Everything with a battery should have an off switch
twitter.com
twitter.com
Bathroom scale: lovely it powers up when you tap it, but eats up 2 button cells a year for a few dozen uses a year
Kitchen scales, bit more use, at least once a day but again 2 button cells a year
electronic caliper: Same, eats up 1 button cell a year, just sitting in my toolbox
Strangely enough none of our kids toys with batteries seem to suffer from this problem
How much extra cost does a hard on/off switch add to the bill of materials?
BoM cost is not the whole story; a physical switch can greatly complicate an assembly, and may require hand-soldered wires. In addition to those cost-related factors, external switches are usually quite ugly, and getting them right can be very challenging, so it’s much easier to just omit them.
It's no different to any other bad design. It could be better.
All phones and computers can be completely powered down. They're hardly ugly switches.
iPhones (and I assume most Androids as well) can't be completely powered off. Neither can Macbooks. In fact, I'd guess that most modern digital things with non-replaceable batteries can't be powered off.
They consume a lot less power, but leave an 'off' phone in a drawer for a month starting at 100%, it will not be 100% when you 'turn it on'.
They all self-discharge [1], some chemistries more than others of course but phone batteries certainly do.
I'm thinking the battery goes dead because of leakage currents through the rest of the circuits, which were probably not designed to the nA standard.
A solid-state Off switch would also need a good silicon device to cut current consumption at the battery. Which also costs something.
This is an emotional, low-quality response to what is a measured comment actually providing nuance to the discussion.
The parent poster is very clearly not saying that leaving off the power switch leads to a better experience for the consumer - they're just giving legit reasons that a design/marketing team might use to justify doing so. Please don't try to cheapen the knowledge that they provided like this.
Where 'maldesign' is my dad's favourite word for every seeming design flaw or 'unreasonable' limitation. (-:
(they would both be shitty without a zero/tare button)
This is less true for a bathroom scale.
Amusing video explanation: https://www.youtube.com/watch?v=KnDype-j3hk
I've found that seemingly similar devices have very different notions of what constitutes a discharged battery.
I have a remote controlled humidifier and RGB LED - controllers for which use one and two infrared LEDs respectively. Both are powered by CR2025 button cells, which when new have an open circuit voltage of 3.3V and 2.7V once depleted.
The two-LED device considers a 3V cell "dead", while the one-LED will happily work at this voltage.
The two-LED device drew more current and thus the voltage dropped more.
I guess a fun middle ground - I have a set of mechanical calipers with a readout in actual numbers on the side like electronics would, but all mechanical, like clockwork or whatever it is in there, going to .001. They're infuriating to use, because the numbers are so small on the side, and at that level of detail, stupidly sensitive.
edited the decimals because it's early and my brain shut off.
For many things like a bathroom scale precision isn’t really needed anyway.
As for language, "thou", short for thousandths of an inch, is kind of the base unit in imperial machining. That's why the next one down is "tenth" in the vernacular. It's very confusing because 100 thou is one tenth of an inch, but you'd never call it a "tenth" even if "inch" is implicitly the base unit. Also confusing is that there are SI prefixes for all of these things, but they aren't in use. (Why not "milliinch"?)
Finally, one more advantage of the metric system; to measure 1 micrometer, you need a tool called a micrometer. That's easy to remember!
>between several persons
This is just wrong. Rule one of precision equipment - do not touch my precision equipment.
I'm also sure it's not calibrated to this level, but that doesn't hurt relative measurements.
Regular tracking of your weight over years does provide valuable information about your health and no normal person regularly enter it into Excell
Every time I want to use it, the 6 batteries are flat.
To be honest, I also have a Dymo label printer (very old model), and it does have a physical on/off switch, but even with the switch off it still consumes batteries like in 6 months.
Tangentially: I used to have a 3 button timer, it was a little LCD countdown display and 3 buttons. One button set hours, one minutes, and one starts / stops the countdown timer. I found this timer when I was a kid, used, and the 1 AAA battery it had lasted into adulthood, literally decades. I have since bought almost the exact same device - 3 buttons, one display - a few different times and the newer ones only last a year on the same battery.
What's going on here? Presumably someone designed a new 3 button timer, since they can't just steal an existing schematic, and I guess just nobody cares to make it good? Why don't all 3 button timers last multiple decades on a AAA battery?
I've seen the same thing with other small electronics, like the dozen or so stopwatches I've owned in my life. A few of them go strong for decades, others have their LCD display start fading in a year.
It has second hand (the kind that doesn't tick but appears to move uniformly) so there is some extra mechanical work.
If anyone knows a reliable model for AA/AAA cells that will absolutely not leak when left unused in a device for year+ I would love to hear it.
I don’t think anyone can say they absolutely won’t leak, but search these types of forums/subreddits and you find other recommendations as well.
These batteries have a regular form factor, with an added micro USB (and USB C on some larger battery sizes) port for recharging. They work really well for us and charge quickly (and conveniently — everything has a USB socket nowadays).
The price may seem expensive, but after 10 or so charge cycles they should have paid for themselves. I expect these batteries will be good for many hundreds of charge cycles.
Or are you saying you can get 1.5V lithium rechargables that work with an external charger?
The only USB chargeable ones I use are 9-volt batteries, those are awkward and I dont want to replace my charger
Some devices care about the difference.
The tiny electronics that allows the lithium battery chemistry to be charged via microUSB also converts the 3.?v output to 1.5v.
They have very low self-dishcharge rate (and even come precharged in packaging) and with recharge cycles ranging from 500 (for the highest capacity 2450 mAh AA ones) to up to 1000 (for 750 mAh AA/1900 mAh AAA ones). Also, they do not leak.
They got them to really low power usage. Some of the ones today would basically last your whole life on a single button cell.
Anyway, my insight is simply that some of them turn off the LCD after a while of non-use, and some of them keep it on indefinitely. Maybe that explains the difference between your two experiences.
The newer ones I am lucky to get a year out of the battery. Even using the on/off switch.
Then, when the remote needs new batteries and I put in the batteries sold in US stores they would last a year or two.
The batteries that came with these devices were brands I'd never heard of and never seen at any store in the US. Usually they had lots of Japanese-looking writing on them and little English except for the brand name.
1) The OEM battery was probably fresher than the one that wound its way through retail distribution and then sat in a drawer at your house for longer than you realize
2) By the time you have to replace a battery, your remote's PCB has had time to adsorb water and contaminants from the air, get soda spilled on it, etc and is going to have higher leakage. It will then always draw more current in standby.
For bathroom scales, sometimes the design doesn't permit batteries larger than button cells, since the entire scale is a piece of tempered glass with four 1/4" tall feet and a lighted LCD display that shines through the glass. I like this particular design because it's thin enough to fit under a door as it swings. Batteries in my scale last over a year with near-daily use, and mine has a very bright display.
When the button cell ran out, I popped in another one we had lying around.
One month later, the button cell was dead, after using the scale a handful of times. Obviously something was draining the button cell battery when the scale wasn't in use.
I could try and debug the scale, but given that even ten minutes of investigation would be more than the amount of time I spent using the scale in one year, I just decided to remove the button cell when the scale wasn't being used.
Two years later, that button cell still works in the scale...
In the 80s I remember seeing expiry dates on AA packs of 99, and my child-brain thought they didn't expire so they put a meaningless number there instead - I couldn't process the idea of the year 1999.
https://en.wikipedia.org/wiki/List_of_battery_sizes#Lithium_...
I've got a bathroom scale that goes several years on it's pair of coins even though it turns on only when stepped on. I suspect there's a physical switch involved, though.
One part is the one you plug the actual thermal probes into, and that then transmits data to a separate display. You can turn that one off.
Then there is the separate display that lights up and beeps and what have you: this one doesn't turn off, presumably so you can admire it say "--" whenever you're not actively measuring the temperature of something...
I had a kitchen scale that would eat through its battery pretty quickly (within a month) even when not being used. I figured it could be an interesting project to open it up and try to figure out where the stray power draw was happening, but the obviously correct solution was just to leave the battery cover off and pop it in and out for regular use.
Anyway my current kitchen scale was given to me as a present and has a mini-usb input to charge an internal battery. I personally found it to be a bit overkill, but at least I wouldn't need to get a new battery all the time. Then I realized that it can't be used when plugged in and charging. As in even if it is fully charged, it will not work when plugged in a charging. I can't believe that design idiocy. The engineers/managers/everyone involved in the production of this product should be ashamed.
In the end I found some review that compares different scales and as a reference they used an entry-level lab scale. So I figured, hey, why don’t I just buy the benchmark scale? It costs about the same anyway.
So now I have an ugly German lab scale that actually has an off button and runs on a 9v battery. Here’s hoping I never have to buy another.
Or you could go with fully mechanical balance scales.
I have just replaced a battery on my laptop, it completely failed after 2.5 years of use - they are a consumable, not a durable good. All headphones, gaming mice, etc. are headed straight for the landfill when their battery fails, it is impractical to replace that inbuilt lipo cell as it has custom size and shape, etc.
I buy devices with replaceable batteries whereever possible.
we have some toys around where the way to turn it off would be to unscrew the battery compartment and take out the battery. The toy cuckoo clock that the kid broke that now goes cuckoo at odd times was fun, but hmm, nobody can seem to find it in the house anymore, wonder what happened to it.
I decided to take it to be recycled and just bought a mechanical one. It was a little annoying at first as it's not as precise, but for my purposes that works fine.
ooo a chance to plug one of my favorite youtubers; I believe it's this video [0] that compares cheap with expensive / brand digital calipers; the TL;DR is that the expensive ones use much less when off, or, the cheap ones use 4x as much. They both keep using power when off though, so that when you move the calipers while off it will still measure the distance.
Had the same problem; got a good Vernier caliper, never looked back. Some things just don't need batteries!
I don't do highly accurate work very often, obviously.
I don't like buying non-rechargeable batteries, though, so cell batteries are out.
A dial caliper is better than a vernier caliper anyway. Still more of a pain to read than a digital one.
After someone nicked my electronic calipers, I also got mechanical ones - they are perfectly adequate for all home use so far.
I used to use electric ones for 3D prints and designs, and haven't tries that since - might not be able to get the same perfect results with mechanical ones.
I recently bought this [1]. It works with friction - you turn the nob once or twice and you're good to go for a couple weigh ins. It feels rather cheap - so let's see if it really lasts. I love the idea though. I wish we had more energy harvesting gadgets like this.
[1]; https://caso-design.de/en/p/caso-kitchen-energy-design-kitch...
Before that I installed a switch in the scale we had to erradicate battery drain.
Based on a 30 year old calculator I have, I would expect them to work fine.
A spring would then reset it for the next time.
Then all the force is on the sensor and is also a switch.
Yet another case of why fix what isn't broken. Only, I expect that an IC or two and an lcd to run a digital scale are cheaper at volume than some amount of mechanical clockwork, once the cost of batteries is pushed off on to the consumer.
But more generally most of those are using alkaleak batteries and you really wanna be changing those out on at least a yearly basis. So it actually makes since to let them have a small always on drain. (Some might be using a 2032 lithium but I can almost guarantee at least the calipers are a 357/lr44 alkaline battery.)
You don’t need power to retain a tiny bit of settings data.
I believe the real reason is because most of them have the “helpful feature” of turning on when you move it. Without something monitoring the sensor, no way to know to turn on automatically.
Of course better calipers do absolute measurements, but they’re too expensive to scribe lines with and therefore useless for hobbyists.
If you have a vintage one that is reasonably exact, they're great.
that being said, UK police considers kitchen scales as evidence that you were dealing drugs
https://www.otelo.fr/waroot/images/products/600/40130350_A.g...
Usually you can pay the vendor to calibrate the tool for a couple bucks more.
of course digital ones are much, much easier to read, but you only really feel that convenience when having to measure hundreds of items quickly to sort them into a tray or something...
And in the end, usually a Nonius type caliper is good enough anyways, and skilled users can read those to up to 0.02mm... and thats already pretty good and most of the time: good enough
and in case you do need to know the length of something down to a micron, there are other tools for that with less sources for measuring errors
so yeah: mechanical calipers rock and: you dont have to always get the brand things to get what you need
Yep.
> and use
Nope, not once you get old and tiny things are increasingly difficult to read with or without assistance.
I just bought a digital caliper with a big LCD display. It will work with Imperial or Metric units, which I really like.
My analog calipers are Imperial. I love them, but run into Metric units a lot. Got those in '88 and they are still calibrating in spec and working great. No battery. Did need light oil and cleaning a decade ago.
I have one Vernier caliper and... it is getting hard to read in some conditions. Bummer. I still prefer these in longer forms, say greater than 12"
I have used vernier calipers but the time savings of not having to scrutinize the vernier scale every time you take a measurement is quite worthwhile. Just buy a good set.
It might not be true, but it feels like battery quality has gone downhill recently (especially AA and AAA). I had very few leaks from the late 90's to about 2015. Since then, a lot more.
When it comes to kitchen scales, most Kenwood food processors double as very accurate and premium kitchen scales with backlight, as as they are plug-in they don;t need battery - great option if you have the counter space - you just place a platform where the chopping bowl goes.
After someone nicked my electronic calipers, I got mechanical ones - they are perfectly adequate for all home use so far.
I used to use electric ones for 3D prints and designs, and haven't tries that since - might not be able to get the same perfect results with mechanical ones.
I may have to try one next time I am shopping.
That being said, there are rechargeable button cells and there is a button cell charger!
Garage door openers can last 10-15 years too; those use the battery to transmit a signal, not just to hold a piece of SRAM alive.
The bathroom scale is just a poor electronic design.
Factory A: Charges 7c per unit with a power switch.
Factory B: Charges 3c per unit with no power switch.
Wanna play the which choice did the C-Suite make game?
I highly prefer AAA cells for those cases, you can get high-quality rechargeable cells from Ikea for a few coins.
I got this fancy tire gauge+filler. It uses a couple coin batteries. They live around 6 months with it "off". I use it a couple times a year, so basically every time I go to use it it is dead. Removing the batteries requires disassembly. A year or so ago I soldered in a couple AA batteries in place of the coin batteries, it's been good so far, but I'm waiting to see.
On the other hand, I have a Mitutoyo caliper, which also uses a couple coin batteries. I sometimes forget to turn it off when I put it back in the case, and so far, over 6+ months, it hasn't run the batteries out.
So it CAN be done. Both devices are basicaally the same: LCD display and they measure a sensor.
Back when I was in HS, I had a Casio calculator watch, but it ate through batteries (3-6 months IIRC). The bracket for the strap broke, so I just taped it up in my locker. Soldered it to a D battery, and never had to replace it again. :-)
But I don't want that 100% of the time. With a power switch, I get my ideal user experience (fast reconnects) and I have control when to apply that user experience (on/off switch).
I’m currently working on a new device that will monitor my weight for 3 years without recharging. (It reads continuously, it’s a scale for Bottomless.)
Same with my Lenovo X13 gen 2, there is a bright white led next to the power button… just why…???
Anyway I've got a speaker in my bedroom with an LED that won't turn off, and a USB-C hub that has a tiny blue LED that requires me to disconnect the whole thing first. Really annoying.
I've found if you use a dremmel with a fine bit, you can usually turn off the LED.
otherwise put something over it.
but yes, my solutions should not be necessary
Thus far this has worked on every ThinkPad I've had; but again, I don't know if it will work on your X13 (or how to do it from Windows or other systems), but something to look at if you haven't already.
I checked, and my kernel doesn't have that option, and I can change it just fine ("echo 0 >/sys/class/leds/tpacpi::power/brightness", as root of course).
I keep a roll of black electrical tape in my office :)
If I turn off my iPhone, MacBook, or Google Pixel phone, are they really off?
What are the security implications of "modern standby" on laptops where the CPU and WiFi are always on even after I fold the lid down?
In fact, even a non-nefarious company might do that so a sudden power loss doesn't cause data corruption.
Pretty annoying that my closed laptop, that's in another room, can drift in and out of being "found" by my bluetooth headset and stop music playing.
New Windows sleep (""sleep"") states are bewildering.
My S/O's new laptop only supports "S0" or "Modern Standby"[0]. It can sit there closed up in its deepest supported level of "sleep" and have its fans spinning and keyboard backlight lit while downloading updates or whatever else it deems important.
Sure, introduce something like S0 and spend time improving it. Try to "perfect" it like how Apple has seemingly perfected its low energy use modes on its devices. It will probably take MSoft 10 more years. But don't also artificially restrict the use of other sleep states. On the new laptop it is supposedly a hardware restriction, but I'm guessing it's more just some firmware toggle (IIRC some laptops have had S1-3 re-enabled via firmware updates, but I'm not entirely sure now).
Edit: it seems that MS has patched a bunch of regedit tweaks that used to allow re-enabling more sleep states[1], though it depends on exactly what laptop model you are on as well I suppose.
[0] https://docs.microsoft.com/en-us/windows/win32/power/system-... [1] https://superuser.com/questions/1179987
When my iPhone's battery is depleted the screen goes blank. If I try to turn it on, an empty battery logo appears and it says "Find My enabled" underneath. So, I assume not.
The battery will shut down when the voltage reaches a cutoff, not when it is truly depleted.
With 5G and future connectivity, it feels as there will be push towards not being able to block the network access of individual devices, such that a TV or fridge will just speak to its server directly via some XG/IoT network which the user does control, without any user ability to block this communication.
After losing control over connectivity, it seems as power switches will be the next thing we'll start to lose control over, for engineering and ergonomic reasons, but also for control reasons. Such a shift may be further pushed by more ubiquitous and long-range wireless charging, if it gets to the point where devices such as our phones can be sufficiently charged simply by being inside buildings.
Just thought I'd throw in my $0.02 vote for the ability to power off things.
Does a smart fridge still work if you rip the modem and Wlan boards out?
Isn't it because the microwave is shielded against a very specific frequency, that is much lower than the 5G one? So that the 5G frequency penetrates the shielding without much trouble?
(I'm really not sure, that's my initial guess, but the questions are not rhetorical.)
How does one ensure that a product is built this way, to avoid it?
I understood that a number of current 5G capable devices - e.g. telephones - can be disconnected the usual way (a button in the (quick)settings that toggles the connection).
However for a device such as a TV, once it is sold with a network plan arranged by the manufacturer, there doesn't remain much of a point to allow the user to normally control whether this connection is active. For a while, an option would remain to disable or configure such an option, but over time it seems likely that the user would no longer be able to control this directly.
It gets even easier for devices such as a smart lightbulb or a smart toaster - devices for which there wouldn't be any direct user interface. Once these communicate via wireless wider-area networks, the trend seems to be clear.
On one hand, it is nice to have an app that shows my stats, and my doctor can also remotely access how my therapy is going.
On the other hand, it also phones home to the manufacturer & my insurance. When I don't hit my usage target for a specific time period, insurance stops paying for it and I have to personally pay out of pocket. So now you add the stress of not only complying with sleeping with an uncomfortable apparatus on your face but also if you don't keep it on enough (I take it off unconsciously in my sleep) you start paying thru the nose for it. Vicious cycle...
And that's why they have tracking built-in. You're perfectly able to buy the machine yourself without insurance and then you don't have to turn on the modem.
And also after you've used it for a while they give it to you and then you can turn off the modem.
Personally, I just bought my machine outright. Yeah, it was close to a thousand bucks. But it's been going strong for years, so I'm happy with the investment. It has never been allowed to connect back to resmed.
However I disagree about insurance - they are paying for it, so they can chose how they want to monitor usage. If you don't like it, then pay for it yourself, exactly like you did.
Also, the older machines don't have remote access, they have a little card you have to bring periodically to the Dr. so they can confirm to insurance you are actually using it. Remote access was meant to make it less annoying for patients, and cheaper too since you don't have to go to the Dr. so often.
Pricing it into the premium is a terrible idea - it would raise premiums for everyone, for very little gain.
Enough to matter? Let's put this in context. A typical CPAP machine will go for years before needing to be replaced, and can be purchased retail for about $800. Compared to routine medical expenses, that's nothing. Heck, the reduction in costs for monitoring compliance would probably lead to -lower- premiums, not higher.
You can buy your CPAP outright for around ~$1500. This is a problem for many people, but probably not HN readers. The telemetry basically facilitates an interest-free loan that lets you pay for the machine over a couple years. More favorable financing than a credit card, so people aren't getting super screwed here.
https://en.wikipedia.org/wiki/Continuous_positive_airway_pre...
Amazon Sidewalk [1] is pretty much already there (assuming uptake is high enough, and I believe it's opt-out, so it probably is). From what I recall, the bandwidth is low, but that didn't seem to be a technical requirement but a political one to prevent leeching so much bandwidth the user notices.
I think you're right, though. Usage is going to proliferate until laws reign it in, or someone develops an "electromagnetic radiation firewall" that allows fine-grained control over physical signals. The only other option I can think of is a soldering iron, but that's not terribly feasible. I'll take a soldering iron to a $40 part, but I'm not going to risk frying a nice new refrigerator or other expensive appliance.
1. https://www.amazon.com/Amazon-Sidewalk/b?ie=UTF8&node=213281...
It won't be many years until the bluebottle fly which came in the window, round the room and off into your house, is indistinguishable from a camera drone from the census bureau, your insurance company, your nosey neighbour. And then many more years until Vernor Vinge's "smart dust".
Check out the size of the cameras on the Tobii 3 eye-tracking glasses, for example: https://www.tobiipro.com/imagevault/publishedmedia/71ru67v90...
AVR Picopower has been around for about 10y now. These devices dramatically drop power consumption to the point where batteries should last a very long time, definitely >1y, more like 3-5y.
The fact that modern devices drain a lot of juice is either sloppy SW design or sloppy HW design, since I believe the cost for microcontrollers is very similar.
That is not trouble, that is the whole point.
Just lookup circuits.
It's the main reason my Xbox sees so much more use than my PlayStation. My PlayStation controller is seemingly always dead when I go to use it, but I can just pop some new rechargeable batteries in the Xbox.
There should be a law for this. Even for cellphones. When corporations market themselves as environmentally conscious, but don't allow the replacement of common wear items in their gadgets and products, then that's how you know they are full of shit.
I'm wearing a Fitbit as I write this. User-replaceable battery?? It's 30 grams, not counting the band. How much bigger and heavier would it be with an openable seal? The seal has to take a lot--it's got 50m water resistance. Likewise, the battery needs to be secured against motion--that either means glue or mechanical pressure against it. The latter likewise increases size and weight.
Other than these cases I agree with you. Batteries should be readily swapable.
Or with things where battery life is a much smaller issue than forgetting to turn the thing on (e.g. smoke detectors).
Or with things that draw unspeakably low amounts of current and where such switches would be hard to implement (e.g. digital wrist watches).
Other than that: Yes please.
What an odd example to pick.
The only place where smoke detectors are frequently switched off are remote huts that might be empty for the better part of a year. And there the best strategy is to unmount the whole thing, pull the battery out and place both things somewhere really obvious so the next guys notice immediately what needs to be done.
Not at all! You might reconsider your opinion when one of these things goes off at night for no reason and you find out its battery isn't removable.
A deactivation button is different from an power switch where the thing is off when the switch is in the wrong position. And if you don't notice people die. I didn't check the ISO norms on smoke detectors but I would be surprised if power switches wouldn't be explicitly forbidden in those norms.
And sure you can get a detector that doesn't conform to the norm, but then the insurance won't pay in case of a fire.
There's more reasons than just being able to plug a different BT controller (which is great in times of supply chain problems)
You're also usually stuck with a different toolchain than you're used to, so you will have to learn all the quirks. And they may not have enough pins or not the right type..
But I get the point. It's kinda funny to read the "w000t my C64 now has Bluetooth". When the BT controller is actually much more powerful than the C64 :P
Lies! You take that back!!
[0] https://www.amazon.com/Fisher-Price-Bright-Beats-Tumble-Beat...
I know there's the baseband processor that does its own thing, but I would've thought the userland "off" state shuts that off, along with everything else except for whatever watches the power button and battery charging.
The FCC for one would throw a fit if airplane mode or power off still actively transmitted.
Battery life is another giveaway, I've had android phones turned off in a drawer for months and they still have nearly 100% charge when turned back on. If they were using enough power to make any kind of cell connection, the battery would be dead in a few days.
The link you have there even says "When your phone is turned off or runs out of battery, you won’t be able to track its real-time location"
You haven't seen hell before you've seen a speaker with Bluetooth that you can't turn off and you can't stop your neighbours from accessing. Or more specifically, heard hell.
My phone is paired with my car--but the bluetooth audio in the car will pick up whatever signal it hears regardless. Including from the car next to me.
For SMD leds it's pretty easy to destroy them directly by using a hot enough iron - they're heat sensitive and die easily. Just touch it with your iron and boom, annoying light gone.
Like, if the battery is rechargeable, 5-25uA is pretty much zero drain.
If it's disposable, you get about 1-5uA before it's A Problem(Because the only good use for a disposable is on stuff that only gets changed once in multiple years).
In a lot of cases the switch is the least reliable part, and it makes case design harder, and maybe compromises waterproofing.
They can also "degrade" when they're not in designated voltage range. This is one reason why most batteries are soldered, because they need to be regulated at all times.
So no, you cannot "turn off" a battery. A battery should either provide power, or be plugged to a charger.
I’ve seen several “present credentials” demands here lately. I find it a bit odd.
Citation: search for the terms “self-discharge”. I dare you to find something less kosher than the wikipedia page. I already find the wiki page less than kosher.
Correction: MUST have a way.
(this is just a random picture of them)
https://cdn.bueromarkt-ag.de/product/1eb892022d09365e7dd2c7b...
PS: foone knows what a chip on board is, right? I'm assuming they just used weird phrasing, but the fact that they called it a "blobbed over IC" kind of threw me off
I understand why a crucial medical device doesn't have an off switch that could be engaged accidentally, but I don't understand why they tie the function to being connected to a power source.
But on the other hand, I think it should not be the job of the consumer to "fix" shitty design choices. If we accept devices that are borderline defective by design instead of sending them back to where they came from, what we'll end up with are more devices that are similarly broken for the sake saving a few cents.
The only way we can truly fix bad design choices is by making them unprofitable for the manufacturers.
Cars are ubiquitous. They are everywhere. We all relate to them without thinking. Cars run on batteries. Even IC runs various systems using battery power when parked. But do cars have an off switch? To turn them truly off, such as when storing for a long period, the procedure is take a wrench and physically disconnect the 12v battery. So I disagree with the author. At least some battery-powered products are doing perfectly well without off switches.
Other IC vehicles, notably aircraft, do have true off switches that fully shut electrical power at the source. So it isn't anything inherent to IC technology. Consumers have spoken: some things do not need off switches.
And I don't agree that "consumers have spoken". Consumers don't design cars; their feedback loop is long, lossy, and easily ignored. Profitability drives car design, not what consumers want.
There's a big Microsoft logo badge right on the center console letting me know who to thank for that.
As a programmer my day job is "this isn't working the way it's supposed to, lets figure out why and fix it." I can't fathom ever wanting to deal with that stuff when I get off work. A lock and key have been working just fine since the Roman times, it's relatively easy to conceptualize how it works, the things you can troubleshoot on it is a relatively short list, the UX surface area is small and predictable (it's just a lock and a key). I could go on for a long time. If something got along just fine before bluetooth and wifi existed then I never want to own a version that requires bluetooth or wifi.
I have a safe with an electronic lock. The batteries that run that lock are on the *outside* of the secure area. There's a little plate you can slide off and fish out the battery pack. That's how it's supposed to be done.
When filtering for products the cheapest one and most expensive one should not even be considered.
That is changing. Several states, including California and New York, no longer allow replaceable battery smoke alarms. They require long life sealed batteries. Other states are in the midst of doing so.
They are doing this because smoke detectors have a limited lifetime. You are supposed to replace them 10 years after manufacture. However, they don't suddenly stop working at 10 years...they just get less and less effective over time.
Many people don't realize they are supposed to replace them, or know it but forget. I didn't know, for instance, until mine was ~17 years old. It still went off when I'd sear a steak if I forgot to close the door between the kitchen and the room it was in so there was no obvious thing to tell me that it was likely losing effectiveness.
By going to units with a sealed battery meant to last the lifetime of the unit they hope to make it more likely that people will remember to replace them. When it is time the units will beep like the replaceable battery units do when the battery is getting low.
I would guess there are also people with the replaceable battery units who when it starts signaling low battery take out the battery planning to pick up a new one next time they are shopping, forget, and take a long time to remember if ever. Sealed units will help there, too.
BTW, when buying smoke alarms it is probably best to avoid Kidde (also avoid them for fire extinguishers). They are prominently featured at Home Depot and Walmart, and can be a little hard to avoid if you don't take a bit of care. Here's why to avoid them, from the Wirecutter.com article on basic smoke detectors:
> With placement on the shelves of Home Depot, Kidde is the most prominent competitor to First Alert, but its overall track record is, in a word, disturbing. In 2018, the company recalled more than 450,000 dual-sensor smoke alarms; in 2016, it recalled 3.6 million smoke/CO alarms, and in 2014, it recalled 1.2 million smoke/CO alarms. In addition, since 2005, Kidde has enacted three separate fire extinguisher recalls of 470,000 units (2005), 4.6 million units (2015), and 40 million units (2017). Worst of all, in early 2021, a federal judge ordered Kidde to pay a $12 million civil penalty “in connection with allegations that the company failed to timely inform the Consumer Product Safety Commission (CPSC) about problems with fire extinguishers manufactured by the company.” For these reasons, we can’t, in good conscience, recommend any Kidde products to our readers.
> Although First Alert is not immune to recalls (it recalled nearly 150,000 smoke alarms in 2006 and roughly 600,000 fire extinguishers in 2000), none of them are recent or on the scale of what Kidde has had to do.
At some point when Apple introduced the touchbar someone decided that you don’t need to truly turn it off… and it automatically turns on when you press a key. Meaning that you can’t cleanup the keyboard without draining the battery first. (I tried looking for undocumented configs to turn this off… for years -well more or less since 2017)
During that time, I maybe tap the buttons a dozen times as i scan around to the app and the thing i want to watch.
I have to replace the 2 AAA batteries inside that thing easily every 3 months. WHY.
Yea, fair
My Kindles all die fairly quickly even when off.
A lot of my electronics is plugged in via bus strips with off switches.
Having a separate "tender battery" for car-is-off power needs is a fine idea. Another idea is for a battery voltage monitor that will cut off the battery if the voltage starts to drop.
I regularly have to recharge it.
I would not like a pacemaker with an off switch you can trigger by accident or have it have user replaceable batteries - seems dangerous to your health
Alright buddy. There are plenty of devices I have without an off switch that work perfectly fine. In terms of controllers, the Xbox controllers I have will go to sleep after not having been used for a little while and conserve battery that way.
The problem isn't lack of off switches, it's the lack of features related to the lack of an off switch.
Nintendo, PDP, hyperkin, 8bitdo, everyone else.
This is endemic to the PDP controllers. If I want to turn off an Xbox One controller, I can pop the batteries or it'll turn off after it's not connected for 30 seconds. PDP wireless Switch controllers? 10 minutes later and they're still hunting for their paired console, often with no way to forcibly turn it off.
Even 8bitdo do in several iterations have Xbox style battery compartments and their fighter pad has a physical switch.
PDP can ship these cheap because they can have loss leaders in the low cost options, making up for the difference with their high end controllers and even their rather extensive selection of other console controllers. Ironically, their switch joycon charger dock is one of the best and some of their headsets are rivaling even the better Gamer ones now.
Also quartz wrist watches.
Turn that quartz watch into an automatic that you wear seldomly. (Aka you need to readjust the time each time you wear it!)
Oh, great, this debate again! I shall use this post as ammunition next time I am arguing with our stupid design team about why we need an off switch on the stupid product we are designing, because it is a giant pain to live with devices you cannot turn off.
(It is especially fun to put something together for the Very First Time, then realize you can't turn it off... and you need to turn it off because it's the Very First Time this thing has ever existed at all, and it isn't quite right, and you need to fix something. But it can't be turned off, and this is somehow by design.)
(Our design team will never listen. This is because they, one, don't listen to anyone ever; and two, never use the products they design. Oh, sure, there might be a user study or two. But they never actually have to live with or work with the damn things.)
> They never seem to just use the bluetooth module as the sole microcontroller
We actually do this! It works pretty well.
> and also it's not like microcontrollers are expensive these days. with a blobbed-on IC of some 8-bit custom non-programmable thing, they're probably spending like 1-2 cents each
Yeah I wish we could get those prices. But we don't do COB (chip-on-board) and we only use microcontrollers we can get and read datasheets for, so we pay at least 20x that price. Maybe that's why we don't have any clients?
Yeah, I do software and Ive found the same thing: if I’m building it, then I’m actually the first user of the software, and I have all sorts of feedback for the designers. Luckily I’ve worked with some great designers who were usually more than happy to adjust the designs though.
Yeah, I see what you mean.
And even if google translate can help, there are of course other problems
20x might be the actual cheaper option
I wonder the percentage of consumer (and business - have you ever used a POS in a pub?!) electronics this applies to? It's gotta be high 90s. I've lost track of the number of times I've lamented "did anyone actually use this?"
We have a combination oven and microwave (yes...) that seems to have been designed with the primary purpose of enraging the owner.
The microwave can only be set in 15 second increments. But you can fine-tune it, in five-second increments, by spinning a dial. Except that you can't do that while the oven is on.
The microwave makes a noise while it's operating. When it's finished, it will beep unobtrusively and then continue to make the operating noise. You can stop it from making the operating noise by opening the door.
The microwave can't be replaced, because it's part of the oven.
Mine is similar, also a combi. You spin the dial and it will go up in 10 second increments to a point, then in minutes, then in 5 minutes, then ten minutes, etc. Fine, I get it, it's like floating point. But it starts doing 5 minutes at the 10 minute mark! So you can't do 11-14 minutes. Which is actually about right for some things in an oven.
You also can't add more than 10 minutes to a short time without stopping the oven (this, I assume, is intended to be a safety thing to avoid being able to accidentally add hours of cooking without positive user intent).
And I do wish it would just stop the fan once it stops cooking, but I suppose the electronics next to the hot oven don't suddenly need less cooling just because the hot metal box doesn't have food in it.
I do like it though because it's much more efficient for small things than a big oven and I don't have space for a toaster oven and a microwave. And it has a much better UI for setting a delay start than the 6-button interface of the big oven.
> I do like it though because it's much more efficient for small things than a big oven and I don't have space for a toaster oven and a microwave.
You may have gotten the wrong idea. This isn't a combination where the same box functions as an oven and a microwave. This is a microwave that is for no reason physically attached to a separate oven - it takes up all the space of a big oven plus a normally sized microwave. The only thing they share is the display panel. And yet that didn't stop anyone from disabling microwave functionality while the oven is on.
There is absolutely no advantage to this compared to having an oven and a separate microwave, and many, many, many disadvantages. It continually boggles my mind that anyone ever designed, manufactured, or purchased this thing.
Perhaps the actual audible alert on yours should be louder.
Actually I guess they both shave off weight/cost and design for built-in model first (that are naturally more heat-constrained) and make standalone variant as an afterthought.
I would happily invest in kitchen appliances that have a silent mode for apartments and shared living spaces.
My microwave beeps at the end, and then beeps again 10 minutes later once the food has been removed. My old toaster beeped 5 times when the toast popped, which is unnecessary when the popping action is loud on it's own.
When we moved into our current house, it had a microwave like that. When we replaced it, my primary requirements were 1. multiple fan speeds, and 2. a nice, simple touch pad for entering the time; preferably one where you didn't need to press a button before starting to enter the cooking time.
The dial thing is... infuriating. It never once found a situation where it made using the device better, and in most situations it made it worse.
Digital dials can be nice in some cases.
My parents had an oven with a dial for the temperature. Spin it either way to increase or decrease the temperature by extra 5 degrees. I loved it. It was so much better than the touchpad design. The range of actual temperatures in an oven is pretty small and I found this much more friendly than punching in the temperature, getting it wrong, cancelling, and punching it in again. It was a “finger tip” dial so it was quick to spin. Probably horrible for people with bad arthritis or other hand mobility issues, though.
I wouldn’t mind a microwave with a similar dial for time. Again, the actual range of times in typical use (for me) is not that high. The design described with a two tier system for adjusting time seems moronic though.
The range of temperatures you use is about 200 degrees?
For the microwave time, it was a horrible experience pretty much every time I interacted with the device. A keypad is simple, intuitive, and easy to use.
“Okay, I want to cook for 3 minutes.”
‘300’
”Ok, the numbers don’t work until I press something else. Hmmm”
‘Time Cook’ ‘400’
“Shit. I hit the 4. Where’s the cancel?
‘Cancel’ ‘Time Cook’ ‘300’
“Where’s the start?”
‘Start’
It's a bit limited, in the sense that you can't set the time to e.g. 17:27. But who needs that?
When my old mechanical kitchen timer died on me, I bought a digital one with a dial, in the hope that it work the same way as the one on my microwave oven. Boy was I wrong. It kinda increases the interval when you spin it faster, but in a completely unpredictable and impractical way.
So I guess it's easy to make impractical dials, but it certainly is possible to make very well functioning dials as well.
Check out Chesterton's Fence.
So, it wouldn't be a good idea to just rip that part out of the design. Even if it may be annoying. Perhaps a better explanation is needed, like some models showing "cool" in the display. But it's per se not a design oversight.
Chesterton's Fence is about not changing things before understanding what led to the current state of affairs. I think the principle easily extends to complaining as well.
I could criticize cars with combustion engines for having one and being so noisy/stinky/... Then I could go on claiming that they should just turn them off, cause there are cars without combustion engines which are less noisy/stinky/... I'd be missing that there is more to it than just turning off the engine. I'd need to replace it with something else, like an EV. Maybe that did not exist yet when that car was build. So, claiming that this old Ford from 1970 having a combustion engine is stupid because "this is not a problem that other cars have" is missing the point. I could go rip the engine out because I don't like that it's noisy/stinky/... But I'm in for a surprise here. The vehicle won't move anymore. Something I'd have understood had I followed the principle put forward by Chesterton's Fence.
> Chesterton's Fence tells us that there is no cost to eliminating this ridiculously awful functionality
No. It tells you to first understand what you are doing. Blindly claiming something misses that part.
Are you me? Lol. Sometimes I use software where I wonder "am I the only one that bought this and tried to use it?" I think part of the issue is that requirements get written by people who have no experience. They've never done the job the software will be used to augment and they don't know anything about software development. Then the requirements are used to outsource the actual design and development.
Using POS as an example (in a country with tipping culture). Imagine if a requirements document for the payment system said "must include an option to tip", but the design and development was done in a country where tipping culture isn't a thing. You'd likely end up with the "tip" option hidden away in a hard to reach place while everyone who uses the software knows it should be prominent and easy to use and could immediately tell you the design is poor.
You might enjoy [1] about a microcontroller that's ~3 cents, in quantity 10.
Of course it's only one-time-programmable, and generally pretty basic. And that's a pre-supply-chain-crisis price - these days prices have risen to 9 cents [2]. Still very cheap though!
[1] https://www.youtube.com/watch?v=VYhAGnsnO7w [2] https://www.lcsc.com/product-detail/Microcontroller-Units-MC...
An ATTiny10 or PIC16F is so much better documented, so much better supported, so much more featureful, and so much better in every way that the average small-volume product design house will never make up the 40-cent price difference per part. How many development hours or support hours can you bill before you're penny-wise and pound foolish?
Their compiler/macro-assembler also seems to be working just fine, once you get past the UI from early 2000s. The error messages that must have been auto-translated from Traditional Chinese, but you squint a little bit and get the idea.
I think people really don't want e-waste. They don't want oceans (and now bodies) full of plastic. The industries have just done a great job of convincing them nothing can be done about the problem.
I think it's a bit more complex than that; a lot of properties that distinguish a "solid product" from "e-waste" are not at all obvious when you buy it, and it's not illegal and it is cheaper, and people got to make rent too, so... yeah. I agree: in the abstract most people don't "want e-waste", but they also look at costs and other factors, and they do "want e-waste" once you factor that in.
It probably shouldn't be strictly illegal too, but now you have a situation where:
1. Consumers blame industry and say that government should do something.
2. Industry says they're just producing what consumers want and that it's not illegal.
3. Government says they want consumers and industry to have free choice and call for different consumer actions and "self-regulation".
So basically, everyone is pointing at everyone else, and nothing gets done. And the arguments from the various sides aren't necessarily bad or malicious either: regulation does come with a cost, and cheap stuff does have its use.
Personally I think factoring in external environmental costs in to the product price would be good, but it's not so easy as it will make things more expensive, and already enough people are struggling to make ends meet (as the current high inflation rates show) so it all ties in to a lot of other issues as well.
> I am pretty sure the market disagrees with you.
Not applicable - efficient market hypothesis requires a host of conditions to be true, and all of the following are unmet: information symmetry, sufficient number of suppliers, lack of monopoly effects, and rational consumers.
I know, I know, the industry doesn't want me to keep my device working longer than the lifespan of a Li-ion cell.
For something long-lived and expensive, I will pay extra and accept a less sleek design for standardized removable batteries. I research most purchases of durable goods. Of course, I may not be the most profitable customer since I don't plan to buy a new widget every six months.
When I use something at my home, I usually have a constant stream of minor annoyances that could easily be fixed by the manufacturer. However, when I'm shopping for a new product, I dont remember any of them.
E-waste is a huge but so far mostly overlooked problem. Built-in obsolescence is great for exploitative businesses and bad for basically everyone and everything else. Unnecessarily restricting hardware repair or replacement might be the largest contributory factor in this problem.
Time to regulate.
More ideas: "online account required for use", "collects location data", "phone call required to cancel subscription", "memory use: 350 MB" (just like calorie counts!), and so many others...
If we require labels, without the necessary strictly defined definitions, today's non-removable battery becomes tomorrow's "removable" battery. Because, hey, it is possible to remove it. Good luck getting everything back together if you aren't an engineer or extremely careful in how you disassembled things.
I'd sooner have a 18650-family (e.g. 14240 or 14500) become standard than a bunch of AAAs.
Not compared to alkalines, not since about 2005. Voltage regulation (i.e. a boost converter) plays well with NiMH rechargeables too, though not with alkalines.
That's assuming one of them is an appropriate size and voltage for the application. If more than one starts to seem like a good idea, consider a larger cell, higher voltage chemistry, or both.
Not in my experience: our main TV is connected to a 2008 computer running Linux Mint that we use for netflix and amz prime. I've been using it in this way since 2016. The remote is an off-the-shelf wireless pointer thingy with a few buttons on it (works like a Wii controller) that takes 2x AAA batteries.
I've been recharging those 2x AAA batteries since 1996, and it still works fine enough that I don't feel compelled to replace them.
Getting a bunch of consumers used to buying spare Li-ion cells and using external chargers might be a non-starter, but it's possible to have both onboard charging and removable cells so the upgrade path is there for those who want it.
There is probably room for standardization of a compact, rectangular, rechargeable battery of roughly those dimensions. Is someone pushing for it?
[1] https://www.apotelyt.com/camera-power/canon-g7-x-mark-iii-ba...
It seems to me the size of the G7X could have been designed around a 14500 cell instead.
So many red flags here.
If you don't mind me asking .. if it's that bad, why do you still work there? Is there a shortage in suitable jobs? (Honest question..)
Does that come from clients?
I used the same 3-in-1 latch circuit for power on/off, pairing, and software power off for 12 years.
A good enhancement mode NMOS FET can be as good as a physical switch at low voltages.
If you really need nano-amp level leakage current AND very low drop, you can consider turning to beefy depletion mode devices, which gate the user will discharge with the button press. A nice benefit is that you can directly sense the button press with an MCU.
I once even did a bluetooth mouse with this "hardware" switch, and a graceful shutdown: we sense voltage on the buck capacitor, and detect when the battery disconnect triggers.
The few microfarads in the capacitor are enough to send the last update with the 0% battery level, and a disconnect command.
Both! There aren't many of those out there, so I probably just outed myself. Oh well. Keep it quiet please?
> A good enhancement mode NMOS FET can be as good as a physical switch at low voltages.
Yes, it can work very well. You are correct that it need not be a physical switch. But it still does nothing if it is never brought out to the device's interface. And physical switches are still preferable if the software is broken, as it often happens to be during development.
> you can consider turning to beefy depletion mode devices
Depletion mode devices are not cost effective. If you think they are, point me to one cheaper than the LND150. (And, yes, I do know how to use them. Many designers do not.)
I increasingly subscribe to the thesis that many designers and design teams design primarily in an effort to impress other designers. They don't have to use the things. They often don't even really care if the things work at all. They care that other designers - their artistic peers - are impressed.
This does not generally result in what we the customers or users would regard as good design. The phenomenon reminds me of software engineers I've worked with who were more interested in building something with event sourcing and stream processing and functional languages rather than solving the problem more readily.
Designers nowadays are in a fashion to get impressed by things that make the results of their work worse. That leads to completely different problems.
You always get some people who build asinine crap in order so show off their "mastery" of their craft. It's just that product design and web development have more money sloshing around to provide for that kind of indulgence than residential plumbing and semi trailer manufacturing do.
To put things in perspective, this is how we get stuff like the Eiffel tower.
I think most people disagree with me so I don't think it's going anywhere. But that's my view of it.
That was a popular opinion when it was built. Eventually it became iconic, but nevertheless I don't see many cities rushing to build their own.
I'm not sure I got the point across. The whole point is that focusing on design is how we get outstanding, high-quality results. It makes the product more appealing and can even make it transcend it's status. High quality design creates value with negligible impact on cost.
The Eiffel Tower exemplifies that. It was designed as a temporary exhibition but it's impact transcends that of a mere temporary structure. It became the landmark of a European capital, and the symbol of one of the richest, most successful and culturally dominant nations on earth, and in the process generates fortunes in revenue due to tourism.
Another good example is the Sagrada Família cathedral in Barcelona, or pretty much each of Gaudi's work. Even the Gaudi's Park Guell, which failed as a real estate project, is a resounding landmark and tourist attraction.
I then heard more about the designers at Microsoft for Vista through Windows 11 which deeply confirmed most of my prior negative assumptions. It sometimes really is as simple as folks who think they know better than anyone who uses their product; people who refused to dogfood or even touch the thing they ostensibly work on, in any given day.
They were bad enough, but then Visual Studio 2013 HAD THEM TOO.
That was triply bad: ugly, took more screen width than Mixed Case, and perhaps worst of all, to a programmer they looked like errors.
Most of the languages we work with are case sensitive, so FILE and File look like two different things. It was a constant irritant.
The chief designer for VS2013 had a blog post about the new design, and naturally in the comments there were a lot of complaints about the uppercase menus.
The designer eventually replied: "Thanks for your feedback. We decided to keep the uppercase menus because we want more energy in this part of the interface."
More energy. I am not making this up.
They eventually came to their senses, and the next releases of Office and Visual Studio went back to conventional Mixed Case menus, just like every Windows app has used since the beginning.
(Note: my use of uppercase at the top of this message is not for emphasis or shouting, it's for illustrative purposes.)
In other words, we jumped off this bridge because all our friends did to. Consistency... with other Microsoft products. What a joke. What about consistency with virtually every other Windows application, both from 3rd parties and Microsoft's own history? Citing consistency to justify such a bizarre departure from the 20 year norm is truly insulting.
It always seems to be new-age mumbo jumbo with these sort of design people. Design needs to be given back over to the engineers again. I know that suggesting is going to draw ridicule; it's easy and popular to point to the worst engineer-designed interfaces to ridicule the design abilities of all engineers. But form needs to be balanced with function and these artsy new-age designers who justify everything they do with faux-poetic metaphors have been a disaster. There needs to be balance.
Engineers can be taught the pragmatic virtue of interfaces designed for common people. Using standardized widgets and guidelines, engineers are perfectly capable of creating interfaces that are intuitive to users, even if they don't satisfy the dedicated designer's sense of aesthetics. Just look at the GUI libraries from the 90s and compare it with the dumpster fire of modern interfaces. In the 90s an engineer designing an interface would use a button widget and the button would look like a button. Today designers make everything look sleek and special to leave their artistic mark and I am left with no fucking clue what I can even click.
No, no, then you get the open source approach:
In case the desktop shortcut for your application is not available with the /usr/share/applications/ directory you have and option to create the Desktop launcher manually. In this example we will create and Desktop application shortcut for Skype application. Obtain the following information for any given application you wish to create shortcut for. Below you can find an example:
Path Application Icon: /snap/skype/101/meta/gui/skypeforlinux.png
Application name: Skype
Path to executable binary: /snap/bin/skype
$ which skype
/snap/bin/skype
The general dysfunction of open source design is much broader than UI design and has little to do with engineers in general being incapable of good UI design. You might as well linux bluetooth woes to claim that engineers shouldn't be allowed to write device drivers. Whether it's designing a GUI, an API, or any other system, there will be some engineers who are good at it an others who suck. To get good design of any sort out of engineers you need an organizational structure that promotes talent and weeds out the hacks. The open source scene broadly lacks such structure and struggles to do this.
What you're talking about is a deficiency with community driven model of development, not a general deficiency of all engineers. Put 20 random chefs into a kitchen with no organizational structure imposed on them and your restaurant will be a disaster. But using that to claim chefs can't run a restaurant is absurd.
Translation: Here is vague explanation that distracts from the fact that I made this choice emotionally rather than rationally.
I see this time and time again. People are simply reacting emotionally. If they try to explain why they behaved a certain way, they offer an explanation with something that sounds rational but isn't actually the reason.
In this case, the designer probably was hurt that people didn't like their work and stubbornly resisted it.
The default white title bars that blend seamlessly into any white content behind them is a perfect example of this and so many people don't even realize they can change it.
Windows usability has declined since Windows 7 which had a lot of things that seemed like the designers were trying to improve their own lives while using the software. My favorite was the thick borders that were easy to grab for resizing, etc., even on high DPI displays. Compare it to Linux UIs where you often have to grab a single pixel border on a 4k monitor to resize a window.
Thankfully Win 10 still has large targets for grabbing window edges even though the border is only 1px.
On Linux you hold the super/windows key and click/drag anywhere on the window to move it, or right-click/drag anywhere on the window to resize it. It's much easier than using thick borders. Discoverability is admittedly not great, but this feature easily coexists with window borders. Microsoft could keep the thick borders by default and also implement this.
“Don’t do it the way everyone else does it or the way you’ve always done it; our designer decided to make that hard for <inscrutable reasons not related to delighting existing users> and instead we offer this hard-to-discover non-idiomatic workaround.
“You will grow to love the workaround because it doesn’t hurt nearly as much, kind of like stopping after you’ve been hitting yourself in the head with a hammer.”
There's a difference between a customer who plans on selling a billion units, and some company like this game controller design team that clearly making some crappy kids toys. I think the OP twitter person just learned a valuable lesson.
I've been a designer for about a long time. I've done many dumb things, but never that. I have definitely had the irrational thought that "users will think I'm a bad designer if we ship this product", but that's usually because it was a bad product that shouldn't have been shipped.
I've never gotten the sense that my peers do this either. I have definitely gotten the sense that some designers want their designs to impress Apple, and Steve Jobs specifically.
Everybody wants to impress their peers to some degree, but to be clear, that usually translates into just doing a really good job. I've never had another designer show me something and say "Nobody else will appreciate this, but I thought you'd like it".
It was applied to a lot more than that. Even the F11 6er still had it.
What is interesting, though, is how much flame surfacing started to flood car designs after Bangle's BMWs had been out for a few years. I'm not arguing that the E65 looked good, but he certainly made an impact on all of car design, seemingly overnight.
Go on to learn what Bauhaus architecture looks like, how that led to modernism, Frank Lloyd Wright, then Mies Van Der Rohe, then brutalism, and finally Frank Gehry in ridiculous mode.[2] Understand why this didn't result in liveable cities.
Then you can talk to designers.
[1] kare.com
[2] https://en.wikipedia.org/wiki/File:Ray_and_Maria_Stata_Cente...
This is so obvious from everyday life that someone really ought to write a paper about it and have it published in Nature. Because it will certainly be published. It will probably become the most cited paper of all times, just because of all the pent-up pissed-offery that everyone carries in them, going around in a world where nothing. ever. fucking. works! dammit.
Pfh. Sorry for the vent. This is what impractical design does to people.
The reasons it's the worst design guidance out there is that it teaches: 1) that engineers should NOT be entrusted/involved with user design. 2) to conceptualize users as blissful idiots. 3) that obscuring errors in a system is actually a good thing.
All of these are things that I see the lasting effects of in products I use in every day. Don't get me wrong, I bought this with the best of intentions. It was the highest rated design book on Amazon on the topic. Reading through was an extremely painful exercise in essentially coming to realize that everything I see broken in the engineering teams I work with and products I use is actually codified in words.
Take this excerpt as an example -- p. 65: "Eliminate all error messages from electronic and computer systems. Instead, provide help and insight."
Are you f'ing kidding me? You cannot seriously think this is right. For the longest of time I wondered why I hated using Apple products. It turns out Norman was schooled at Apple and they absolutely practice this teaching. I bought an early-day iPod touch circa 2007 and proceeded on to loading all my music on it. A few upgrades later, some albums disappeared without any error. I searched and searched and searched, trying to describe my problem in a number of different ways on Google to see what I could find. And all I could find were semi-answers. But you know what? If it had actually displayed "error 0xFHJAD1234" when failing during the upgrade, I could've actually google that. And I'll spare you the story of what I had to do when the family Mac's HDD failed ...
So this is what we actually get, devices that SILENTLY fail, designed for people who are assumed not to know what they're doing and who likely do NOT have any technically-literate person around them that could assist them if they could in fact identify the problem.
Worse, this philosophy probably hints at a level of intellectual ignorance/arrogance that is quite fascinating. You see, in my line of work I have to be able to walk engineers I work with from the gate-level silicon all the way up to cloud and AI. Without any false modesty, I think I have a fairly good idea of how modern computer architecture works. Yet, I will be the first to tell you that I have no idea how any of these systems work in full. In fact, no one does. Not the best of kernel developers nor the best of chip designers. To claim that you can somehow inventory every possible problem in a computerized system and then provide a graceful exit from that or useful info is either ignorant or arrogant or both. But since you've eliminated engineers from the design loop (see #1 above), no one will tell you you're full of it.
So yeah, no. Do read Norman's book. But look at it as everything you should NOT be doing.
It sounds like your complaint isn't with the advice from the book, but rather that Apple didn't follow it. They did hide away the error messages, but they didn't provide help and insight.
If they had done so, you'd have (presumably) been told in the software that there was a problem with certain albums, and been walked through a flow that would fix said problem. Or maybe a link to a support page that'd do the same. This, if it was done, would have been much better than showing "error 0xFHJAD1234".
Thus I think it's fairly good advice, as long as you read the "instead" as a critical instruction. You should eliminate opaque error messages so long as you can provide help and insight for those cases you have removed.
So not only are systems increasingly complex, but those maintaining them at any given point in time likely don't understand them nearly enough to accomplish what you suggest: understand all potential outcomes and provide useful error recovery. Therefore, the "provide help and guidance" advice provided by Norman effectively becomes "silent failure" for those cases not anticipated by the current-day release team. And since anticipating all outcomes is by definition impossible, Norman's advice is awful.
Yes, that's why I said that I read the advice as saying it should only be followed when you can provide help and insight. You don't have to do it in an all-or-nothing way.
You can have common problems with a helpful flow that tells the user how to fix them. You can have less-common problems which just show an error code. Removing the error without providing help is the problem here, and is the thing that I feel goes against the book's advice.
The best you could do is report it to Apple (or whoever). But, the error alone doesn't achieve anything. You would have been better off with "Something broke, click here to send log to Apple." (which is roughly what the book recommends - replace the error with something actionable).
It turns out that if you actually do take the time to report the error, no matter how obscure it is, someone in a dark corner of the internet will spend enough time on it that they'll even see fit to document it for others to help their own selves out. And given that, per my initial post, nobody is smart enough to understand how the computer system/software they're creating is going to work under all circumstances, providing error verbosity is sometimes the most empowering thing you can give to your users.
However, I suspect this has also made it easier to write and (kinda sorta) maintain larger and more complex systems. Like the phenomenon where you widen a road to ease traffic congestion and it works for a while but then encourages more traffic. The reduced cost of creating and maintaining complexity may have encouraged it overall.
- - - -
As an aside, can I quote you, like, on my blog?
> in my line of work I have to be able to walk engineers I work with from the gate-level silicon all the way up to cloud and AI. Without any false modesty, I think I have a fairly good idea of how modern computer architecture works. Yet, I will be the first to tell you that I have no idea how any of these systems work in full. In fact, no one does. Not the best of kernel developers nor the best of chip designers. To claim that you can somehow inventory every possible problem in a computerized system and then provide a graceful exit from that or useful info is either ignorant or arrogant or both.
I don't mind the quoting, but, just in terms of mental hygiene, I generally dislike posting unequivocal opinions on the internet ... because it's often come back to bite me ;) I'm especially skeptical of my own writing when I use labels such as "ignorant" or "arrogant". So long as you understand that I don't take myself very seriously, I mostly stand by that quote you mention.
Giving a user an error ID gives them a partially-if-not-completely unique identifier that they can then use to find other people struggling with the same error and possible solutions.
This is infinitely better than having no error message and having to try to type different permutations of your symptoms into Google in order to try to win search engine bingo.
Sure, if Apple quickly and responsively fixed issues, then you wouldn't need the error code. But, they don't! The problem is that "hide the error message and replace with actionable advice" only works in the idealistic case where the vendor will quickly fix the issue and/or the advice consistently fixes the problem. But, they don't, and it doesn't - the advice doesn't work unless implemented flawlessly - it's not robust.
The robust approach that will actually survive contact with the real world is "include an error message and ID code - even if you have to hide it behind a "more details" button".
Bonus points to the mythical developer that includes a google-this button on the error display.
In my experience sending crash data to software companies has never resulted in my satisfaction, nor has using some 'wizard' to diagnose an error. As alluded to above these systems are just not smart enough to actually know what's wrong. If the original programmers could create a fully automated flow to repair all error conditions then why bother notifying the user at all?
Sending them to large software companies does not have the same outcome. But then, that's certainly not a flaw with the format.
If they're not tech-savvy (and you would never talk to Apple store employees besides at the checkout counter if you were), they likely won't be able to reproduce the error anyway. They'd fumble around trying to explain what went wrong, and the employee would run through multiple scenarios trying to reproduce it.
> If they had done so, you'd have (presumably) been told in the software that there was a problem with certain albums, and been walked through a flow that would fix said problem. Or maybe a link to a support page that'd do the same. This, if it was done, would have been much better than showing "error 0xFHJAD1234".
The parent post addresses this: there are so many ways for for the system to fail that it’s not feasible to “inventory every possible problem in a computerized system and then provide a graceful exit”. So where an error is not gracefully handled, the system should allow for a human to see the error and solve the problem, rather than throw away the error.
>> Without any false modesty, I think I have a fairly good idea of how modern computer architecture works. Yet, I will be the first to tell you that I have no idea how any of these systems work in full. In fact, no one does. Not the best of kernel developers nor the best of chip designers. To claim that you can somehow inventory every possible problem in a computerized system and then provide a graceful exit from that or useful info is either ignorant or arrogant or both. But since you've eliminated engineers from the design loop (see #1 above), no one will tell you you're full of it.
Yes, that's what I said as well. It's why I said the problem was that Apple didn't follow the book's advice, insofar as they removed error messages without providing help for them... and so it's unfair for the grandparent-comment to blame the book when its rule wasn't followed.
If it's not possible to always give help and advice, sometimes you have to give an error message and unless the book mentions that then it is definitely fair to blame the book.
Since it says to hide error messages and instead provide help, anyone who hides error messages and doesn't provide help can't really be said to be following the book's advice. It feels wrong to me to blame the book for not, in every bit of advice it offers, saying "don't half-ass this incorrectly". That seems inherent to me.
That way, when the help and insight inevitably fails to solve the problem at some point, the user can still dump the error code into Google to see if other people have run into/solved the problem.
First off, Norman wasn't schooled at Apple, he basically created the HCI guidelines with Nielsen. Apple's design guidelines have long been based o years what Norman worked on so you have it the other way around.
In no part of the book does it say that engineers shouldn't be part of the design process. What the book advocates is that if engineers are entrusted with design, they should at least understand the users and design their products based on their needs and for the users. Anyone can be a designer, anyone can apply design thinking. No one is excluded.
And your comment about error codes, I would always say that an error message is more useful than an error code. You say that you would Google the code, wouldn't it be more useful if you could just read the actual error code and figure out what went wrong without habing to look it up first somewhere else?
What Norman advocates for is to help the users help themselves, show the state of the application and don't hide things behind error codes.
Again, silently failing is definitely not recommended by the book or anyone really so if thats been your experience with Apple, it's not because they follow the teachings of Norman's book, it's because they _don't_ follow it.
So while Norman wasn't a neophyte when joining Apple, he clearly credits his experience at Apple for having heavily influenced the specific book we are discussing.
Norman certainly doesn't recommend silent failure verbatim. But what I'm saying is that this is the net effect of his recommendations.
And it gets much worse in "The Inmates are Running the Asylum", that says engineers shouldn't do design right in the title, on the most sensationalist way possible.
Later he toned down that message a lot. Nothing from his group will say that anymore, and you will get a clear message that making your engineers think about design is better than nobody thinking about it. It still carries a message that you should leave design to experts, but you won't find anything there saying that engineers can't be UX experts anymore.
If you know how the product works then you already have a mental model of interacting with it. That mental model is NOT the same as the user will have, and thus an engineer will think that something is obvious even though it is not obvious to the user.
The same goes with anyone else who is close to it during the development.
You need outside users to make it clear how people new to a product can understand and interact with it.
Unless you're an engineer who can forget everything they know about their own product, you shouldn't try to design everything yourself with no outside feedback.
That is the correct message, and the one every article about design should push. Neither the engineers nor the designers can design a product in a vacuum.
Those are both old books that do not deny this message, but focus on less relevant subjects, and have less than clear advice. We shouldn't recommend those books for people without previous knowledge on UX design, because they will be harmful.
Besides, given that Nielsen was himself a very important voice on the creation of the modern user-focused design, there is very likely a newer book from him to recommend instead (I stopped reading his books and started reading his papers at the time of the change, so I don't know one).
I don't think you read the book. I read the book in 2000/2001. The emphasis was on helping the user.
Not a single line in that entire book, IIRC, advocates silently failing. I have no idea how you came up with that interpretation, when almost every single example in that book highlights "silent failure" as something to avoid.
https://www.youtube.com/watch?v=5GCPQxJttf0
Don Hopkins and Donald Norman at IBM Almaden's "New Paradigms for Using Computers" workshop
Talks by Don Hopkins and Donald Norman at IBM Almaden's "New Paradigms for Using Computers" workshop. Organized and introduced by Ted Selker. Talks and demonstrations by Don Hopkins and Don Norman.
Norman: "And then when we saw SimCity, we saw how the pop-up menu that they were doing used pie menus, made it very easy to quickly select the various tools we needed to add to the streets and bulldoze out fires, and change the voting laws, etc. Somehow I thought this was a brilliant solution to the wrong problems. Yes it was much easier to now to plug in little segments of city or put wires in or bulldoze out the fires. But why were fires there in the first place? Along the way, we had a nuclear meltdown. He said "Oops! Nuclear meltdown!" and went merrily on his way."
Hopkins: "Linear menus caused the meltdown. But the round menus put the fires out."
Norman: "What caused the meltdown?"
Hopkins: "It was the linear menus."
Norman: "The linear menus?"
Hopkins: "The traditional pull down menus caused the meltdown."
Norman: "Don't you think a major cause of the meltdown was having a nuclear power plant in the middle of the city?"
(laughter)
Hopkins: "The good thing about the pie menus is that they make it really easy to build a city really fast without thinking about it."
(laughter)
Hopkins: "Don't laugh! I've been living in Northern Virginia!"
Normal: "Ok. Isn't the whole point of SimCity how you think? The whole point of SimCity is that you learn the various complexities of controlling a city."
One solution I try is to redesign the cause of the error out of the system. For example, compilers often have maximum quantities of language constructs that are supported, like the maximum length of a string literal. Then, when the length is exceeded, an error message is concocted and generated, then error recovery has to be done, then the compiler has to not generate an object file, etc.
I don't know what other compilers do, but one day I realized that it was less work in the compiler to not have a limit, but to keep enlarging the string literal buffer. There was only one limit left on all these things, that was globally running out of memory. Globally running out of memory is a fatal error for compilers, and so error recovery isn't necessary. Just print a message and exit.
This works great. Large numbers of errors just go away, like "line length too long", "string literal too long", "too many cases in switch statement", "too many symbols", etc.
There are, of course, still some limits, like the object file formats often have hard limits, and of course you don't want to overflow the program stack.
30 years ago I had to live with problematic memory limits and made some design decisions that over time I would come to hate because I had to shoehorn data into EMS memory banks. Data objects ended up sliced and diced into separate arrays, never did they point to the relevant things because such pointers would always have been into a different bank and the only possible allocation was the whole bank.
When it’s failing is when it comes to the forefront. Crappy touchscreens seem to be a major component in lots of them, had an experience with an elevator “system” with a touchscreen that was mildly annoying.
From cars and elevators to software. Anything that interaction designers touched is broken.
Kind of like Yogi Bera says "nobody goes there anymore, it's too crowded".
I'm two months (at least) without data on my phone. I didn't need to use it and I'm a lazy ass so I didn't try to fix it. When I call them, a kind youg man on the other end of the chat says he's reset my service and within a couple of hours I'll definitely have data.
In a couple of hours I have no data, but now I also have no service: "emergency calls only". Result!
Of course I can't log into my account to see what's up, because it's trying to send me a TFA code. It can't send it to my mobile, since it's dead, but it sends it to my landline which is in the UK. I'm in Greece now.
I try to get help by clicking on a link in the page where I fail to login. I'm taken to a page with an "AI assistant" called TOBi. The page has the assistant's icon, smiling at me encouragingly and a wheel that's spinning, spinning, spinning... and spinning.
I think I pass out from all the spinning and all the screaming and pulling at my pigtails. I finally realise I can actuall call a number to ask for help. Fortunately my friend still has service.
I have to blow a few raspberries down the line to convince the (other) chatbot that it can't understand my speech, but I'm finally connected to a human being.
The kind young woman on the other end of the line tells me she has reset my service and I will definitely have my service back in 24 to 48 hours.
Three days later I definitely don't have service. I try to log in to my university email to pick up some work I need to do, but it asks me for a TFA code. That it sends to my mobile.
I call Vodafone for a third time. This time, I'm put through to a technician. He asks me to take my sim out of my phone and read the S/N off it "to make sure it's the right sim".
Hang on. To do what?
I put the sim back in the phone and I have service. Ah. So that's what he wanted me to do. He could have said so.
And I should have thought of it.
So, nothing works. Not even my goddamn brain works anymore. If that was a dumb take, that's because I'm dumb and useless and so is everyone else, it seems. Except for the guy who doesn't know how to tell you "did you try taking it out and putting it back in again?".
Because that's the only thing that works anymore.
Then, silly me, I remembered the good old DOS days. I did a cold reboot of the phone, and voila! people could hear me.
At the very least it's terrible UX design though.
> buys the cheapest controller one could find
> "Oh my god it's all cheap components/design!"
> buys a transparent one
> "Oh my god I can see the charger indication on literally the entire surface area which triggers my ADHD!"
This line is sarcastic, the whole thing is somewhat tounge in cheek.
> Hardware / software necromancer, collector of Weird Stuff, maker of Death Generators. (they/them)
Please find a more useful hobby than this worthless language policing.
I see someone drank the Kool-Aid.
This statement was brought to you by Kraft Heinz.
If the only thing stopping you from being rude would be the police literally dragging you away from your computer, yes, by all means, try to prove a point by being intentionally rude just because they can't stop you.
Man buys cheapest product, is surprised by its lack of quality.
If one of their threads is interesting enough to show up on HN isn't it also likely to have been retweeted, shared around elsewhere too?
They certainly pop up in my feed constantly due to RT's from other popular tech people on twitter
"I HATE GETTING POSTED ON HACKER NEWS"
"Smart" remotes are a different picture. audibly sighs and walks off the stage
I'd rather have Bluetooth use infinitesimal amount of my iPhone/Macbook battery than them being lost/stolen which could be potentially tracked when off with "soft off".
But besides some use cases like this, I agree with the post.
Why not? I mean obviously they wouldn't be findable if they were off, but that's a feature. Apple could add a physical switch and leave everything else the same. Your devices would function as they do now in "On" mode, but you'd also be able to turn the off properly. Isn't that better?
I know there are other ways to do it (a sophisticated theif might just put it in a Faraday Cage), but Find My would just work in its current form otherwise, and putting a physical switch that physically switches it off kills the whole purpose.
(Edit: if you disagree, please tell me why. With proper design, it's possible for every button on this type of controller to act effectively as an on button, and I even cited a working, shipping example.)
The twitter thread complains about controllers which empty their battery between uses because they keep trying to connect; but it's not something that's mandatory: an at-rest device can stay at rest without user interaction. I've used hardware on which the battery was still usable after years without use; it does requires very careful design, though.