To Touch or Not to Touch, That is the Question

This is an excerpt from an analysis on the strategic errors of Windows 8 and the philosophy behind the product that was written for our Tech.pinions Insiders Members. To learn more about Tech.pinions Insiders click here or to see all Insider topics and articles click here.

Adopting a New Posture

While I was at Microsoft’s build conference last week, I decided to make a point to keenly observe those attendees who have embraced touch on notebooks and watch how they use them. The plus to being at a Microsoft conference was that I saw more touch notebooks, and Surfaces for that matter, in one location than I have ever seen out in one place.

What I observed was interesting. Those who had adopted touch on their notebook would type with the device at arms length, but then move their body and face closer to the screen as they sought to use touch input. In essence to use touch they actually leaned in, performed the action and either stayed or leaned in to scroll a web site for example, and then leaned back to start typing again.

Interestingly, Surface owners had adopted an entire experience built around leaning in. I can only speculate that this is because the screen is so small that staying leaned in closer to the screen makes it easier to read the text, etc. Surface owners would even type with arms bent significantly more because of how close they were to the screen.

Alleged-Microsoft-Surface-Phone-Emerges-in-Official-Photos-2

My key takeaways from these observations were that to use a Windows 8 notebook, or an aspiring hybrid like Surface, adopting touch as a paradigm is one necessary component, but so is adopting new body language to operate it in a useful and efficient way.

So the question we need to ask ourselves is this: Is this better? Does touch bring so much to the notebook and desktop form factor that we should consider this new, somewhat un-natural required body posture worth the effort?

Let’s look at it this way. Is adding touch as a UI mechanism to something like a desktop or notebook a more efficient input mechanism? In notebook and some desktop form factors, I would argue that it is not.

I absolutely condone touch on smartphones and tablets. In these devices touch is natural, and the best and most efficient input mechanism for the use cases they are best at. This is because they are truly mobile and you use natural motions to touch the screen to navigate. But notebooks and desktop are different beasts that succeed at very different use cases for very different reasons.

WHY TOUCH?

What I’ve tried to bring out, both in public and in private, is this: does using touch as an input mechanism on a notebook or desktop make me more efficient in my workflow? I’m yet to find that it does.

When you sit behind a notebook or a desktop you are prepared to get work done. In this context speed, efficiency, and ease of use are keys to make these devices the best tools for the job. So for touch to be compelling, it must be better at the above experiences than a solid trackpad or external mouse. Does it do this? The answer is no.

Take the trackpad for example. My hands have less distance to travel for me to reach the trackpad on all installations. To use a trackpad I bring my hands closer to me a very short distance (maybe 2-3 inches). Contrast that with using touch as an input mechanism and rather than bring my hands in a short distance I must reach for the screen (approximately 5-6 inches). This requires more effort and more time than using the trackpad and is more tiring to the arm, by keeping it fully extended to operate. Unless you hunch over or lean in, which is also uncomfortable for any length of time. I concede that for some the amount of time and effort may not be considered much difference by some, but it is still a key point.

When I discuss this with those who advocate touch screens on notebooks, they propose that touching the device for input is a preferred mechanism to the trackpad. My counter point is that this is because most trackpads put on Windows PCs are downright terrible. Sometimes I wonder if Microsoft pushed OEMs to do this on purpose to make touching the screen seem like a better experience, simply because the trackpad is so bad, that it makes touching the device appear to feel like a better alternative.

I’d like to quantify this sometime by having a race with a Windows user and challenge them to a similar task, like creating a few slides and graphs in Power Point. Them on their touch notebook and me on my MacBook Air. We will see who can finish the task the quickest.

Published by

Ben Bajarin

Ben Bajarin is a Principal Analyst and the head of primary research at Creative Strategies, Inc - An industry analysis, market intelligence and research firm located in Silicon Valley. His primary focus is consumer technology and market trend research and he is responsible for studying over 30 countries. Full Bio

38 thoughts on “To Touch or Not to Touch, That is the Question”

  1. The reason why touch makes sense on a laptop (I’m against its use on desktops of any kind) is that, while ergonomically less energy efficient than using a mouse or a trackpad, it is indeed faster and more precise provided the target is large enough. While trackpads on laptops are universally terrible in Windows, they aren’t the reason touch has started to proliferate. Run this simple experiment:

    Take note of where the mouse is on your screen. Then place your finger on the trackpad and move the mouse to one corner of your screen. Take your finger off of the trackpad.

    At this point, the mouse should visibly be seen in a corner of your computer screen. For the rest of the example, let’s say it is in the upper right corner of your screen.

    Now what needs to be done in order to move the mouse from the upper right corner to the lower left corner? If I place my finger on the lower left corner of the trackpad, will my pointer automatically appear in the lower left corner of my screen? No, I have to slide my finger along the trackpad to move the pointer from one end of the screen to the other.

    However, if I want to move my pointer from one end of the screen to the other using a touch screen, this is what I do (this can be done with Ubuntu on a touchscreen PC and I’m pretty sure it can be done with Windows 8 as well):

    Touch the upper right corner of the screen to place the pointer at that end of the screen;

    Touch the lower left corner of the screen to place the pointer at that end of the screen.

    Though it took far more physical energy to perform, moving the pointer from one end of the screen to the other using a touchpad is far, FAR quicker.

    This illuminates the problems with Windows 8 and even a few basic ergonomic issues with trackpads and mice use in general. When a target is large enough it will almost ALWAYS be quicker, though less energy efficient, to use your finger to navigate to it. The problem is that most of the navigational targets on Windows 8 (particularly desktop mode) and other general purpose OSs are too small. The fact that most general purpose OSs have a mixture of large and small navigational targets creates confusion and reduces automacity. The average user has a hard time because they must constantly switch modes.

    However, trackpads and mice aren’t as efficient as they can be because there is no mechanism that allows for the concept of PROXIMITY. The pointer must always be actively moved from one point to another. This makes sense with mice but why is this the case with trackpads? Why am I not able to touch the right corner of the trackpad and have the pointer automatically appear at that location on my screen? Or move my finger to the left corner on the trackpad and have the pointer appear there on my screen.

    This is simply a matter of making the trackpad more analogous with the screen. Areas on the trackpad should correspond with areas on the screen. This would make trackpads far more efficient and eliminate the need for touch on laptops because the trackpad would not only be more energy efficient but also pretty much just as fast and precise as using a touch screen.

    To sum it up, touch is becoming popular on laptops because, when the target is big enough, it is much faster and more precise than using a trackpad. If trackpads were designed to be more analogous to the actual screen, it would obsolete touch on laptops. Barring that, if UI elements become large enough and consistent in size, touch will supplant pointing devices at least on laptops. The benefits in speed would be too great to ignore.

    1. Very good thoughts James. I don’t disagree with any points. Part of this is an exercise in future UI methods. One thing that has stuck in my brain with this discussion around touch, is whether touch is even the right solution to be investing in for notebook and desktop computing modes, or whether voice, or perhaps 3d space (like leap motion) makes more sense.

      Again for the reasons I point out, I will never use a cursor dominated task, like doing something creative in photoshop or illustrator, etc, if I must stay leaned forward for a period of time with my arms not resting on anything. For quick navigation stuff, as you point out, touch is a bit more intuitive, but again my question is around whether touch is the investment the industry needs to make or whether it should be in something else, like more contextual trackpads for example as you point out.

      Overall, I’m not convinced yet touch is the right new natural UI solution for desktop and notebook modes.

      1. My preference would be “contextual” trackpads for PCs, as you have coined them. Gives you the speed of touch for navigation with the energy efficiency and greater fine precision of a pointing device. I like touch but it is tiring.

    2. No, sorry. That’s not how you do it with a trackpad. To get to the top right hand corner, I just flick my finger across the trackpad. Then to get it to the bottom left corner, I just flick it back again. I can do that without looking, and much faster than lifting my arm up to touch a screen. At least that’s what you do on a Mac. As Ben said, perhaps Microsoft made the trackpad experience terrible on Windows.
      The last thing you would want to do with a trackpad is to make it correspond with the screen. You’d have to make it the same size as the screen in the end. The only thing I have thought touch on a larger screen might to be useful for is to jump around cells in a spreadsheet.

      1. I just tried to perform that same thing with my laptop running Ubuntu. It took several swipes in either direction and that is without trying to actually hit a target at either end of the screen. My example was meant to simulate attempting to hit a target at opposing ends of the screen. Any movement is quicker if precision is not necessary.

        As for the trackpad having to be the size of the screen, that isn’t accurate. It only needs to be the same proportional size. In other words, a 16:9 screen should have a proportional sized 16:9 trackpad. Granted, the larger the trackpad, the more precisely the pointer can be placed on an object relative to the screen.

        1. On my MacBook Air, I can send the cursor from corner to corner with a flick, but not with enough precision to hit a target (unless its a big one.) But on a proper touch screen, that is, not Windows 8 desktop mode, you don;t want a cursor because you know where you are pointing.You need an insertion point in a text app, but you don’t need or want a little arrow following your finger around.

          1. I only used the example of using the touchscreen to position the pointer as an example of the advantage in speed and precision of the touchscreen. The pointer appears at whatever spot I touch, it doesn’t follow my finger around and I’m not advocating for that.

            I am advocating for the pointer to appear on the screen relative to the position I place my finger on a trackpad. If I place my finger on the upper left corner of the trackpad, the pointer should immediately appear in that area. The pointer will immediately be closer to whatever target I have identified in that region, which will significantly improve speed.

          2. Regardless of speed/precision, you still have the ergonomic problems. It is OK, if you are doing an occasional poke at the screen, but if you doing any significant cursor movement, then “Gorilla Arm” will soon be a factor, even on a laptop.

            Having only tried Windows trackpads, my reaction has always been to demand a mouse when given a laptop to use.

            I think touchscreen only makes on a laptop if it is a convertible to some kind of tablet mode, or the touchscreen is essentially a freebie for the marginal real usage.

          3. I agree that touch on PCs presents ergonomic issues. I like using touch on my laptop but I have to switch too much to the trackpad, the experience isn’t consistent. It actually causes me a great deal of mental fatigue. It’s so inconsistent, I haven’t gotten a chance to test the “Gorilla Arm” theory yet.

          4. I still contend that there are enough examples of doing things at arms length outside computing that “Gorilla arms” is a bit over played.

            What I am wondering now, however, is to what extent is the one-computer-to-do-all computing paradigm, itself, beginning to recede. I mentioned earlier about appropriateness of task and mentioned drafting in particular. When one considers tasks like audio recording, video editing, drawing and painting, and the like, a keyboard and mouse is really a poor interface. Whether the solution is a touch screen or not is still TBD, but the idea of a computer effectively being a dolled up typewriter has got to go, and mobile computing is showing and making that a reality.

            In live performance computer lighting control, the GrandMA is the hot new console and utilizes a touch screen as does the ETC ION lighting console. The best programmers can absolutely fly with (and prefer) touch screens and no evidence of gorilla arms in sight even when sitting at the console for hours on end and days at a time teching a new show.

            Joe

          5. “What I am wondering now, however, is to what extent is the
            one-computer-to-do-all computing paradigm, itself, beginning to recede.” – jfutral

            I’m wondering this myself. General purpose computing wasn’t so much of an issue in the early days of PCs because you couldn’t do as much with them. Now, their capabilities are so vast, I’m beginning to wonder if we should be designing computing devices that are more tailored to specific tasks or categories. I’ve always felt that the visceral experiences of computing would be improved by doing this. I conceptualized a tablet in 2010 that was, for all intents and purposes, an “infinite notebook” which allowed you to create using various writing and drawing implements and various paper styles. Not too long after that, an app called Paper hit the market for the iPad that utilized the same basic philosophy. It’s a great app but I think the visceral experience would have been greatly improved if I had a tablet specifically designed for creating.

          6. Why do you have to switch. Just pretend you have no trackpad and you will quickly get a sense of “Gorilla Arm” issues.

            People get RSI problems just from using a mouse and there you have your arm resting on your desk. Do the full suite of cursor movements with your arm held in the air and you will quickly get “Gorilla Arm” IMO.

          7. In the end, I really don’t dispute that there is a “Gorilla Arm” problem with touch on PCs. What I CAN’T do with current systems like Windows 8 is determine if the benefits in speed make “Gorilla Arm” worth it. Like you stated, it’s possible to get RSI problems just using a mouse. Would a well designed touch system be so much better that I’m willing to put up with other trade offs? I don’t know and can’t know with the current touchbased systems. The reality is that even the iPad’s touch targets were designed with the assumption that it would be used at much closer distances than a laptop or PC. Touch targets on a laptop have to be MUCH larger at the distances at which they are used, even larger on a PC. I think that is why Microsoft went with the Live Tiles concept. One problem that I found was that the navigational elements WITHIN Microsoft’s apps were too small.

            There may very well be benefits to using touch on PCs but it would require serious work to create a system that properly exploits it. I might not have any problems using touch and putting up with the trade offs if it really made my computing experience flow more fluidly.

      2. Microsoft didn’t make the trackpad experience miserable, the hardware OEMs did, though some of the new Windows 8 systems are finally getting it right.

        But you hit the nail on the head for why James King’s one-to-one mapping of the trackpad to the screen won’t work. For amusement, see Jorge Luis Borges’ very brief story, “On Exactitude in Science.”

        1. I don’t see how he has substantiated why such a system won’t work. We are all allowed are biases, but, so far, if you are attempting to prove that what I am suggesting doesn’t work, you haven’t. I think I’ve made a clear case why such a system would be much faster. If you can provide technical detail for why it is not TECHNOLOGICALLY feasible, please do so by all means.

          1. Your idea wouldn’t work because it would create a usability nightmare. The trackpad has to act like a mouse (move your finger on it a bit, it moves the pointer a bit) so that you can have fine control. But if you also try to have it be an analogue to a touchscreen, where touching it here moves the pointer to the same spot on the screen, then there’s no good way for you to make the driver software “know” whether you’re touching the trackpad because you’re about to swipe your finger to move the pointer mouse style, or whether you’re touching it because you want to move the pointer to that spot.

            The non-technical people I know are all constantly accidentally clicking with their trackpads (to their great frustration and annoyance) because the “tap to click” functionality can’t tell the difference between their touching the pad and tapping it. I don’t want to think about how unusable they would find the trackpad if sometimes the pointer would move to a new spot on the screen when they touched it to start a swipe.

          2. “Your idea wouldn’t work because it would create a usability nightmare.” – Glaurung-Quena

            Gee, I guess that settles it then. /s

            “The trackpad has to act like a mouse (move your finger on it a bit, it
            moves the pointer a bit) so that you can have fine control.” – Glaurung-Quena

            What I’m proposing wouldn’t alter this in the least. Once the finger makes contact with the trackpad, as long as the contact is maintained, the pointer would react in the same fashion as it already does with current technology. It operates on the same basic principle as multi-touch.

            “But if you also try to have it be an analogue to a touchscreen, where
            touching it here moves the pointer to the same spot on the screen, then
            there’s no good way for you to make the driver software “know” whether
            you’re touching the trackpad because you’re about to swipe your finger
            to move the pointer mouse style, or whether you’re touching it because
            you want to move the pointer to that spot.” – Glaurung-Quena

            Incorrect. It could be done using simple contact. The relevant event would be REMOVING your finger from the trackpad. As long as the finger remains in contact with the trackpad surface, it would work exactly as current trackpads do. For all intents and purposes, the technology is already in place.

            “The non-technical people I know are all constantly accidentally clicking
            with their trackpads (to their great frustration and annoyance) because
            the “tap to click” functionality can’t tell the difference between
            their touching the pad and tapping it.” – Glaurung-Quena

            This is easily remedied by returning to the use of ACTUAL mouse buttons. Tactile controls are always superior to virtual. That likely one of the reasons the new Razer Blade gaming PC has a trackpad with real buttons.

            “I don’t want to think about how unusable they would find the trackpad
            if sometimes the pointer would move to a new spot on the screen when
            they touched it to start a swipe.” – Glaurung-Quena

            I wouldn’t be any different than using a current multi-touch trackpad. The only difference would be that the pointer would appear in the relative space on the screen that your finger is placed on the trackpad as opposed to maintaining its last position.

            Exchanges like this remind me of why Apple refuses to use consumer input to create its products. I seem to be getting quite a few ego-driven comments. What I’m proposing would be a relatively simple change that should greatly improve the speed of trackpad use. If someone from Synaptics or Apple provided technical reasons for why it is impractical, I could understand the resistance. But the comments I’m getting on this seem to be contrarian just for their own sake.

          3. So every time I unintentionally lift my finger and set it down again, the pointer would jump to a new spot on the screen. Awesome. Your proposal sounds wonderful in an ideal world where everyone always uses their touchpad in exactly the way the engineers who designed it think it ought to be used. Sadly for your idea, that world does not exist.

          4. If you unintentionally lift your finger when you are in the process of using your trackpad, the cursor is not going to end up on your intended target regardless:

            If you are moving the pointer to your target and you unintentionally lift your finger, your cursor will not move to your intended target even using current technology;

            If you are using a “contextual” trackpad, you actually AVOID a user error because the pointer deactivates. The worst that happens is that you have to redo whatever action you have initiated with the pointer. That is pretty much the EXACT SAME EFFECT of using current technology.

            Look, you don’t have to like it or want to use it. But the reality is that, unless there are technical reasons of which I am unaware, the idea is feasible and will likely perform as intended.

          5. You are getting pushback on your idea simply because nobody believes it would ‘greatly improve the speed of trackpad use’ as you think.

            Right now, if you want coarse movements, you move your finger a small distance but quickly, i.e. a flick, and then for fine corrections for a precise motion, you move your finger a similar small distance, but slowly. So your finger on the trackpad never has to move much, and the motion is restricted to one finger or possibly a hand, moving briefly from the keyboard and back, rather than lifting it up to a touch screen, or moving the whole arm a large distance across a screen-scale touchpad. A mouse works similarly. The degree of speedup obtained by moving your fingers a distance quickly vs. the same distance slowly is the response curve, and at least in the old days was a linear function in Windows, and a curve on the Mac, making the Mac mouse much more sensitive. This response curve may be why your Ubuntu trackpad requires several finger swipes to move your cursor from one corner to another, while on my Mac it is one quick flick.

            Secondly, in your idea, you have to take your eyes off the screen to see that you have put your finger in the right place, whereas on a normal trackpad, you keep your eyes on the screen, and your finger on the trackpad. Again, I suspect your idea would be slower, as it is much faster to touch type than to look at the keyboard.

            What you are talking about exists; it sounds like a Wacom tablet for artists to ‘draw’ on a screen, where the direct spatial mapping from touchpad to screen is useful. These can be used as general -purpose trackpads, I guess, but I don’t think they have ever caught on as such.

          6. “You are getting pushback on your idea simply because nobody believes it
            would ‘greatly improve the speed of trackpad use’ as you think” – James Owens

            The problem with a “belief” is that it is not based on empirical evidence. So far no one has shown any technical or ergonomic reason why a “contextual” trackpad would not work or not do as I have outlined. I don’t mind my ideas being challenged, I just ask that criticisms be based on quantifiable factors and not knee-jerk reactions.

            “Right now, if you want coarse movements, you move your finger a small
            distance but quickly, i.e. a flick, and then for fine corrections for a
            precise motion, you move your finger a similar small distance, but
            slowly. So your finger on the trackpad never has to move much, and the
            motion is restricted to one finger or possibly a hand, moving briefly
            from the keyboard and back, rather than lifting it up to a touch screen,
            or moving the whole arm a large distance across a screen-scale
            touchpad.” – James Owens

            My example already shows the advantages in speed and precision offered by a touchscreen over a trackpad. As I pointed out, your example does not disprove the example because you are not aiming for a specific target. Smartphones used to have optical trackpads (Windows Mobile, Blackberry, etc) but now are pretty much exclusively based on touch. Optical trackpads for smartphones are more energy efficient and arguably more ergonomic. But touch is faster and more precise. The issue is not that touch isn’t faster or more precise (provided the target is large enough) over a trackpad or mouse; it simply is. The issues relate to other ergonomic problems that are created when touch is applied to laptop or desktop PCs, the bulk of which deal with greater personal energy usage.

            “Secondly, in your idea, you have to take your eyes off the screen to see
            that you have put your finger in the right place, whereas on a normal
            trackpad, you keep your eyes on the screen, and your finger on the
            trackpad. Again, I suspect your idea would be slower, as it is much
            faster to touch type than to look at the keyboard.” – James Owens

            This is at least a legitimate point. But this poses a far less significant challenge than typing on a touchscreen, an action comfortably performed by millions of people at this point. Even now, I can place a finger in the proper proximity on a trackpad relative to my screen without looking. Some tactile elements, like small nubs or indentations placed in a grid configuration on the trackpad, would improve that ability.

            “What you are talking about exists; it sounds like a Wacom tablet for
            artists to ‘draw’ on a screen, where the direct spatial mapping from
            touchpad to screen is useful. These can be used as general -purpose
            trackpads, I guess, but I don’t think they have ever caught on as such.” – James Owens

            EXACTLY. What I’m proposing is not a major departure and Wacom tablets are an excellent example. There shouldn’t be that much of a difference from using a stylus to using a finger.

          7. You demand evidence, and complain about ‘beliefs’, but then present your own beliefs without any evidence. Oh, and you can’t spell my name right.

            You said that your example had ‘already shown the advantages of a touchscreen over a trackpad’ but I have presented evidence that moving rapidly and precisely around a screen is trivial with a well-designed trackpad, negating your example.

            You said that on your Ubuntu laptop, moving the cursor precisely from one end of the screen to a target at the other is slow, and that touch would be quicker. However, as a counterpoint, I described how I would move a trackpad cursor from one end of a screen to another extremely fast, with a quick flick on my finger on the trackpad, achieving the same thing. So, either I move my hand a large distance from one end of a touchscreen to another as you suggest, or I make a quick flick of one finger, followed by a more careful finger movement on my trackpad, without lifting my hand from the desk. This is not only less energy, but my hand doesn’t have to move so far, because the movement is magnified by the trackpad, so inevitably this will be quicker than your suggestion, not slower.

            You haven’t shown that your suggestion would be faster or more precise as you assert, just that your trackpad doesn’t work as well as it could do. I’m sure the hardware is similar, so perhaps Apple has done something clever with the software, probably in the response curve, which runs their trackpads.

            Furthermore, when people reviewed keyboards for the iPad, many noted that the lack of a trackpad or mouse, which wouldn’t work with the iPad, was a handicap that they had to get used to, as lifting a hand to the iPad screen to move the cursor was more cumbersome.

          8. “You demand evidence, and complain about ‘beliefs’, but then present your own beliefs without any evidence.” – James Owen

            I presented an example that anyone with a laptop with a touchscreen can duplicate. It doesn’t get more empirical than that.

            “You said that your example had ‘already shown the advantages of a
            touchscreen over a trackpad’ but I have presented evidence that moving
            rapidly and precisely around a screen is trivial with a well-designed
            trackpad, negating your example.” – James Owen

            No, it doesn’t negate my example. It doesn’t magically make touch less quick or precise than a trackpad. I already exposed the flaw in your counter-example by pointing out that speed is easy to achieve without precision. I can PRECISELY place the pointer ANYWHERE on my screen by touching it.

            For that matter, you can’t even duplicate my example using a Mac but I actually have a trackpad on my laptop. You are arguing from a position of ignorance.

            “You said that on your Ubuntu laptop, moving the cursor precisely from
            one end of the screen to a target at the other is slow, and that touch
            would be quicker. However, as a counterpoint, I described how I would
            move a trackpad cursor from one end of a screen to another extremely
            fast, with a quick flick on my finger on the trackpad, achieving the
            same thing. So, either I move my hand a large distance from one end of a
            touchscreen to another as you suggest, or I make a quick flick of one
            finger, followed by a more careful finger movement on my trackpad,
            without lifting my hand from the desk. This is not only less energy, but
            my hand doesn’t have to move so far, because the movement is magnified
            by the trackpad, so inevitably this will be quicker than your
            suggestion, not slower.” – James Owen

            This is completely incorrect. The combination of making two movements, the second being a slower gesture for precision, means that the entire process will be slower than simply touching a point on a screen or on a trackpad. It doesn’t matter that you are using less energy, what matters is that it is less precise so will require more time to execute.

            I used the example of trackpads on smartphones to illustrate this point.

            “You haven’t shown that your suggestion would be faster or more precise
            as you assert, just that your trackpad doesn’t work as well as it could
            do. I’m sure the hardware is similar, so perhaps Apple has done
            something clever with the software, probably in the response curve,
            which runs their trackpads.”

            I’ve proven the advantage of touchscreens over trackpads for speed and precision. Applying the same principles for a “contextual” trackpad should yield similar results.

            You can’t credibly argue this point because you can not duplicate the example I’ve provided to compare it to your trackpad.

            “Furthermore, when people reviewed keyboards for the iPad, many noted
            that the lack of a trackpad or mouse, which wouldn’t work with the iPad,
            was a handicap that they had to get used to, as lifting a hand to the
            iPad screen to move the cursor was more cumbersome.” – James Owen

            This only proves that touch is more energy intensive under those circumstances. That’s a point I have aready made.

            My apologies for misspelling your name.

          9. “I already exposed the flaw in your counter-example by pointing out that speed is easy to achieve without precision. I can PRECISELY place the pointer ANYWHERE on my screen by touching it more quickly than I can place it with a trackpad.”

            And I pointed out that I have speed and precision with my trackpad, so there is no flaw in my example. I cannot speak for your experience with trackpads, except to say that your description of your trackpad on your Ubuntu laptop sounds clunky and slow, compared to my experience of my trackpad on my MBP.

            “The combination of making two movements, the second being a slower gesture for precision, means that the entire process will be slower than simply touching a point on a screen or on a trackpad.”

            Nonsense. The combination of flying from Dallas to Boston, and then taking the train to Back Bay is still going to be quicker than driving from Dallas to Boston. It may be quicker to touch, but it may not be. It won’t automatically be slower to do something in two steps rather than one.
            In fact, I think moving my finger a inch here or there on a trackpad is always going to be much quicker than moving my whole hand or arm up to and around an 7-10″ touchscreen. It’s not that it’s more energy-intensive, which it is, but it’s also going to be slower to move my whole arm.

            “For that matter, you can’t even duplicate my example using a Mac but I actually have a trackpad on my laptop. You are arguing from a position of ignorance.”

            [I think you mean you have a touchscreen on your laptop. Naturally there is trackpad on my MBP.] Of course I can. I can touch any point on my screen on my Mac with my finger, thus duplicating the experience of a touchscreen, it just won’t activate anything, as it is not a touchscreen.

            “You can’t credibly argue this point because you can not duplicate the example I’ve provided to compare it to your trackpad.”

            As I said above, I can duplicate it very easily. I’ve done so.

            I have an iPad of my own, but I don’t need it here.

            Launch app on my Mac: Wham cursor down to Dock, can’t overshoot as it’s a screen edge, slide along to app if necessary usually no more than 1 app, click. Don’t need to be ultraprecise, as Dock magnifies just a little bit, to increase the target size. (Or open Launchpad, but I never use that.)

            Pretend to launch app on my Mac: Reach out to screen with finger, press icon in Dock. No magnification in this case, as my screen doesn’t know there’s a finger coming, so target is smaller and therefore slower. Need a touchscreen with proximity detection (which exist) to do the same target magnification as for the Dock, to make this a viable alternative. Can’t make icon targets any bigger, as I have lots of them in my Dock for everyday use.

            Can’t say there is any speed advantage in either of these. I notice that my finger slows down as it approaches the screen to ensure that I hit the right icon because I can’t touch the screen without activating the wrong icon, whereas if I miss with the cursor, I can scrub along the Dock before clicking. The further out I have to reach, the worse this will get, to hold the long lever of my arm and make precise movements. My laptop screen with a small distance is not so bad, but on my big desktop screen, it’s terrible.

            “I’ve proven the advantage of touchscreens over trackpads for speed and precision.”
            No, you’ve asserted quite a lot, but you have proven nothing at all. You keep making definitive statements, which crumble under the slightest scrutiny.
            I’ve presented plenty of empirical evidence that there is at best, no speed increase from having a touchscreen on my laptop or desktop compared to the trackpad on my machine.

            Try something for me: go to the nearest Apple Store and try the trackpads on the nearest MBP. Try flicking your finger around the trackpad, try out some of the 2,3,4-finger functions which are available. When you’ve done that, come back and let us know how it compares to the one on your laptop.

          10. “And I pointed out that I have speed and precision with my trackpad, so there is no flaw in my example.” – James Owen

            The problem is that you can say anything you want. Anyone who duplicates the example I have provided will clearly see that a trackpad is no match for a touchscreen provided the target is large enough.

            “The combination of flying from Dallas to Boston, and then taking the
            train to Back Bay is still going to be quicker than driving from Dallas
            to Boston. It may be quicker to touch, but it may not be. It won’t
            automatically be slower to do something in two steps rather than one.
            In
            fact, I think moving my finger a inch here or there on a trackpad is
            always going to be much quicker than moving my whole hand or arm up to
            and around an 7-10″ touchscreen. It’s not that it’s more
            energy-intensive, which it is, but it’s also going to be slower to move
            my whole arm.” – James Owen

            I’ve given you a way to objectively test your theory using your MB and an iPad. If you think your trackpad is as quick or precise as a touchscreen, simply put it to the test.

            “Of course I can. I can touch any point on my screen on my Mac with my
            finger, thus duplicating the experience of a touchscreen, it just won’t
            activate anything, as it is not a touchscreen.” – James Owen

            Really? So you can duplicate something by not actually duplicating it? Okay.

            “Launch app on my Mac: Wham cursor down to Dock, can’t overshoot as
            it’s a screen edge, slide along to app if necessary usually no more than
            1 app, click. Don’t need to be ultraprecise, as Dock magnifies just a
            little bit, to increase the target size. (Or open Launchpad, but I
            never use that.)

            Pretend to launch app on my Mac: Reach out to screen with finger,
            press icon in Dock. No magnification in this case, as my screen doesn’t
            know there’s a finger coming, so target is smaller and therefore slower.
            Need a touchscreen with proximity detection (which exist) to do the
            same target magnification as for the Dock, to make this a viable
            alternative. Can’t make icon targets any bigger, as I have lots of them
            in my Dock for everyday use.” – James Owen

            So, rather than performing the very simple real world test that I presented, you are running a similation in your mind? Okay.

            “No, you’ve asserted quite a lot, but you have proven nothing at all. You
            keep making definitive statements, which crumble under the slightest
            scrutiny.
            I’ve presented plenty of empirical evidence that there is
            at best, no speed increase from having a touchscreen on my laptop or
            desktop compared to the trackpad on my machine.” – James Owen

            And what empirical evidence would that be? The similations that you’ve run in your head? I’ve provide a real-world example that anyone can dispute provided they have a different ACTUAL, NOT VIRTUAL, experience. No one has come forward yet to dispute it.

            I’ve also given you a REAL-WORLD example that you can perform with your MB and your iPad which can simulate pretty well the ergonomics and usage distances of a touch screen laptop vs trackpad use. Hell, put your iPad on a stand or attach it to a BT keyboard attachment. You won’t do it because you already know what the results will be.

            You’ve provide no REAL-WORLD test to prove your point. The science of this has already been confirmed. But I guess you are allowed to BELIEVE what you want to believe.

            “Try something for me: go to the nearest Apple Store and try the
            trackpads on the nearest MBP. Try flicking your finger around the
            trackpad, try out some of the 2,3,4-finger functions which are
            available. When you’ve done that, come back and let us know how it
            compares to the one on your laptop.” – James Owen

            Why do you think I already know your assessment is wrong?

          11. “Anyone who duplicates the example I have provided will clearly see that a trackpad is no match for a touchscreen provided the target is large enough.” I thought I warned you against making definitive statements that crumble at the slightest scrutiny. So I have duplicated the example you gave, and it didn’t give the outcome you think it should, so with a single counterexample (this is the scientific way), the whole theory collapses.

            And no, I haven’t done them in my head, I did them with my fingers on my computer screen. It doesn’t need to be an actual touch screen to be able to touch it, silly. What it needs to be is a physical surface to touch, that’s all. I didn’t realise there was some magic about a touch screen that made it so much faster to touch than some other glass surface.
            I have a laptop with its smaller screen, and with a secondary larger monitor, so I tried different variations of test. The bigger and further away the screen gets, the worse touch becomes. So a phone or an iPad, which are held close, are much quicker with touch, but an ordinary computer, not so much.
            The tests I did were indeed real-world, and yet sadly, because it does not chime with your expectation, you tell me it’s virtual, or that I won’t do it, when I have indeed done it.

            You keep telling me to do this simple, real-world, empirical test, yet when I do that, you deny the result that I get, probably because you are incapable of accepting that the result might not be what you assert it should be. Not very scientific. Clearly, you in fact are the one believing what you want to believe, ignoring or denying in this futile fashion all the evidence that I have presented to you.

            And just to put the last nail in the coffin of your credibility, there you refusing to try my empirical experiment, which is to go check out the trackpads in the Apple Store. Hoist with your own petard. Good night.

          12. “And no, I haven’t done them in my head, I did them with my fingers on my
            computer screen. It doesn’t need to be an actual touch screen to be
            able to touch it, silly.” – James Owen

            So you didn’t actually use a real touch screen? You used a device that could give no results? That’s great. Real credible.

            “So a phone or an iPad, which are held close, are much quicker with touch, but an ordinary computer, not so much.” – James Owen

            You had the opportunity to use a stand or BT keyboard with your iPad. It was a simple challenge: You with your MB vs. someone else with an iPad. You open an application using your trackpad, the other person opens an app with touch. With a stand or a BT keyboard, the ergonomics of the iPad would have been similar to a touchscreen PC.

            It’s so simple but you won’t do it. I wonder why 🙂

            “And just to put the last nail in the coffin of your credibility, there
            you refusing to try my empirical experiment, which is to go check out
            the trackpads in the Apple Store. Hoist with your own petard. Good
            night.” – James Owen

            Yeah, I’m totally broken up about it.

      3. Back in the old days, when having two fingers on a trackpad was a mistake and confused the bejeezus out of the laptop, I used to be able to jump the curser from one end to the other by using two fingers spread across the track pad and lifting one finger or the other depending on which way I wanted the curser to go. Can’t do that anymore, though.

        Joe

  2. Drill down a little deeper and there are other sub-questions. Is the way we sit at a desk the best way to use a computer? Back in the old days, when we drafted by hand at a drafting table, a big thing was the chair where you essentially knelt which shifted your body weight forward as well as making it easier to reach the top and bottom of the surface. Some people still prefer those chairs for desks.

    Also, these days, the standing desk is getting a lot of attention.

    I still liken using a touch surface that is not a tablet more to using an easel and painting.

    The point is, we have lots of ways already that we interact with something both visually and physically without the typical office desk set-up. So a lot of things have to be rethought or revisited, not the least of which is the particular task at hand. But what this requires is what most of us are uncomfortable with because we have developed a comfortable way of interacting with a computer.

    I mentioned metaphors in another thread (regarding Windows 8 and Windows RT) and I don’t think Steve necessarily understood what I was talking about (no surprise there, I can be rather abstract or inscrutable). What I mean is, with new ways of interacting with something, sometimes the old metaphors get in the way and especially the underlying assumptions have to be re-examined. Especially for those of us who have gotten it ingrained in us to do something a particular way. Sometimes, you have to go back a little further and ask not just how we do things, but ask why we do them that way.

    And these days, considering how much time we spend at the computer vs. how much time ergonomics says we should step away from the computer, maybe not having the computer at our desk is not such a bad thing.

    Joe

    1. I actually have no issue with the posture. My point was comparing it to how we have used computing it could get laborious keeping arms up and leaning forward.

      Touch has to prove its usefulness in a sit down and operate a desktop or notebook form factor for long periods of time. I’m not saying this isn’t possible or won’t happen just that it is not today, nor does it seem to be an area of focus to problem solve by MS or even other OEMs. I have brought this up with decision makers at the OEMs so we will see what happens.

      1. “My point was comparing it to how we have used computing it could get laborious keeping arms up and leaning forward.”

        Right. Which is why I pointed out this is a problem that was addressed when we had to draft by hand with the chair I referenced. (like these: http://www.bizchair.com/kneelingchairs1.html) or the standing desks some people are all hot on. A touch computer (I would think primarily laptop) in those environments is far more appropriate if not at least more usable.

        I could be wrong, I kind of don’t think the OEMs will give it much thought and will expect the user to figure it out. Kind of like how hammers don’t come with instructions. Not that they shouldn’t, the manufacturers just don’t understand why they should.

        My bigger point is that maybe it isn’t the hardware that needs to be rethought, but what is the most appropriate environment to use whichever UI makes the most sense for a given task.

  3. The only computer my mother has ever owned is an iPad. Other day I was visiting and handed her my MacBook to show her some photos. She kept touching the screen. She found the whole notion of a mouse/trackpad ridiculous.

    1. Keyboard shortcuts are similarly less intuitive than using a mouse to perform many tasks, but they are ultimately far more efficient. The two are independent, something can be more intuitive, but vastly less efficient (though this isn’t always the case).

      The iPad or Surface are far more intuitive than a laptop and for some tasks they are equally effective or even more effective, but for many tasks, the laptop is vastly better. If you are working on a spreadsheet, programming, or editing a long document, a tablet is vastly less effective. The touch screen is less accurate than a mouse, keyboard, or trackpad. Simple tasks like selecting text or precisely placing the cursor become more cumbersome and require a bigger effort to manipulate.

      If you are a professional who needs to do these jobs, it pays to have the right tools and the cost of learning the trackpad or mouse is a small investment versus the better efficiency.

  4. When using a desktop or notebook computer, I absolutely have to have a mouse. I can’t use a trackball or that magicpad thing from Apple. Even with a touch screen on my netbook, it is so much easier to use the mouse. For one thing, I can get more precise than with my chubby finger or the bulbous stylus.

    Eventually, when using my tablet, I will wean myself from the keyboard. It’s a long and arduous journey, but I’m getting there.

    Pointing devices are absurd on tablets, just as touch screens are nearly useless on standard computers. Eventually, I expect tablets and smartphones will replace desktops and such as standards. After that, possibly some kind of projection and simulation combination of software and hardware will most likely take over. I expect it will eventually get down to gestures with hands and eyes as well as the spoken word.

  5. Ben, I think you are missing the bigger picture. Not only is touch just wrong on desktop and laptop computers but the reasons why we are seeing laptops with touch built in is not only because most if not all Windows OEMs took the cheep way out and included horrible trackpads but because Microsoft saw the writing on the wall and we have pivoted from the mouse and keyboard input to touch. They are once again trying to be considered modern by forcing touch UI’s on users in a shell layer the same way they did with Windows to DOS back in the 1990’s to make Windows look like it was the same as Mac OS. They are doing this in the hope that they can buy the time they need to pivot Windows once again in to a more modern OS. The last time it took Microsoft until 2001 (Windows XP), 6 years after Windows 95’s release to get Windows fully on the GUI.

    In Microsoft’s eyes “Windows” is just a name they can slap on any technology and then sell it to their OEM partners who then go off and sell it to the end users. Witness Windows RT. It really is NOT Windows at all. But they sold it as such.They figure that most users do not know enough about the guts of how Windows works to really understand what really is going on under the hood. For the most part I will agree with them. But the difference this time around unlike the 1990’s is that we have real choice. Between what Apple is doing on both the desktop (Mac OS X) and mobile (iOS) and the various flavors of Android and other mobile OS’s too. Also, I do think that Google will continue to push Chrome OS too.

    The thing is that I really do not think that they have 6 years to redo the guts of Windows to make it and all of their 1st party applications touch native. Microsoft just can not develop that fast. Not only that but developers are spending way more time writing mobile and web apps than desktop apps. They see where the money is and are going for it.

  6. To me this issue is really what you are doin with your computer. If you are a light user it doesn’t really matter much. Heck you might actually not even need a computer but could be fine with a computing device like a tablet.

    But for someone like me who is doing advancing editing work on photos and videos, a touchscreen computer could prove a disaster. IF it is the traditional form. I work on a Wacom Cintiq display connected to a Mac Mini all the time and it is better than using a mouse etc cause I can write directly on the photos I’m working on. But when I’m back to email etc I still tend to use keyboard and trackpad. If only because I need to change my posture to mix things up and avoid strain.

  7. All anyone needs to do is sit in font of one’s PC or laptop and mimic conducting touch gestures on their screen interspersed with keyboard input and after five minutes it will become quite obvious from the pain in your shoulder, elbow, or wrist that a screen touch-keyboard hybrid is a non-starter. MS could have paid me a token consulting fee and I would hav told them that. But noooooo, they preferred to spend a billion in development and writeoffs.

    Touch-keyboard hybrids can be ade to work but it’s trackpad touch not screen touch. Steve Jobs knew this and he actually said it in public that the way to implement touch on a desktop machine is with a trackpad. But why would anyone listen to Steve Jobs, eh?

    I’m still waiting for a trackpad that maps to a screen pixel for pixel with animated graphics on the screen to show the user exactly where his fingers are positioned.

Leave a Reply to steve_wildstrom Cancel reply

Your email address will not be published. Required fields are marked *