Based on this question.

While I loved reading the sassy replies about caves and leopards, the larger debate seems to be whether or not we assume the user is irresponsible and I wonder, as a user experience designers, or interaction designers does putting in safeties in apps like this fall under our description?

My specific questions are as follows:

1) How do we factor in user irresponsibility into our design? Do we have to do that at all, or do we just accept that no matter what we design, users will always find some way to be irresponsible about it and thus don't even bother?

2) If we do design safeties for our applications - how do we enforce that? For a car-driving app the app probably shouldn't just lock up all together in case a passenger wants to use it. How do designers handle and tackle corner cases like this IF they are taking user irresponsibility into account.

A somewhat similar scenario that came up in a design class was whether there should be some way of detecting whether someone is smoking with a child in the room. While we spent a while discussing solutions to this problem, a question most students had was "Why would something like this NEED to be designed? Shouldn't people know not to do this?"

  • Your title and your "TL;DR" statement are asking two different questions. "How far do we go to protect users" and our "responsibility... to factor in user irresponsibility" are two very different things! Are you asking how the UX process (since it isn't just design), when done properly, protects the user, or if it is possible to "design out irresponsibility"? May 21, 2014 at 18:50
  • apologies, Tl;DR has been redacted. May 21, 2014 at 18:51
  • A related question: ux.stackexchange.com/questions/57055/…
    – CJF
    May 21, 2014 at 18:54
  • I'm still torn on the direction of the question: "protect" vs. "user irresponsibility". Are we talking about (1) adding a label to a hair dryer to avoid using it in the shower, or (2) designing the hair dryer to not work in the shower? May 21, 2014 at 18:55
  • the second one, the first seems like somewhat of a cop out/easy way out. May 21, 2014 at 19:03

7 Answers 7


User Protection vs. User Irresponsibility

The original question and several answers given show a gap between the understand of "protecting the user" and "avoiding user irresponsibility," so let's get that out of the way.

Protecting the user is of paramount importance within the UX process.

Avoiding user irresponsibility is impossible. Simple. Irresponsibility implies a conscious action against ones "better judgement."

A lot of little examples of this distinction could be nit-picked back and forth. A better example might actually be a very complex one... airplanes. Airplanes are designed to protect the user (i.e., the pilot) in a number of different ways, but no amount of design effort can prevent user irresponsibility on the part of the pilot (e.g., purposefully pushing the yolk full forward and hitting the plane in the ground).

Several actions in an airplane (and other systems) are purposefully made more difficult on purpose. This is to protect the user! Guarded buttons, which require the user to lift a guard before hitting a switch/button, are actions that are purposefully made more difficult so that (for example) you don't jettison all your fuel unless you really meant to.

We spend a great deal of effort to protect the user in aviation (and other fields) and to prevent them from making mistakes or errors.

User Experience is About "Easy"

No. It's about "error".

ISO 9241-2101 defines user experience as "a person's perceptions and responses that result from the use or anticipated use of a product, system or service". According to the ISO definition, user experience includes all the users' emotions, beliefs, preferences, perceptions, physical and psychological responses, behaviors and accomplishments that occur before, during and after use. The ISO also list three factors that influence user experience: system, user and the context of use. (Source: Wikipedia)

Notice that "easy" isn't in there. "Easy" is good, "easy" is an element, but it is the overall experience (the "anticipated result") that is the overall goal. Safety is a huge part of that.

The US Military has a Military Standard that focuses on Human Engineering (of which User Experience is related). MIL-STD-1472F, "Department of Defense Design Criteria Standard", discusses at length how systems should be designed to promote effective work flows and safety of the user. An excerpt of the Objective of MIL-STD-1472F reads:

Military systems, equipment and facilities shall provide work environments which foster effective procedures, work patterns, and personnel safety and health, and which minimize factors which degrade human performance or increase error.

Emphasis added by me.

The standard also has the following requirement:

4.8 Safety. Design shall reflect applicable system and personnel safety factors, including minimizing potential human error in the operation and maintenance of the system, particularly under the conditions of alert, battle stress, or other emergency or non-routine conditions. Design of non- military-unique workplaces and equipment shall conform to OSHA standards unless military applications require more stringent limits (e.g., maximum steady-state noise in personnel-occupied areas).

When something is "easy" the chance for human error is likely reduced. But sometimes something needs to be made hard in order to achieve the user experience you are looking for. We protect the user as a result, making it more difficult for them to commit an error.

Everyday Things

The term user experience was brought to wider knowledge by Donald Norman in the mid-1990s. He never intended the term "user experience" to be applied only to the affective aspects of usage... In an interview in 2007 Norman discusses the widespread use of the term "user experience" and its imprecise meaning as a consequence thereof. (Source: Wikipedia)

Indeed, let's look at one of Donald Norman's most famous books: The Design of Everyday Things.

We could quote so many portions of this book, but I'll call out a excerpt from the "User-Centered Design" chapter. In this example (Figure 7.4) a picture of a door is presented with the unlock handle all the way at the top -- making it rather inconvenient to open. Why?

7.4 A School Door, Deliberately Made Difficult to Use. The school is for handicapped children; the school officials did not want children to be able to go in and out of the school without adult supervision. The principles of usability espoused in POET can be followed in reverse to make difficult those tasks that ought to be difficult.

Emphasis added by me.

Indeed - "difficult" is sometimes good! Difficult tasks protect the user from doing something that they should not do, unless they really mean it.

"We're just designers... It's not our job!"

User Experience is a process. This design chain is commonly called "User Centered Design". A quick visit to your favorite search engines yields many results, with slightly different flows, but here is just one example:

enter image description here

(different examples call out different stages and names, but all to the same end goal)

If you fancy yourself "just a designer" then great. Design, and title yourself accordingly. Don't pretend to do "user experience".

There are elements to protect the user in every stage - except arguably the "design" stage.

  • Strategy: What is the product we are attempting to make? This sets everything in motion - wanting to "build a plane" guides your design process.
  • Analysis: User analysis, environmental analysis, task analysis, needs analysis. They are all designed around making sure the user can effectively do their job and to reduce the potential for error. Sometimes that means making something easy, sometimes it means making something hard. **A great deal of proper research should go into this step to make sure
  • Specification: Requirements are important to keep everyone on track and not decide what to do on their own; rather keeping what the research has shown us.
  • Design: This takes the analysis and specification and executes on it.
  • Evaluation: Does all our research and design hold up? Can the user effectively do their job, with minimal errors... meaning some tasks might be a little more difficult then others, so they aren't done accidentally.

Should I Take This Hair Dryer Into The Shower?

No, you shouldn't. So how do we design for that?

If it is possible to prevent the user from taking the hair dryer into the shower, or to prevent its operation in the shower, that is absolutely the goal of the user experience process. Other stakeholders may push back for various reasons (Engineers: really hard to do; Finance: really expensive to do) but the process should allow for that discussion to happen.

Sometimes the best we can do is just put a sticker on the hair dryer. For whatever reason the user experience process ran its course and that was decided what the best way to protect the user was. Perhaps the technology didn't exist to prevent the dryer from working in the shower; perhaps it would be too expensive to integrate (who wants a $300 hair dryer!?). But the process to prevent a very real potential for error was followed and the most appropriate solution was found.

  • Thank you so much for this! It called out a lot of misconceptions I had about what it means to be in UX and really helped me answer not only this question, but see a much bigger picture. I really appreciate all the time and thought you put into this. May 22, 2014 at 16:56
  • 2
    +1 "User Experience is a process". Good way to tell the difference between design and user Experience .
    – Misters
    Jun 5, 2014 at 12:54
  • Hair dryer needs a short power lead. Hair dryer can go in shower, just not while plugged in.
    – Josh
    Jun 6, 2014 at 3:25
  • @Josh, hopefully the home owner doesn't have a spare extension cord laying around. :) Jun 6, 2014 at 23:37

There's also another conversation to be had, because there's irresponsibility of the user "during" using our app; irresponsibility of the user "using" our app; or irresponsibility of the app itself.

That is, the user may be driving while using our app (irresponsible, but unrelated to the app), OR the user may be posting nude pictures with the app (irresponsible, using the app), OR the user could unintentionally transfer all of his money to some foreign account (irresponsibility of the app). These aren't all great examples, but I hope I've conveyed my point.

Addressing those, this question seems to be asking about the first situation--the user is being irresponsible, but it's no fault of my app. For example, in the driving question that was cited, my app can say "Sorry, can't do this while driving", but there's nothing that says the user won't just open a different app that doesn't have that restriction, so the user is choosing to be irresponsible, regardless of what app they're using. Now, whether that's on our conscience or not, may be worth considering.

I've seen some interesting approaches to tackle the second situation-- where the user is doing something irresponsible with my app. This would be like a texting app that won't let you text if you're drunk or a photography app that won't let you post pictures of nudity (via computer vision algorithm, not "I agree to the terms..")

The third situation seems to be the only one we can control as designers. We can make sure that our app is clear about what it's doing and that it won't do something irresponsible without the user being irresponsible. Certainly if you type in a random bank account number and click transfer all savings, my app might could warn you, but perhaps that's what you actually want to do.

Now to your question, it seems like its a personal decision, because irresponsibility is subjective. There are things that have been found to be bad (smoking around children, texting and driving, etc), but people still do them. That's the definition of irresponsibility. But somethings are more subjective and come down to moral disposition. Perhaps I don't want you to take pictures of women in underwear or bathing suits in my app, but you don't see that as wrong. Should my app still prevent that action? On the other hand, with the aforementioned driving app, if I disable input while driving (assuming there's some magic way to tell if you're driving), then it's not on my conscience if you wreck while using your phone, because it wasn't while using my app.

  • "Whether that's on our conscience or not, may be worth considering" is essentially the whole reason I asked this, thanks for expressing it so eloquently. In the example of designing a smoke detector that can detect cigarette smoke and children in a room - is that something that's actually needed? Would/Should it weigh on my conscience if I ignored that possibility? May 21, 2014 at 19:40

Regarding what users know and should do, it is a matter of needs and desires vs. awareness, e.g.

  • User may think that something like that can not happen to him/her

  • User may experience some urgency which causes him/her to overlook common sense

  • User may have to change plans while driving and not find somewhere to stop

For these reasons and others, counting on a user's common sense regarding safety might not be a good idea.

Regarding solutions, I think there are a few dimensions to this:

  • Legal responsibility and attempt to enlighten user - try to make sure user understands the dangers in a way which is easy to perceive and hard to ignore.

    E.g. warn user on first usage of app in a short sentences with large font with a next button per sentence and a timer to make sure user can't skip immediately.


download bmml source – Wireframes created with Balsamiq Mockups

  • Attempt to enable safe usage of app while driving via a different UX approach - try to enable as much as functionality as possible without looking at phone and preferably without touching phone too.

    E.g. Instead of/in-addition to using visual menus, use simple auditory menus e.g.:

    • On click anywhere on screen have app say "Say 'navigate' or slide down to navigate", "Say 'report' or slide up to report". (Talking doesn't require looking at or touching screen, simple up/down/left/right sliding doesn't require looking at phone to find buttons.)

    • On slide down, say "Say 'home' or slide up to go home", "Say 'work' or slide down to go to work, "Say 'other' or slide left to select another saved location", "Say 'search' or slide right to search for a location".

    • and etc...


download bmml source


The UX required to guide or enforce a more responsible or careful action in very likely, just by it's nature, to negatively impact on a user trying to achieve their goal.

In the case a nuclear missile launch the tedious protocol that requires two people and a locked safe, etc is a clearly reasonable imposition. One big red button on a desk would be much easier and more efficient, but too easy to have an accident.

Unless pursuing a very important goal users will abandon a task when there is an intentional or accidental "blockage" in their path (see data on Captcha's impact on conversions). Do be sure that the goal you support is worth the risk of intrusion from the 'failsafe'.


There's two issues here, in my opinion - which is based on my previous (dreaming of) riding leopards in caves ;-).

Either we talk about a clear statement of consequences, like pregnant smoking. The important point here is to be factual - this will be a point to debate. If an app (or web site, or any other product) can lead to damage, this must be conveyed to the user clearly. In these cases, UX has a responsibility to inform about the consequences in a clear, unambiguous way.

Or we talk about enforcing rules/laws, like undistracted driving. I value democracy, which in essence means all "our users" are electing the lawmakers, and as such are involved in making laws. I do not quite see the point in requiring UX to enforce the laws again. So in these cases, I think it's not right to patronize our users.

Regarding the first category, the interesting question comes up what is "factual". Tobacco companies have long lobbied against cancer research. Climate change is a current topic, where similar bending of facts happens, probably in both directions. So in my view, there is some caution required as to what you accept as factual while applying UX work in the first category.


Please forgive me for making this observation, but it almost sounds like you are trying to avoid making your own value judgement.

As some have mentioned, it's really a question of whether you don't want an irresponsible user's fiery (if Darwin award-worthy) death on your conscience, enough that you'd be okay with the (many) issues created by your decision to protect the user.

For example, if you decide to prevent users from using your app while driving by disabling the app when the GPS shows the phone is moving faster than 10 mph, you should be prepared for users who get angry because:

  1. They are a passenger in someone else's car or mass transit
  2. They are in an elevator
  3. They are irresponsible, and insist "it's a free country"
  4. The GPS on their phone is broken

In all cases, the user in question will probably stop using your app. The simple (but not easy) answer is that you must judge whether protecting irresponsible users is worth enough to you to have the losses that will result.

The problem can be removed, ever so slightly, from the moral sphere to the financial by considering the likelihood of being sued by either the irresponsible user (could happen, especially in the U.S.) or an "innocent" victim. But that's really a side issue, IMHO.


Have you considered to use Bluetooth 4.0 (BLE) distance functionality? Bluetooth 4.0 ios distance calculation

I am working on a similar solution, and one option would be, to use a OBDII dongle which is connected to the OBDII port. This uses BLE 4.0. So you can see, how close the iphone is next to the dongle, and test, if the driver is using the phone. I think there are already Bluetooth beacons around, which are capable of this.

But still it is not 100% safe, because user would learn to cheat (imagine a user leaning to the far right, so the distance is far enough to use the phone....)

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.