Because that was the whole subject of the article -- replacing buttons that move with haptic-feedback buttons. And your comment was a response to the article.
And despite your lengthy reply, you still haven't suggested why replacing a physical depressed button with a haptic one has any ergonomic implications, given that the user interaction appears identical.
> I never make comments on HN that are deliberately intended to offend... given your nonsensical comment.
Well calling my comment "nonsensical" continues to offend, and it's awfully hard to see it as anything but deliberate. Again, I suggest you look at HN guidelines which you are continuing to break.
2. Both a mechanical push-button switch and an Apple's Taptic switch are BOTH haptic devices/interfaces by definition as both involve the sense of touch albeit each having a different modus operandi. Here's part of the Merriam-Webster dictionary definition of 'haptic':
"haptic, adjective
1: relating to or based on the sense of touch
2: characterized by a predilection for the sense of touch
https://www.merriam-webster.com/dictionary/haptic
3. Again, the thrust of Allison Johnson's article is that the more familiar mechanical switch interface works better than an Apple's Taptic capacitive-type switch. There is no doubt about this as switches without tactile resistance cannot provide dynamic feedback to the human who pushes it as they have no inherent mechanism to do so; the consequence is that they've a dead feel and are dynamically unresponsive. It's why so many users hate smartphone keyboards, they only use them out of sufferance because there's nothing better. (If this haptic interface worked well then there'd be no complaint, however, even with nearly 20 years of smartphone engineering and development, the problem still remains.)
Nothing illustrates this better than competitive gamers who demand mechanical keyboards. For them, everything about a keyboard is important, its size, rigidity, stability, visual feedback (whether the keys are backlit etc.) and especially the feel of its key switches—all of which contribute to the main haptic feedback (interaction) that users experiences. Diehard gamers demand 'Rolls Royce-type' key switches of the highest quality, these are usually made by the industry leader Cherry. Even then that's not enough, they'll select a specific type of key that has acceleration dynamics and actuation pressure to their liking and particular working style. It's why Cherry and other keyboard manufacturers grade their keys by color, red, blue, black, green etc.: https://keebnews.com/mechanical-keyboard-color-switches-guid....
The same goes for the original IBM PC keyboard, anyone who has ever owned one would never let it go (here read, 'It Sounds and Feels Satisfying'): https://www.howtogeek.com/661430/why-i-still-use-a-34-year-o....
4. A well-made mechanical button/switch is very predictable in its operation. It provides precise and consistent tactile feedback to the user by increasing mechanical resistance to human pressure, simultaneously its moving in sympathetic analog along with the user's finger and does so consistently up to the trigger/switching point whence the pressure is released. Its physical properties including size, shape, resistive pressure, positional change during operation as well as the texture and feel of its surface all integrate into a specific dynamic feedback pattern that defines overall haptic characteristic for any given device (it's dynamic feedback because it's constantly changing right up to the release point). As human familiarity and confidence grows, the button/switch essentially becomes an effective extension of the user's body (here the user's fingers).
5. All these switch parameters are very important because tips of human fingers are exquisitely sensitive to touch and feel. A human notices even the slightest change in contact, acceleration, pressure and other properties of touch (most people have little appreciation of how truly sensitive the sense of touch in their fingers is).
6. 'Taptic'-type buttons/switches are made to mimic mechanical ones, they can be made to respond to the speed of a user's actions by measuring positional information and a change in capacitance. As a finger approaches or withdraws from the switch surface the time rate change of capacitance is inputted into a haptic emulator. Similarly, pressure difference is measured as the area of the finger on the surface changes, the harder a user pushes the switch surface the more skin comes in contact with its surface hence the capacitance increases. Here, the change in capacitance only has limited capacity to convey information. Note: this is a vast subject, and I've barely scratched the surface.
We know Taptic-style interfaces work as they are the basis of modern screen operation in smartphones. However, they are far from perfect. I'll illustrate what I mean by example. Above, I moved the order of two sentences about, and as this is comment is being prepared on my smartphone, I selected the text to be moved by highlighting it with my finger. The trouble was (as is so often the case with many others and me) selecting text this way was difficult, in fact it's often a damn nuisance. Either one's pressure is too high or too low, or the conductivity of one's fingers changes with circumstance, and or the granularity of selection is wrong: finger too big, text too small, etc. Irrespective, selecting text on a smartphone is much harder† than with a mouse as one doesn’t have the same fineness of granularity as one does with typical well-honed mechanical devices.
I again return to the exquisite sensitivity of human touch and how important it is whether it's interaction with other humans or inanimate objects such as tools and smartphones. Simply, if dynamic tactile feedback is missing there's a problem, if it's not perfect then there's still a problem. For human contact a crude analogy would be the withdrawal method of contraception—it sort of works but it's not very satisfying. Simply, the human body requires very subtle feedback if it's to work at its best (the type of coupling is critical if things are to be optimal).
Designing a tight effective coupling (interface) between a human and an inanimate object to the extent that it effectively becomes an extension of the human body is a very complex matter. Especially so if it has to be intuitive and its use second nature to a wide body of people with a minimum of learning (a prerequisite requirement of smartphone features).
There are any number of excellent examples from other areas but most require extensive learning if using them is to become second nature. An experienced pianist can make dazzling use of the feedback from his piano keys but it can take years of training to do so—even then that's with the dynamic tactile feedback that piano keys inherently provide. Moreover, much of this learning is not only very involved but by nature it's also very subtle and specific. When for the first time a pianist tries to play an organ, harmonium or an electric keyboard that has little or different dynamic feedback then he/she is often thrown completely, the result is often horrible. Designing Taptic-style interfaces is inherently difficult as without moving parts one starts from an essentially 'static' base; in effect, all haptics have to be emulated (and that's a huge task and to date the results have been far from perfect).
There's no doubt that Apple's engineers will go to inordinate lengths and use combinations of algorithms to optimize the Taptic performance and no doubt it will work but the fact still remains that without tactile feedback that possesses dynamic range (as opposed, say, to a finger hitting a glass or other hard surface) then effectively the tool/human (haptic) interface is only half done. QED!
You wanted a finely-ground answer and now you have one. My argument was and still is that whilst the author's choice of words could have been better—in that her usage of the word 'haptic' is loose to say the least—the thrust of her point/argument is nevertheless crystal clear. I still can't understand why you cannot see this. Essentially, with your understanding of ergonomics (and presumably haptics) you should have skipped the somewhat mangled terminology and cut through to the core of Johnson's argument.
__
† For those who would question otherwise I'd ask them to check the size of the text compared to their finger size. Often this is sub-optimal, especially so if one has large fingers and one chooses small text as I do (for me, the more text per screen the better).
This type of limitation/problem is not a function of the brand of O/S (Apple, Android etc.) but rather it is intrinsic to its modus operandi for reasons stated. That said, the quality of the implementation differs between brands/systems, however the fundamental design limitation remains.
Whilst I'm not doing so here, I often use a Bluetooth mouse and keyboard with my smartphone as this overcomes the limitations of the interface—I even get to choose the type of keys on my mechanical keyboard.
The fact that PC peripherals can be used with a smartphone is clearly a tacit acknowledgment by smartphone designers of the limitations of Taptic-style interfaces.
2. Both a mechanical push-button switch which actuates (that is, parts move to switch on) and Apple's Taptic switch are BOTH haptic devices/interfaces by definition. Here's part of the Merriam-Webster dictionary definition of 'haptic':
"haptic, adjective
1: relating to or based on the sense of touch
2: characterized by a predilection for the sense of touch
https://www.merriam-webster.com/dictionary/haptic
3. Again, the thrust of Allison Johnson's article is that the more familiar mechanical switch interface works better than an Apple's Taptic capacitive-type. There is no doubt about this as switces without tactile resistance do not provide dynamic feedback to the human who pushes it. It's why so many hate smartphone keyboards—they only use them out of sufferance because there's nothing better. They've a dead feel and are dynamically unresponsive.
Nothing illustrates this better than competative gamers who demand mechanical keyboards. For them, everything about a keyboard is important, its size, rigidity, stability, visual feedback (lit keys etc.) and especially the key switches—all of which contribute to the main haptic feedback (interaction) that the user experiences.
Diehard gamers demand 'Rolls Royce-type' key switches of the highest quality, these are usually made by the industry leader Cherry. Even then that's not enough, they'll select a specific type of key that has acceleration dynamics and actuation pressure to their liking and particular working style. It's why Cherry and other keyboard manufacturers grade their keys by color, red, blue, black, green etc.: https://keebnews.com/mechanical-keyboard-color-switches-guid....
Same goes for the original IBM PC keyboard, anyone who has ever owned one would never part with it (here read, 'It Sounds and Feels Satisfying') : https://www.howtogeek.com/661430/why-i-still-use-a-34-year-o....
4. A well made mechanical switch provides consistent tactile feedback, it offers increasing mechanical resistance up to the point where it switches, moreover, it not only switches cleanly and repeatedly but also has a predictability about its operation. As such, the human has confidence in it and thus it becomes an extension of his/her body (fingers). The more predicable the switch's action the more effective it is as an extension of the human anatomy.
5. All these switch parameters are important because the tips of human fingers are exquisitely sensitive to touch and feel, a human notices even the slightest change in acceleration, pressure and means of touch. 'Taptic' type buttons are made to mimic a mechanical one; that is they can be made to respond to the speed of a user's actions (time rate change of capacitance as a finger approaches/withdraws), and the pressure applied (capacitance changes as the area of the finger on the surface changes—it increases when the user pushes harder as more skin touches the surface). Note: this is a vast subject and I've barely scratched the surface.
No doubt these Taptic-style interfaces work as they are the basis of modern screen operation in smartphones. However, they are far from perfect. I'll illustrate what I mean by example. Above, I moved the order of two sentences about, and as this is comment is being prepared on my smartphone, I selected the text to be moved by highlighting it with my finger. The trouble was (as is so often the case with many others and me) selecting the text this way was difficult, in fact it's often a damn nuisance. Either one's pressure is too high or too low, or the conductivity of one's fingers changes with circumstance or the granularity of selection is wrong—finger too big, text too small, etc.). Irrespective, selecting text on a smartphone is much harder† than with, say, a mouse.
The fact is that whilst these Taptic-style interfaces work they do not provide the dynamic tactile feedback that a proper mechanical switch, mouse, joystick or stylus does (the stylus is why I loved my old Samsung note3). Simply, the dynamic tactile feedback is missing. A crude analogy would be the withdrawl method of contraception—works sort of but not very satisfying. The fact is human body requires this subtle feedback if it's to work at its best. Another good example is a pianist and the feedback from piano keys. When a pianist tries to play an organ, harmonium or an on/off electric keyboard without dynamics for the first time then he/she is often thrown completely, the result is often horrible.
There's no doubt that Apple's engineers will go to inordinate lengths and use combinations of algorithms to optimize the Taptic performance and no doubt it will work but the fact still remains that without tactile feedback that possesses dynamic range—as opposed to a finger hitting a glass or other hard surface—then effectively the tool/human (haptic) interface is only half done. QED!
6. You wanted a finely-ground answer and now you have one. My argument was and still is that whilst the author's choice of words could have been better in that her usage of the word 'haptic' is loose to say the least the thrust of her point/argument is nevertheless crystal clear. I still can't understand why you cannot see this. Essentially, with your understanding of ergonomics (and presumably haptics) you should have skipped the somewhat mangled terminology and cut through to the core of Johnson's argument.
__
† For those who would question otherwise I'd ask them to check the size of the text compared to their finger size. Often this is suboptimal, especially so if one has large fingers and one chooses small text as I do (for me, the more text per screen the better).
This type of limitation/problem is not a function of brand or O/S (Apple versus Android) but rather it's intrinsic to its modus operandi for reasons stated above. That said, the quality of the implementation differs between brands/systems but the fundamental design limitation remains.
Whilst I'm not doing so here, I often use a Bluetooth mouse and keyboard with my smartphone as this overcomes these limitations—I even get to choose the type of keys on my mechanical keyboard. The fact that PC peripherals can be used with a smartphone is clearly a tacit acknowledgment by the designers of the limitations I've been discussing.
