Tablets: How Would You Solve the Input Problem?

247 Comments

View All Comments

• TPC-Daneel - Wednesday, March 30, 2011 - link

  I don't see that there is an input problem.. A tablet is an augmentation device, not a primary one - if you need to type long emails or documents then you should be using a keyboard. something like a tablet is only designed for shooting off a few lines in an email or maybe posting a comment like this one, and for that the on screen keyboard works flawlessly.
  
  It's not a work device, it's a light interface to view documents, read news, browse the internet, play some casual games if you like and quickly check email - I don't think devices like the iPad were designed with high end word processing in mind, a fact alluded to by the fact that pages doesn't come bundled.
  
  Anything more is expecting too much I fear, and there's already another segment of the portable market for users who need to input large amounts of data.. They're called notebooks.
  
  I tend to slump horizontally on the couch with my knees up, the tablet resting against my legs. It that works perfectly for me..

• samirsshah - Thursday, March 31, 2011 - link

  If you are so much into typing on a tablet get a foldable bluetooth keyboard. And if your so much into intricasies of using mail programs etc, get rid of the tablet, nobody is forcing you to use it. But finally it will be your loss because the new tablet paradigm is going to replace old paradigm your 'trap' input question or not.
  
  I remember similar thoughts of mine when PCs came out. Oh, oh it does not have a good editor like Emacs. But within a year or two it had Brief and with Windows the whole paradigm changed.
  
  Nobody is forcing you to use tablets, carry around a notebook, your muscles will be stronger because of it!!!

• jcandle - Thursday, March 31, 2011 - link

  Tablets and even smartphones are primarily consumption devices. In terms of productivity they will always be limited. But that doesn't mean they don't receive a plethora of input. Chat, games, and even web browsing is bound to generate a large amount of touch and taps to screen and pocket sized keyboards alike. However, as much as any people will text these are short question, statements, and responses not essays or articles. As an owner of many tablet devices including both generation iPads and a Xoom the larger form factor of a tablet does entice a user to desire more productivity. After all a device that consumes so much space and equally as much space when looking for a spot to set it down should be more productive. Phones can be easily pocketed or tossed aside, but a tablet has a bit of heft and slight unwieldiness to it. The "cycle time" between switching from a tablet to another device for productivity is increased thus requiring the tablet to have additional functionality. Currently its function is restricted to an extension of smartphone activities. Solutions to the issue of increased ergonomic function can have many facets. I suggest the following...
  
  + Voice dictation for short input. Voice is great for short queries like searches or selection. Voice is poor for lengthy input where the amount of time spent correcting input can make the process frustrating.
  
  + 3D tactile input. This is not tactile feedback. The ability to sense fingers in space is a requirement for analog touch perception. At the moment touch is evaluated in binary. You either have touched the surface (or approached close enough) or you have not. Sensing in 3D means finger can rest of the tablet surface without registering a touch. Used in conjunction with accelerometers (multiple for more accurate detection of multiple simultaneous events) a tablet can sense the hammer-like action of a finger striking the screen to register a key press. This also means you palms can rest of the screen without worry of misregistration. Tactile feedback can enhance the sensation but is unnecessary for proper text input. If you can touch type without staring at your fingers you can use this input mechanism. Just try it out on a desk. Your fingers move and register in the proper locations when simulating the act of typing. A slight amount of grit in the desk surface provides a semi-comfortable response to your actions. The same texture can be applied to screens to create a similar response with the added benefit of reducing glare with a matte look.
  
  And then there's highly sophisticated options. Many of these options require complex sensor fusion techniques to determine proper input.
  
  + The on-board cameras track your eye movements and use them to infer correct or incorrect input. The system then learns and autocorrects mistakes without user intervention. This is used in conjunction with tactile input to determine event interval and thresholds. As you compose text many people think the words out in their head this also causes slight lip movement. Cameras also measure lip movement to increase accuracy when used with voice augmentation.
  
  + Voice augmentation. The microphone constantly monitors your conversation. If you speak out loud while you type or in place of physical input; the voice input is combined to resolve more accurate text. Voice profiles is also constantly in effect. The system will constantly monitor you, regardless of performed activity on the tablet, visually and auditory for speech patterns. These can be words and phrases you often use; also incorporating slang and other non-dictionary speech for better accuracy. There could be privacy concerns with this method.
  
  + Additional infrared cameras, ultrasonic sensors or laser plane is used to track finger movement to predict finger landing zones. This helps produce the correct text even if the finger does not actually strike the proper position on the screen. The angular and linear displacement between finger movement is converted into a gesture by which the tablet can infer the correct entry of text. This is in conjunction with predictive text.
  
  + Photosensitive LCDs that double as an optical sensor. These could be used in-lieu of additional sensors to detect finger movement and touch events.
  
  + Piezoelectric, ferroelectric polymers, or ferrofluids can be used to create varying degrees of tactile feedback. When used with a polymer topcoat these techniques can result in electronically variable screen surface; this is likely at the expense of visual clarity.
  
  + Polymer substrate LED/LCD can be used for touch sensory, tactile feedback and when implemented with multiple layers can result in a thin and flexible screen that can also sense touch and produce feedback. Its more exotic tech that's just making its way out of the lab.

• gescom - Thursday, March 31, 2011 - link

  "So how would you fix the problem? Is there even a problem from your perspective?"
  
  The problem is that you can't fix it. It's nearly useless for serious tasks. Simple as that - no matter how perfect we'd like it to be.

• handaxe - Thursday, March 31, 2011 - link

  What is a serious task?
  
  I just passed three lobbyists in the hallway using tablets, and I see a lot more using them in committee hearings. They would seem to be perfect for people who need to: (1) refer to a lot of printed documents; (2) write and receive short e-mails; and (3) check websites for calendars, schedules, and changes to bills. Particularly for people who are away from their office for 8-18 hours per day and who have to carry the device with them, and who end up spending much of their time sitting in a hearing, sitting in a meeting, or sitting on a bench outside the chambers (i.e., people who have to use the device on their lap).
  
  To actually produce complex documents, they would, of course, use their desktop/laptop in their office. But doing that is only part of the job.
  
  I think that there are probably many jobs with a similar division between: (1) producing the work (which may well require a full powered computer); and (2) presenting the work, which basically just requires access to the material.

• edge929 - Thursday, March 31, 2011 - link

  I think voice has the most potential to help the odd nature of tablets. Take a look at Microsoft and Ford's Sync software. It's voice recognition abilities are the best I've ever seen. The main problems it has are non-convential pronunciations of normal words. For example, my wife's name is pronounced like Megan but she spells it Maagen. When I say "call Maagen" it looks through my contacts (spelled Maagen) and can't find it. I then have to say "call Mahgen" since it translates the double-A like ah.
  
  Another solution could be ALL suggested inputs, pen, voice, attachable keyboard a la ASUS Transformer, swype, etc. It may be confusing for new users to handle all the various inputs but I'd rather have them all for options. Having to tow around a separate docking keyboard makes the entire form factor of a tablet pointless though. If there's one input action that touch is king of, it's scrolling. The scrolling pad section on notebooks I've used never worked consistently.

• jesterbenvolio - Thursday, March 31, 2011 - link

  Get the gorgeous Apple keyboard/dock unit! If DRAGON(tm) dictation will give you a good first-pass, then correct using the keyboard/dock! If the Apple Store is too pricey, try eBay for a used one; a great design under-marketed and so little-known. I love mine! Fits and functions with the iPod Touch and 3/3GS phones, too.

• Th3rdparty - Thursday, March 31, 2011 - link

  I think the point of a tablet is not to replace a typical laptop but to compliment it. That said, I have barely touched my laptop other than for downloading media since I got my iPad. My use is obviously different in that I don't write very often but for the short emails and web browsing plus consuming content it serves its purpose wonderfully.
  
  To answer your question I think I have a good solution. Make a case for the iPad like the Dell duo tablet/laptop combo. You could use the embedded magnets as a dock so to speak and would only need some small feet on the opposite side to hold it secure. It could rotate all the way around so that when you wanted to use it in laptop mode you would have a Bluetooth keyboard at your disposal or when in tablet mode flip it on the hinge and fold it closed so the back is resting on the keyboard.
  
  In this way it could act as a case when folded closed as well. You could include an invisishield to protect the back from scratches when folded Ike a laptop. The bezel surrounding the iPad could have a small slit for the dock connector. If the Bluetooth keyboard had a lithium ion battery it could be very flat and could also be used as a back up power source for the iPad. Adding an extra 1/2 inch bezel around the iPad would also allow the keyboard to be somewhat larger as well.
  
  The more I think about it the more perfect it sounds. I'm off to kickstarter to get this baby funded. Hopefully no one on here will steal my idea but if they beat me to it than I will buy it.
  
  Oh yea and I wrote this whole post on my iPad :)

• Midwayman - Thursday, March 31, 2011 - link

  Laser projected keyboards might be an answer. I remember a few years back that they were out for PDA's etc. You could easily integrate one into a tablet form factor. That would solve the input issue for desk use with no increase in bulk. Mobile use is another problem. One handed operation is a must since the majority of the time you will be using one hand to support the device. Looking to phones is obvious here, but what makes a tablet different than a phone? Mostly the screen size. You could implement slightly larger versions of various on screen phone keyboards, but that's a lost opportunity. I know I've seen a number of one handed hardware keyboards over the years. Implementing something like that in software would probably be the most efficient, but they aren't simple. In reality you're going to have to have multiple options on a tablet, as no one method will be best for all users, in all usages. I think the real magic will be in detecting how the user is trying to use the tablet and bringing up the right input method.

• TDM - Thursday, March 31, 2011 - link

  I think the best we will ever do is to match the speed of a touchscreen smartphone. Voice doesn't work because there are to many situations where you don't want to or can't use it.
  
  They should make the bezel smaller and "split" a keyboard from a phone (maybe double the middle vertical columns of keys). You type with both your thumbs...