)
    Hand me a doughnut while I work on this would ya? Everyone take a moment and look BETWEEN the keys on your keyboard. Now put that all on your monitor.
  • by streak (23336) on Sunday October 29, 2006 @11:31PM (#16638427) Journal
    Ok, everyone realizes this was recorded in February right?
    Last I checked its the end of October.
    Jeff Han has been covered I don't know how many times on how many sites (probably on Slashdot too - haven't checked the archives yet).

    There's no such thing as no interface btw.
    Yes, you can remove a lot of the mode-switching with different gestures, but there is always going to be some sort of interface to allow you to access other functions.
    In my mind, once you get above about 4 or 5 gestures, things start to become confusing for people again - what was that gesture again? Thus defeating the purpose of no interface.
    • Re: (Score:2, Informative)

      by Tremor (APi) (678603)
    • by Aqua OS X (458522)
      Word. I saw this months ago.

      That said, "no interface" is a figure of speech. Technically, every tool we use has an "interface" of some sort. For example, the interface for a pencil is arguably the wood stalk that's intended to reside in your hand.

      By "no interface" we're really talking about interfaces that are intuitive and reference more natural metaphors of interactivity.
      • by bogado (25959)
        The pencil is actually a good example, many of us "grown ups" don't think twice when we get a pen or pencil to start writing or drawing. So the pencil/paper technology is almost "interfaceless", right? Wrong, just go to watch some children drawing with pencils and you are likely to see them getting the pencil in an awkward way that makes it very hard to have any kind of precision, the paper below is not well fixed and the strength applied in the pencil almost aways end up breaking the tip.

        So we do learn how
    • Re: (Score:3, Informative)

      by catwh0re (540371)
      I've seen various incarnations of this video all through the year. In particular it's often referenced when speaking about Apple computer's gestural patent applications which detail very similar techniques as shown in this video.

      In particular we can see the zooming (into maps, etc) gestures in the patent imagery. http://hrmpf.com/wordpress/48/new-apple-patents/ [hrmpf.com] Although these ideas are not too-new, the related apple patents date as far back as Jan 2005 (through to Oct 06)

  • That has to be the coolest thing I've seen in a long time.
  • in the computer industry.. and yes, the fact that this stuff isn't actually available to buy yet means it aint coming out of the lab. Why? Cause it takes people who are willing to accept risk to turn research into products and not every grad student is into taking risks with their life.
  • Oh for fucks sake (Score:4, Insightful)

    by glwtta (532858) on Sunday October 29, 2006 @11:38PM (#16638481) Homepage
    Sure, it's neat - giant touchscreen with multiple points of contact and gestures that zoom and pan.

    And this is what's going to "change the way we interact with computers"? Odd as it may sound, most application interfaces don't revolve around zooming and panning; there are considerably harder problems left to solve.

    Funny he should mention RSI too, because that keyboard that will free you from the bonds of conformity, and that's displayed on a hard surface, will kill your wrists in a matter of months.

    The thing is great for the Earth-type applications, but that's about it. It's cool technology - why must every innovation promise to change all future computer interactions?

    (btw, if that picture viewer's "Pile of Crap" metaphor is where UI design is headed, I'm never upgrading again. I have my desk to act as a pile of crap, it won't make me feel more comfortable with my computer if it emulates that)
    • by kaan (88626)
      Not only is the keyboard an issue, consider the rest of his body! He's bent over the screen, neck bent to view the screen that's 2 feet below eye level. Any basic ergonomics advise says you should put the top edge of your display at eye level. Anything lower than that and you'll experience neck and back pain. Keyboard-related RSI will go nicely with a stiff neck.

      I swear, if this were from a business selling some new product, I'd say they were trying to boost sales. But he's a researcher. I guess they must b
      • by parc (25467)
        He's giving a presentation. Would you rather see the back of an input device or the presenter?

        Cut him some slack.
    • by jargon (75774)
      It actually reminds me a lot of Raskin's ZoomWorld [raskincenter.org]. "Pile-of-Crap" isn't so good, but having a zoom in interfance for storing data and applications is actually super intuitive and useful.
    • by chill (34294)
      The thing is great for the Earth-type applications, but that's about it.

      Just because you lack vision, don't assume everyone else is just as blind.

      This would provide a great addition for editing video, audio or image collections. I've just scanned in over a hundred images related to genealogy and this could provide an excellent and FAST interface for sorting them. Combine it with audio annotation and decent voice recognition (dictation) and it would save me weeks of work.

      My kids take tons of photos, and or
    • by AusIV (950840)
      I agree. It's a fun toy, and it might lend itself to some useful new applications, but for current computer uses, it's not worth much. Web browsing wouldn't gain much if anything from that kind of interface. Word processing on such a computer would be a complete pain in the ass. Their photo organizing app looked pretty useless, but I could see that being made more practical. Web browsing, word processing, and media organization are the three main things the average joe uses their computer for, and media org
    • Sure, it's neat - giant touchscreen with multiple points of contact and gestures that zoom and pan.

      Yes, it's neat. Hey, we are geeks so that's the point!

      If you want to see some applications that can benefit from it immediately, check out this video [google.com] that shows both Google Earth and Warcraft III being run using this multi touch interface combined with voice recognition.

      I've been using a multi touch interface for over a year now via my PowerBook's touchpad and iScroll2 [sourceforge.net]. Let me tell you that panning around ap
  • Well it makes for a great demo but I notice he didn't actually manipulate any information there, just graphics.

    What I'd be really interested in is seeing some kind of email or office app done this way. I suspect it's much harder to apply these techniques to very data-heavy displays or data based around language rather than graphics. That's not to say it's a bad idea - multi-touch will probably arrive on our desktops at some point, but I see it as being a supplement to what we have now rather than replacin

  • In fact, it absolutey does have an interface. Granted, it's a simple interface, and one that contextually changes with each application, but it's still an interface. Basically, all he's showing is an interface that essentially has two mice, not one, and instead of using your hand to manipulate a physical mouse (which is then translated onto the screen), they've built a complicated touch-screen system to eliminate the mouse altogether. And then they added a second one. Don't get me wrong, this stuff is neato
  • Minority report??? (Score:4, Insightful)

    by www.sorehands.com (142825) on Sunday October 29, 2006 @11:58PM (#16638611) Homepage
    This is not from the Minority Report that was released in 2002. This was shown in shuttle interface in Earth the Final Conflict [tv.com] which aired in 1997.

    This type of interface was also in The first $20 millions is the hardest. [imdb.com] But that came out in the same year as The Minority report.
    • by mqduck (232646)
      This is not from the Minority Report that was released in 2002. This was shown in shuttle interface in Earth the Final Conflict which aired in 1997.

      This type of interface was also in The first $20 millions is the hardest. But that came out in the same year as The Minority report.


      Thanks for clearing that up. I guess this is why the call it "news for nerds."
  • by prof_peabody (741865) on Monday October 30, 2006 @12:04AM (#16638647)
    I recently attended a demo of a similar device at my company. The pentagon already has purchased units and the company is trying to branch out to private sector applications. They were using for collaboration with geographical software (gis data).
  • It's a very cool looking product, and I'm sure there are some uses for it, but I don't see this becoming commonplace by any means. It just doesn't seem to be a really better way to interface with our machines then what already exists.
  • This is good. One of the major problems with graphic design systems, both CAD and animation, is that it's only possible to select one thing at a time. Many operations involve two objects, and you're forced to some sequential select-and-manipulate interface. This gets you past that.

    Many high-end animation systems will accept multiple input devices, from MIDI keyboards to knob boxes to articulated skeletons. At the low end, we have the scroll wheel, which was a big improvement. Finally, you could do tw

  • I think that streaming media is very [buffering]....
  • by Alsee (515537) on Monday October 30, 2006 @12:24AM (#16638757) Homepage
    As he was manipulating the map application it really jumped out at me how cool it would be to run a Mandelbrot set app that way. It would have made a fun and awesome addition to the presentation. If I were working in his lab that would almost certainly be the first thing I would add to the system.

    -
  • The GUI (Score:3, Insightful)

    by mqduck (232646) <mqduck AT mqduck DOT net> on Monday October 30, 2006 @12:48AM (#16638849)
    Very neat and completely useless. I don't know about you guys, but I find it much easier to find my data in my nice hierarchical filesystem than by digging through a garbage can fan full of papers, which is what this GUI simulates.
  • This reminded me of this video [youtube.com] I watched a while back, with Google Earth and Warcraft III on a giant touch-screen display using two fingers to zoom or select units, etc. Pretty cool stuff. Too bad it's probably still to expensive to hope for it to become mainstream anytime soon.

    • by crovira (10242)
      What cost?

      Its a piece of glass that has the image back projected.

      If you have a projector, you're already good.

      The interesting part of this is the multiple detection points.

      There are a few of ways to do this most of which involve stress deformation of a laser beam through the glass.

      Wake me up when he stops having to touch a flat surface.
  • "The only intuitive interface is the nipple. After that it's all learned." Anon [greenend.org.uk]
  • As the guy notes that they should have this in google lobby. More or less it is good for 'removed' visual data manipulation. Nothing serious, like office work. What is missing, something that was learned ago, is safety brackets data removal or change. This interface will allow less technical people to use the computer. What is missing for the demonstration, is critical data alteration. You won't see this being used in the office for spreadsheets, word processing, etc etc - real meat of the computer industry
    • by vidarh (309115)
      You won't see this being used in the office for spreadsheets, word processing, etc etc

      Really? The first thing I thought when seeing it was that this would be perfect for things like Visio or other diagram editors, but also for moving text and images around in presentation apps like Powerpoint. It would also be a great interface for doing things like cut and paste, or for changing text sizes etc.

      The uses may not be huge in all kinds of apps, just the same way as mouse usage varies, but they might mean pr

      • by zoftie (195518)
        Mouse was successful because it was cheap. It did cost alot less then system's value. Such display will only be successful in changing the way people interact with computers, if chinese can crank out a millon of these in say less then 500-1000 range. And the researcher is excited about joes and janes using multipointer display.

        On other note, pressure sensitive displays are prone to wear, since one has to flex material in places. So displays have to be cheap to warrant replacing them now and then for low cos
  • by tygt (792974) on Monday October 30, 2006 @01:00AM (#16638927)
    I realize that the point of this (TFA) is about trying to make things more intuitive and natural. But, as others have pointed out in other words, interfaces are a natural aspect of life.

    I have an interface in front of me right now. I have pen, paper; I've got a camera... if I want to record a visual of something, I have to pick up my camera. Never mind that the camera has one of these "non-intuitive interfaces" that we (rather, the article) are trying to remove, I still have to do something to get it done. Anything that I do interfaces with reality.

    One of the goals of the iconic desktop originally was to duplicate the real desktop in some fashion to make things simpler for humans to interact with their work on a computer, so that there wouldn't be too much of a translation layer to build between real and virtual work. Similarly, some try to implement handwriting recognition to remove the interface of the keyboard from the writing process.... until they realize that geeks like us can't write for crap and can type ten times faster as well.

    Regardless, of course, there's got to be some way to tell the computer that you actually want to resize the strange hand-like object on that screen the guy had (I think it was a hand, my sound was off and I lost interest rapidly) rather than add to the drawing. There's got to be some way to change modes, as he did between drawing the outline, getting it filled in, and then moving it around - that's all interface. Sure, it looked sweet that there wasn't any menu pull-down happening, no mouse, but really, you've got a pretty damn simple application that can be manipulated in this fashion.

    Do anything complex, and you'll have to have a more complex interface suddenly.

    "Computer... Computer... (McCoy hands Scotty the mouse) Aye. Hello computer." -- Scotty

    Even talking to a computer would be an interface..... a pretty complex one, though definitely one that could be considered intuitive, if you could use your chosen language for commanding it rather than some cryptic "ok, list the files, sort by date then name.... uh.... ok that one no that shit fucking computer where's my mouse"

  • Keyboard (Score:3, Interesting)

    by WhatsAProGingrass (726851) on Monday October 30, 2006 @01:22AM (#16639035) Homepage
    His keyboard Idea sounds pretty cool. I would like to see some more practical applications than what he showed. Games would be cool with this interface. I think the idea is great, moving objects on your screen as if they were actually on your desk. But gestures will still need to be learned. Also, we would all get neck problems from staring down all the time at the screen rather than looking straight ahead. All in all, this technology seems very interesting.
    • I envisioned this as having a screen in front of you, and a touch screen at your finger tips. Both displaying everything.
      I don't think it would take much effort at all before you'd rarely have to actually look down and see what you were doing.

      I also think that we'd have to keep an interface, but it could be more like OS X's menu (Or like how he changed things in world wind).
      You would still need to have menus for some things, but for the most part you could remove a lot of the menus.

      I also liked the idea of
  • slashdotted that.

    I think /. covered it when the video was first done. you still need a file manager etc.
  • the linked talk also won the price for " [google.com] The Most Gratuitous Use Of The Word Kinda In A Serious Scientific Presentation".
  • How many times are we going to see this posted? This is getting ridiculous. It was neat the first time. Now it's getting old.
  • Combined with this Pen-based Interface [youtube.com] and the many different ways of interacting with files in this system, we could be on to something extremely cool, functional, and versatile.
  • by hey (83763)
    Why is there no manual (or instructions) necessary to know that touching the bottom
    of the lava lamp adds heat? It doesn't in real life.
  • The real significance of this is that any technique he uses in a video release for public view becomes part of the body of work of prior art that can be used to prevent future attempts to patent anything shown in the video.

    When I first developed, but didn't patent, the graphical point of sale software paradigm in the mid 80's I began a fifteen-year effort to travel the world and show it to thousands upon thousands of people. As a result, none of the aspects of the graphical POS software user interface coul
  • It's past time for /. to warn us when a link goes to a video. The last thing I want is to preload a page in a background tab and suddenly have loud music inexplicably start blaring at me. If pdfs deserve warnings, videos deserve far stronger ones.

An authority is a person who can tell you more about something than you really care to know.

Working...