Hacker News new | past | comments | ask | show | jobs | submit login
The Worst Interface Ever: The Self-Destruct Switch (2004) (asktog.com)
114 points by waterhouse on Jan 2, 2019 | hide | past | favorite | 70 comments



The follow-up post is a must read: in the panic ensuing after a failed switch flip, they made a series of disastrous mistakes that almost caused a fatal road accident.

The animality of the human mind is fascinating, all it takes is a small change in the perception of external threats to drive people utterly bonkers. Puts all kinds of things from wars to conjugal murder into perspective - what is utterly nonsensical to a calm and rational individual may appear as the only way to go for someone who is distressed.


Only one thing stands out: "I was certainly ill-equipped."

So far in these posts everything has been blamed --engineering decisions, the psychology of panic -- except the mindset of the adventurers.

"Fight and flight" is not a normal way to describe a reaction to an ordinary oil-spill, even in a slightly less than extremely convenient location when you have the comfort and convenience of a complete house, which is even mobile by itself, at your disposition.

I think some psychological factors are left unmentioned here, with hints of spousal disagreement, grudges, an unwillingness to admit failure, and a desire to divert blame to external factors (the posts in question being an example by themselves).

There was definitely a chain of ill-considered decisions that almost lead to a disastrous outcome, but it seems like a lot of the actions were outside of the realm of the normal default choices, making me suspect there were unmentioned factors at play.

Either that or the adventurers were hopelessly unsuited for their chosen adventure at that time.

"The animality of the human mind" would have significance in describing high-pressure and continuously mounting situations where in retrospect some baffling choices can be explained by instinctive choices that turned out to be counter intuitive.

I guess:

- "I don't want to deal with this right now!",

- "Fine, I don't need a navigator anyway!"

could fit the description, but seeing how it could easily have been:

- "I don't want to deal with this right now!"

- "Maybe you're right. I'm sorry. I'll put on the kettle and make some tea. Then I'll get online and see what I can drum up."

the story really ends up having a "Why the stupid things that happened weren't really my fault" tone.


Where is that post?



Thank you!



Thanks!


Reminds me of when PCs came with a voltage switch. With one flick of a switch... goodbye PSU. Of course the difference is it only needed to be set once.


Except when you move the PC. About 15 years ago I had an awesome demo all set up and ready to go on one of those cube PCs. Only problem was I did the setup and testing in the US, and the demo itself was in Germany. I came prepared with the correct power cable for Germany, but of course I forgot to switch the PSU to 240V and there went my awesome demo.


I have an old external HDD with a switch on the back labelled "NTFS / Fat32". Flipping the switch will format the drive.

...as one would expect.


I blew up a PSU like that a few months ago after moving to Europe. Didn't realize it wasn't auto-switching. It actually blows (loudly, and with smoke) the moment you plug it in.


If you're especially lucky it'll catch fire with the fan running. I've seen that happen to someone once, panic would be an understatement.


I had one of those multi-power adapters for a pair of infrared headphones when I was about 14 years old. You had to get the plug the right way around. Plugged it in, and BANG, the headphones were toast.


Except when a curious child flips the switch when the PC is running.


I can confirm that a curious 10 year was indeed responsible for the loss of at least one PSU. At least he learned something.


Ha, I can't count the number of times I've said to my kids some version of, "Don't push buttons if you don't know what they do!" I've mostly drilled it out of them by 6 and 4, but it was a process.

Of course the new manifestation of childhood curiosity was my six year old daughter testing out the piece of sandpaper that came with a craft set she got for Christmas... on our kitchen table. (Fortunately it was already a bit "rustic".)

Can't blame them though—I clearly remember at that age putting the garden hose into the dryer vent, with no clue what would happen. Dryer was in the basement too. I recall being mostly bemused by my parents' reactions. (It was on for quite a while too—I think it flooded more than one room.)


"Don't push buttons if you don't know what they do!"

That statement makes me a little sad, it's pretty much the exact opposite of the way I was brought up and I'm pretty sure I wouldn't be where I am without the freedom of curiosity I was given at home. The general rule was if it's not going to kill me or maim me then I could have it.


Uh, I'm sad for people who think you have to break things or press random buttons in order to figure out what it does. Like the people in stores who press buttons and then wonder why the police is knocking with a full SWAT team. Documentation exists. The resources exist to inform yourself if you're actually curious. Emphasize the need to satiate that curiosity, but there are ways to do it without being an idiot.


I generally agree with this, which makes the existence of buttons that do kill (you, someone else, or an expensive device) a problem.

I did have to teach my toddler not to press the power button of the PC while I was working. What kind of idiot puts a led in a power button?! Of course children will push that. Clearly PCs need to be put as high up as possible, to keep them out of reach of children.


Gave up on that and taped the button over with black tape. No longer lit up, no longer pushable, not interesting to toddler.


Interesting, got a couple responses in this vein. I can certainly see how my initial post would give the impression of not allowing exploration. I certainly encourage them to explore the world, learn about things, and figure out what things do. They can of course also ask what things do. I don't think you need to have wide open freedom to try anything to explore though; part of learning to navigate the world is to develop an understanding of what sort of experimentation is safe and prudent, and in what cases learning should be done first. IE: experiment with sandpaper on a fallen branch or scrap lumber, great. Experiment on the kitchen table, not great; potentially expensive.

Likewise, regarding button pushing specifically—say a kid happens to get their hands on a power tool—I don't want them experimenting with it to see how it works; I want them to ask me, and I'll show them how it's used safely. (And if using it safely is within their capabilities, I'll let them try it, or help them to try it.) There are many other similar cases where blindly pushing a button or flicking a switch could be at best inconvenient or expensive, and at worst dangerous. (And of course many more where it would be innocuous.) By teaching kids to differentiate between these, and not to take actions without thinking through the consequences, you can then give them more freedom to explore, confident that they are unlikely to take dangerous or destructive actions out of ignorance.

So yeah, I'm not teaching my kids to never experiment with anything. I'm teaching them to first think about what could happen. In many cases it's obvious, even to a four year old, that experimenting with something couldn't hurt anyone, and is unlikely to damage anything. (And in your example, "if it's not going to kill or maim me, then I could have it," it's implied that someone has already vetted "it" for you, in which case again you know it's safe.) In other cases they might not be sure, in which case it's better to ask (or for older kids, study) than to just blindly try it.


Imo, it's a bit unfortunate to drill the curiousity out of kids. I'm very thankful for being allowed to mess around unsupervised on my father's work computer as a 7 y/o. Pressed a lot of stuff I didn't know what it did and learned in the process.


How does this drive out the curiosity? If anything it encourages it: kid should find out that this button does before pushing it. Though to be fair I did my share of button pushing at 5 years old. It was a typewriter at my mother’s work and I taught her a couple of new tricks after finding out what buttons do. On the other hand I struggle to understand people who refuse to read manuals and then struggle for months or even years with some basic functions of the product, when spending two minutes reading the page four of the manual would have saved lot’s of time and frustration.


This is not realistic for childhood curiosity. A 4 year old is curious about things like what happens when I bang a pot on the ground. It's unreasonable to expect the 4 year old to read a manual to inform them it makes a loud sound, and also unreasonable to expect them to know the difference ahead of time the difference in scale of danger between banging pots and flipping a switch. In fact, for most areas of curiosity no such manual exists. Experimentation is the primary mechanism of learning.


It even has a name - heuristic play.

Here are early years examples, but it applies to all children.

http://www.kathybrodie.com/articles/heuristic-play-a-simple-...

https://schoolhouse-daycare.co.uk/blog/heuristic-play/


>kid should find out that this button does before pushing it.

I bet you're the same parent that got an infant a toy covered in giant buttons that light up and make noise. We as a society design button pushing as play.


When I was seven I got a (used) Commodore 64, and learned a ton experimenting with it. Started me on a path to eventually running a small tech company. So I hear you there. My oldest is 6 now, and I definitely want to get her started on similar things.

I certainly won't be giving her unsupervised access to my work computer though. Instead I'll start her on an old laptop with some flavour of Linux, and no internet connection. I'll also take an LVM image, so she can tinker to her heart's content and we can easily revert if it goes wrong. I expect she'll learn far more than she could using my work machine, where there are all kinds of things that could she'd have to be careful with, and that could cost me hours of work (or even worse, potentially, with email access and such).

Separately I'm also introducing her to minecraft, which is fantastic for exploration. Of course it needs an internet connected computer, but there's more supervision for that one when she's outside of the game.

It's funny, thinking back to getting that Commodore 64, I remember the best piece of tech advice I ever got, again when I was 7-8 or so. My parents' friend, who had previously owned the computer (my parents, like most people in the 80s, didn't have a computer) was teaching me how to use the word processor. He taught me how to create a new document, I think, and I was going over the steps, like "press alt, press f, press n..." And he says, no, no... Don't just memorize the steps—understand what they do, and you won't have to remember them. Then he goes through explaining how alt tells the computer you're going to start a command, f refers to the file menu, which has commands relating to... etc. Stuck with me like nothing else, and that attitude was hugely beneficial in not just computer science, but all aspects of life.

It's a bit tangential to the "don't push buttons" thing, but only somewhat. In both case the point is that you want to not just do things, but think through them and understand them. In my mind, that's a great lesson for kids.


I teach my children the opposite. Perhaps I've worked too much tech support.


When I was a curious ~10 year old, yes, I too did!


This is why when I got my RV I also got a Jeep wrangler to go along with. Its one of the few vehicles left that can be towed "4 down" without significant modification.

It has no steering wheel lock and you simply put the 4wd transfer case in neutral, no transmission lubrication issues.

Mopar even sells a wiring harness that ties into the rear lights directly to properly operate the brake lights from the RV.

The jeep is also a great vehicle for sight seeing especially in remote areas.


Slightly off topic, but in the current context of environmental awareness, let me get this straight: They're travelling across the continent in what I can only describe as the vehicle from Meet the Parents, and for what I presume are short local trips, they bring along... a Lexus RX !?

This is the standard European solution to the same problem: https://www.watling-towbars.co.uk/motorbike_scooter_carrier....


I have an RV and tow a 4 door Jeep behind it, the difference in gas mileage towing the jeep vs not is maybe 1 mpg(6.5 mpg vs 7.5 mpg).

I have two kids, towing just a scooter would not be that useful. The jeep works really well for sight seeing, including off-roading while at the destination.

A large RV is not the efficient way to get around regardless of towing a smaller vehicle, but it is a great way to see the country. I am somewhat torn about the environmental aspect of it, I am not sure how clear cut it is vs air travel and hotels and related infrastructure.


For those outside the US: 6.5 us mpg = 36 l/100km ... it's funny how different fuel prices have given radically different local optimisations.


I've seen more tagalong cars in Europe than I have motorcycles, but usually very small ones (Nissan Micra etc). More common is bicycles - you are after all on holiday.


I'm not sure "buy another vehicle" is necessarily the environmentally conscious approach, nor is a scooter likely to be the best approach in many areas of the US you might vacation to - for both distance and weather reasons.


Sure, but an RX? That's the biggest Lexus you can buy in Europe..


In the US, an RX is in the "mid-size crossover" class. Things are bigger here, including the RVs themselves. Check out the pics, it's the size of a small apartment: https://asktog.com/misc/countryCoach.html


This article is from 2004. The RX was a smaller vehicle 15 or so years ago. At this point, what Lexus is selling under the same nameplate is considerably different.


You would be amazed at what Americans tow behind their RV. Often you get a pickup truck being towed with a couple of dirt-motor-bikes in the pickup truck with a trailer on the pickup truck to carry a boat. Then the RV is as big as a European coach, with untold junk on the top to further impair the aerodynamics.

The joke of this is that all of these toys take a lot of time to get out and pack away. So in the National Parks where people join the dots between places like Bryce and Zion the humble bicycle becomes quite competitive. On the bicycle you can rock up at a camp ground at 7 p.m. with no booking needed. Then, by 8 p.m. one can be fed, watered, washed, clothes washed, stove washed, tent ready and tired enough for the best sleep ever. Food might be basic but there is a lot you can do with a simple camp stove.

Meanwhile, the RV crew have arrived five hours earlier, having set up their BBQ, chairs, TVs and what-not they are not having much time left. They also have to have arguments about plates and other petty things.

Then, the following morning, when on a bike there isn't much to pack away if it all fits into two pannier bags, even with a leisurely lie in the road can be hit many hours before the RV crew have got their first breakfast prepared.

On the way the bike also pays dividends. Say you need to shop for groceries then you can park right next to the store entrance with no time wasted walking half a mile across a car park. Same with attractions en-route, there is no need to dismount. Meanwhile the RV crew are having to park a considerable distance away, have an argument or two about maps and fritter away the day doing RV things. The 'panoramic sunroof' is also a lot better on the bike, one's view of the world is not framed by a windscreen.

Admittedly cycling is not for everyone, particularly if carrying a tent, sleeping bag and stove but, when you are covering the same distances as the RV crew in the same time and doing a better job of seeing the sights rather than picnic tables, you do have to wonder what it is all about. The other thing the RV crew don't get is meeting the natives, American hospitality is the best in the world, not that you would believe it from watching TV, but it is. The RV isolates the traveller from that, the bicycle opens it up. People you meet on the way do invite you onto their boats, show you their gun collections, even put you up in the RV that is parked outside their house. So all these things American people take with them on their RVs are kind of there for you if you go by bicycle through the splendid US scenery.



Another anecdote about an arrogant technician and UI's. When Ebay was a young company, I decided to give it a try. There was an auction status notice that was supposed to appear on the screen that I missed. My tech support session went something like this:

Me: I didn't know that such and such occured because the status messages appear on the right of the screen, and my monitor is too small to show the right-hand side.

Technician: You are correct. I suggest you adjust your Windows DPI setting so that more fits on the screen.

Me: I already tried that, but it makes the characters too small in general. It's hard on the eyes. I'm using a fairly old monitor.

Technician: I suggest you buy a newer and bigger monitor then.

Me: Other users may have the same problem; not everybody has new monitors. Can't you move the status messages so they are not on the far right?

Technician: Sorry, we can't do that. Please buy a new monitor if you want to use our site.

I stopped using Ebay for a while. At least back then you could easily talk to an actual technician who knew the system. Now you'd probably have to go through middle-ware AI and script-reading minimum-wagers who may be diplomatic yet useless.


Would seem to me pretty simple to interlock the switch with the starter so that it would be impossible to start the car with the switch in the wrong position.


That's Tog's point. It could have been designed to make it much easier and safer to use. The people who made it just didn't think those kinds of considerations were worth putting time into.

> Calling the company was of no help. The engineer who answered responded that nothing was wrong with the design of the switch that extremely careful operation would not overcome. He’d been using it for months with no problem.

How this analogizes to software is left as an exercise for the reader.


If a lawyer would look at the contract, he would say we fulfilled the requirements of the contract.

If a customer would look at the contract, he would say "Never again, and will warn everyone i meet".


I don't know about more recent vehicles, but on older cars it would have been as simple as man-in-the-middling the ignition wire which runs to the solenoid. Switch is in tow, the key won't start the car but things like accessories will still work.


That still wouldn't stop you from towing the car with the switch in the wrong position, which also destroys the transmission.


Correct me if I am wrong but with the switch off (i.e. in position for driving) the lubrication does not occur same as if the modification had not been added. So wouldn't it still be an improvement as functionality is just as good as before and 50% of the failure modes are eliminated? Obviously you want to find a solution for that other 50% but at least you've made an improvement without any regressions.


Yeah, a 50% solution is definitely better than nothing (but you'd still need to be very careful since a mistake could still fry your transmission).


Towing will destroy the transmission in most automatic transmission cars. That is why most of them specify flatbed towing in the owner's manual.


1. If a car has front-wheel drive, why would towing it with the rear wheels on the ground destroy the transmission? The rear wheels aren't even connected to the transmission. (I can see that it might be a problem for four-wheel-drive vehicles.)

2. The article specifically states that Lexus originally thought that towing this car would be OK:

"Lexus didn’t know they had a problem when they released the car, so they put a four year warranty on the transmission, specifically certifying it for towing."



To read the article:

    document.body.style.maxWidth = '800px'
    document.body.style.margin = '0 auto'
    document.body.style.font = '14pt Georgia'


Or reader mode in Firefox. But agree, there's something hilarious about a guy complaining about bad design interactions - meanwhile his website is almost unreadable on a desktop PC, with whole paragraphs stretching the length of the window.


I'm willing to cut the website some slack. It's 15 years old and still online. It's also only 35KB and loads fast. By comparison to the cesspool that is any modern monetized website, this was a joy to read.


I have to agree. Having managed to read both articles, which loaded instantly, while on a high-speed, cross-country train journey using mobile hotspot to my laptop, small is good, and even on a 4K, 15.6-inch screen, make the window smaller if you don't like the flow, or use reading mode. Or zoom in :D


Yeah, fair enough. In 2004, I think I was still running 1024x768, so this would have been less of a problem then, and newer articles have a better layout: https://asktog.com/atc/principles-of-interaction-design/ (2014)


Snake text (ie, copying existing learnings from newspaper column widths) was already a thing in 1996

> The column was the first example of what the Web magazine Salon (www.salon1999.com) dismissed as "snake text," meaning the story ran in one long, narrow column. While the reader has to constantly scroll through snake text, the narrow width is far easier to read than, say, oh-I-dunno, Salon's ripped-from-a-magazine page format.

https://www.wired.com/1996/11/web-dreams-2/


And three lines of CSS would make it readable on wide displays. Somehow not included in those 35KB.


I for one am kinda sick of websites that shove all their content into a narrow band in the middle of the page. What's the point in having all this complicated layout bullshit if it doesn't actually make use of the space it gets on a larger screen?


Human eyes are bad at tracking long lines of text. A general guide is 100 character line widths, maximum.

Unfortunately, multiple columns are seen as ugly.


>"meanwhile his website is almost unreadable on a desktop PC, with whole paragraphs stretching the length of the window"

That is not a fair point at all. Resize your window and the content flows with it. What better layout can you demand from a 15yo fast loading webpage that is still online?


So a key element in this story appears to be the role of fear in switching off the brain and reverting to old behaviour while skipping the checklist. And twice it seems to be caused by the human environment: middle America seems to be a hostile place?


Sometimes we get the interface we deserve.


> Calling the company was of no help. The engineer who answered responded that nothing was wrong with the design of the switch that extremely careful operation would not overcome. He’d been using it for months with no problem.

My blood pressure rose reading this. Absolutely infuriating response. I can't stand it when engineers fail to recognize how human factors relate to the systems they build. Makes me want to grab them by the ear like a child and put them in timeout until they come to their senses.


I recently re-watched "This is broken" and I think there's a subtle way things are broken related to "not my job" - called "I did my job". In any kind of non-trivial engineering, you have a spec to meet, and it is difficult to meet that spec under given constraints, so you don't go and look for more problems. You are happy that you did your job.

0. https://youtu.be/aNDiHSHYI_c?t=235


There could be something else at play in the call... they could have been scared that if a representative of Lexus admitted there was a problem with the design, they would be admitting liability.


He didn't call a representative of Lexus.


True. And TFA doesn't name the company:

> ... but it is a real, aftermarket device for Lexus RX-300 series cars being sold by a company in Florida.

However, parent's argument applies to "company in Florida".

Also, ironically, to the author of TFA. I'm guessing that it's not an accident that he doesn't name "company in Florida".


Another variation of "it works fine for me".


Just say, "It's a practical problem for us and if you don't fix it, I'll take our business elsewhere." If you can't take your business elsewhere, then you or your org dug itself a hole and is thus partly responsible. Never depend on one lone idiot. (Remember Jurassic Park 1?)




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: