Hacker News Reader

Comment by systemvoltage | Hacker News Reader

I don't understand why people find novelty of touchscreens appealing.

We've got to do a better job of educating consumers. The marketing behind touchscreens is so powerful that it is unanimously considered as a feature, something you must have.

I get angry just thinking about the amount of marketing bullshit around touchscreens. Just go to Garmin Aviation website and now a lot of operation in airplane cockpits is offloaded to touchscreens. What happened to the scratch pad and press an index key in FMCs? Age old UI and it is so crisp and clear, there is absolutely zero ambguity. Write something in a scratch pad, click the button where you wanna insert the data.

This is a cost saving measure disguised as a feature. Public is dumb and does'nt understand UI/UX in the slightest bit.

It is just sad. As long as consumers keep buying, nothing will change.

View on HN

Touch controls are only warranted when you need to support a large number of varied interactions which can't be supported by physical controls, or if you have to support arbitrary gesture-based input (like a drawing tablet for example). If that's not the case, physical controls are always better. If you're operating a vehicle, or can't look at the display for any reason, tactile controls are leagues better.

I don’t think it’s quite so black-and-white although I agree roughly with where you draw the line.

My car has tactile controls but they control a cursor on a non-touch screen. In this case touch screen controls would be superior because it would be faster and there would be less time looking at the display.

An argument could be made for more controls to get away from the need for a cursor/menu at all, and while I largely agree, at some point you’re likely to have so many controls you end up looking at the controls anyway, even with tactile hints.

Even if not, in this case the major controls are basically shortcuts and they’re the most commonly used and important shortcuts. Adding more buttons and knobs for things used less frequently or of lesser importance would muddy the waters and make these controls more difficult to use.

To muddy the waters further, some controls can do double duty with things like cycling through functions, double clicks, contextual presses, etc.

General purpose computing requires generic control systems. If your device knows its purpose, it shouldn't need a touchscreen.

Another benefit, for kitchen appliances: a flat screen is easier to clean than physical knobs.

Agreed and you nailed it - an iPhone with touchscreen makes total sense.

I disagree wholeheartedly that cockpit touchscreens (as a concept) are worse than physical knobs and switches. What are you doing in an FMC that requires you to split focus between operating the plane and operating the FMC? In what scenario are you going to need to create a new flight plan while pulling stunt maneuvers? When are you going to need to do anything involving navigation if you do not first have control (i.e. straight and level) of the airplane?

>This is a cost saving measure disguised as a feature.

Cost saving is not bad. Cost saving allows thousands of new aviators to enter into a field they otherwise would be price locked out of, and enables them to have a safe and enjoyable experience. Consider the AV-20 and AV-30 touch screen instruments:

https://youtu.be/hhRrNQsBl2w https://uavionix.com/product/av-30-exp/

For $1600 they provide in one instrument what you usually would need 10-13 instruments for. These instruments are not cheap; they all cost multi-hundred dollars to buy on top of drilling the holes in the cockpit and the associated maintenance to keep them all airworthy. If one component of your AV-20 fails? Who cares; throw it away and install a new one. It's dead simple.

As for UX, physical interfaces are not immune either. There have even been fatal crashes attributed to putting a lever in a stupid place, labeling a button incorrectly, not flipping a switch on time, etc. Just because you may see some horrible touchscreen interfaces out there does not mean that touchscreens are horrible.

The difference is that the touchscreen can receive software updates. At any point you can completely rearrange menus and control to address shortcomings in the interface. Doing this is practically free; the only downtime is waiting for the firmware update to complete. Addressing shortcomings in physical interfaces? You're looking at dozens to hundreds of man hours on top of buying the new equipment to replace the old.

It costs Tesla $0 per vehicle to add a new button to their cars. It costs millions for everyone else. This is not a useless capability; it makes a difference every single day.

>It costs Tesla $0 per vehicle to add a new button to their cars. It costs millions for everyone else. This is not a useless capability; it makes a difference every single day.

Complaint #1 about Teslas' from owners that I have spoken to is that the UI/UX requires total re-learning as opposed to other cars, and then when the re-learning phase is complete, it's still difficult to operate without full attention.

Sure, the argument 'Oh, they shouldn't be dividing their attention from driving.' is all well and good; but try convincing normal commuters that they should pull over to adjust the volume on their stereo. I don't think you'll win that one.

Tactile/physical components are just easier to operate without know-how, and without focus being pulled to them in its' entirety. The linear movement of a volume jogging dial is intuitive; there are two diretions on both the knob and the mechanism which is being affected -- it's direct, and obvious.

Touchscreens, usually, are none of these. Using a touch screen telephone is not an eyes off exercises, but dialing and talking to someone on any of my old flip phones or Nokias was. There was a fundamental shift from worrying about being able to do one thing very well to doing all things OK in the cell phone world, and I think that it's a shame that companies like Tesla think that this is a philosophy that should be pursued among auto-makers.

Thankfully, many automakers that were dead set on touch interfaces have begun to realize that they're not good everywhere.

Mercedes/BMW/Audi went away from buttons just to find themselves crawling back... and i'm okay with that. It makes the cars easier to operate, and less focus dividing on the road.

> but try convincing normal commuters that they should pull over to adjust the volume on their stereo.

But they don't have to. On a Tesla you can adjust the volume using the physical knob on the steering wheel; similarly you can activate or deactivate turn signals using the stalks. This is an example of touchscreens and physical controls working together, just like it would be outlandish to expect pilots to touch-and-drag on a screen rather than using a joystick.

> Using a touch screen telephone is not an eyes off exercises, but dialing and talking to someone on any of my old flip phones or Nokias was

Is it not? Can't you use Siri or Google Assistant to call a contact, or in the case of Tesla the on board voice control? Of course the voice control may suck; but that again can be fixed with software updates just as we've seen in the mobile phone space.

> Mercedes/BMW/Audi went away from buttons just to find themselves crawling back...

Because their implementations sucked. Try using an iPad vs whatever you found in their cars; why is the iPad so much better? It doesn't have to be. You could very well have had an iPad-like experience with your car's touchscreen if the manufacturer actually put in the effort to refine it. Instead we got menus in menus in menus: a horrible experience.

In the aviation world, one of the most popular apps is ForeFlight, which enables you to navigate and perform common tasks on an iPad:

https://www.foreflight.com/products/foreflight-mobile/

It is popular precisely because you can interact with information on the iPad, rather than having to deal with the shitty UX on your on board GPS or weather instrument. ForeFlight is an example of touch done right, as opposed to doing touch for the sake of touch.

Touchscreen (lack of physical response, unable to locate position without seeing) is really bad for while driving cars. I won't saying for other use cases like airplanes, tables. IANAP but I suspect that pilot needs less continuous attention for the air during fly compared to driving a car.

> I don't understand why people find novelty of touchscreens appealing. We've got to do a better job of educating consumers.

You’ve answered your own question: it’s the novelty. Plus people have been educated — by their phones.

Of course HN readers know that other than a few special cases (of which the phone is one) the touchscreen interface is worse. The problem is that white goods (washers, stoves) are cap ex and have a lifetime measured in decades. So there are plenty of people who have learned the hard way, but many more who have not yet learned the lesson.

As gratuitous touch screen is currently a fashion item attempting to differentiate between largely similar items. In another decade or so it will be considered an artifact of the 2020s, much like avocado-colored appliances and fold out stoves (we had one in the 70s) scream “1960s”.

But notice I said “2020s” (though touch screens have been around for a while), and that we had an item that screamed “60s” in a later decade. The latency in this sector is enormous.

Very high-end or professional ovens will not have this. These are gimmicks for people with too much money, and too little sense.

I don't want my toaster posting on Facebook. I don't want to link my refrigerator with my Google account. And no, I do not expect my microwave to post pics of my food to Instagram. I want a toaster that toasts properly and can be cleaned. I want a refrigerator that runs forever, doesn't leak, and doesn't build up frost in the freezer. I want a microwave that won't burn out after a year and will heat evenly.

I've managed to find all of these things, but it is becoming increasingly difficult to find appliances that aren't disposable, favoring internet-enablement to high-quality construction. I'll pay for the stainless steel... leave the wi-fi hardware out, please.

Looks like some connectivity is creeping into the high-end stuff like Gaggenau as well, but at least they still stick to TFT screens and not huge LCD touch displays, and it looks like the connectivity stuff is actually useful for stuff like remote diagnostics.

My Samsung fridge can connect to wifi, but lord knows why. I can adjust the temperature, but why I would want to do that on my phone and not on the appliance itself I do not know.

My recent Samsung washer and dryer units have wifi options because I couldn't find any local available models without, but I'm also confused what the real use case is for it.

If I'm home to deal with them, I can hear them sing. Yes, they play cute little songs when they complete their loads. If I'm not home, it makes no difference to me whatsoever if my dryer can send me artisanally curated messages about the glorious dryness of my garments.

I could see a use case if you had some sort of home SCADA controller that was locally networked and could operate a vent fan based on appliances being in operation or tell your fridge and HVAC to switch to power-saving modes based on the local energy market's posted power prices. But that's invariably not what corporate interests are offering because their incentives are so totally divorced from that of the end users.

We have the ability to design things better. SV folks should get toghether and start a startup in this space - it is ripe with many opportunities. One company comes in uses the truthful marketing to change the tune of GE/Haier/Samsung and others.

Unfortunately, we're battling vicious hardware market forces and whatever you do - it won't beat the cost of existing players who cheapout in parts and UI, especially when mass produced in China. Hardware requires serious funding to get started and ME/EE skills are disappearing from the west.

Physical controls are getting more and more expensive due to dwindling demand. Many encoders from ALPS have been EOL'ed.

I don't think we're gonna get out of this horrible UX/UI rut.

> What happened to the scratch pad and press an index key in FMCs? Age old UI and it is so crisp and clear, there is absolutely zero ambguity. Write something in a scratch pad, click the button where you wanna insert the data.

I am not a pilot but am always interested in seeing examples of great user interfaces. Can you post a link or more information about what this device is? (Bonus: the version you love and the version you now hate)

They look like this: https://i.ytimg.com/vi/RZawxJ6TLuY/maxresdefault.jpg

The amber boxes are required fields and ---- are optional. The bottom line is a scratch pad. You first punch in the data there and push one of the index switches on the side to insert it in place.

Yea its not beautiful to most people (I disagree) but it is very clear in the way it works.

Here is the new one: https://www.aviationpros.com/tools-equipment/maintenance-it/...

Got it. Thanks!

> its not beautiful to most people

I think the same could be said about audio mixing boards, for example. I've found that interfaces for professionals often look ugly to the uninitiated.

I've said many times that you should give an interface two weeks, not two seconds. Interfaces that you love after two seconds, you may come to hate after using it every day. Interfaces that are intimidating on the first day, you may come to love after a few weeks, and even more and more through the years. For me, a great example is vi.

> Interfaces that you love after two seconds, you may come to hate after using it every day

I wish more people knew this tradeoff.

I think people assume it's a step forward without thinking it through. It's a classic example of 'just because you can, doesn't mean you should'.

Reply on news.ycombinator.com