“Sit in the driver's seat,” he instructed me, “and now try to changethe radio's volume.”
I complied, fiddling with the impressive, and baffling, iDrivejoystick in the BMW 745i. Eventually frustrated, I gave up and askedfor help. “I keep forgetting,” he replied.
He was one of the engineers who designed the system.
The iDrive replaced some 200 controls in high-end Beemers, but wasso confusing that some dealers reputedly offered a one-week course fornew owners. Too many features meant that, for many drivers, none wereuseful.
Famously-bloated Microsoft Word offers so many features that manypeople find it impossible to do simple tasks. Some wags suggest thatonly 10% of its capabilities ever get used. It seems each new releasebrings truckloads of even more obscure operations. Yet it has been afine word processor for years. The only reason I can see for an upgradeis to follow file formats.
An elderly couple, great friends, recently told me of struggles withtheir cell phone, complaining about deep menu structures that madeusing the address book too complicated. Yet theirs is a simple modelwithout the seemingly-mandatory MP3 player, camera, video, TV, webconnection, and plethora of games.
Skipping down a couple of generations, my kids laugh that I haven'tprogrammed in different ring tones for each caller. I neither know how,nor have I any interest in developing an intimate relationship with aphone.
Now a company is exploiting some consumers' frustrations withoverly-complex phones. Jitterbugsells two models: one does little but make calls, with big buttons easyon older eyes. The other doesn't even dial calls. Press “911” or”operator.”
In either case a real person – an operator (remember those?) -answers and either invokes emergency services or completes your call.Instead of fumbling on a keypad at 70 MPH, or trying to find a clearfocus point through bifocals to locate tiny buttons, a human assistantdoes the dialing.
The user doesn't struggle with menus to set up the address book. Itcomes pre-programmed, from a list of numbers customers provide whenordering the phone.
Instead of cramming every possible feature into their product,Jitterbug has stripped features till the phone does nothing but actlike a phone.
While we engineers obey marketing's commands to add ever morefeatures in a mad race to keep up with the competition, one can't helpbut wonder if there's another sort of consumer who desires a clean,elegant and simple product, a product that does one thing well. Onewith a clear UI, with large buttons and no gimmicks.
Or, I wonder if Jitterbug will be able to resist the urge to add”just one cool feature.”
What do you think? Do we really need all of the features crammedinto every electronic goody? Or is it all a marketing ploy?
Jack G. Ganssle is a lecturer and consultant on embeddeddevelopment issues. He conducts seminars on embedded systems and helpscompanies with their embedded challenges. Contact him at . His website is
You said: “While we engineers obey marketing's commands to add ever more features in a mad race to keep up with the competition, one can't help but wonder if there's another sort of consumer who desires a clean, elegant and simple product, a product that does one thing well. One with a clear UI, with large buttons and no gimmicks.”
I say: “Both of those customers are typically the same customer”
– Ken Wada
We choose to live and operate in conspicuous consumption.
We choose quantity over quality.
We choose to complicate our lives.
I like this article because it demonstrates how important it is to get back to basics and meet the needs of the targeted and intended customer. Not (always) attempt to do everything a customer could ever imagine and then some, yet somehow only frustrate, bewilder, and even lose them in the process.
Simplicity Simplicity Simplicity.
There will always be a market for the “Convergence Gone Wild” types of products, but we also need to keep developing simple, reliable, and very affordable products that only do a few things, and nothing else, extremely well.
Well, I think what we are confusing over here is a BAD-UI versus feature-sets. When we add in new features , its basically for the consumer for whom the “need” can be developed . Adding a red -911 button on the phone , need not mean removing the MP3 player.
– Sachin Panemangalore
Adding features without adding complexity is a tricky business. For a great read on this topic, check out Alan Cooper's book, “The Inmates Are Running The Asylum”. Alan is the proclaimed “father of Visual Basic” and his book is an insightful and humorous read covering many over-engineered products. He makes a strong case for why engineers need to focus on human interaction design to make things simpler.
– Jan Liband
And where is the feature i've been longing for in cell phones, teleportation.
Just asking? 😉
– Aki Peltonen
Apple's iPod and upcoming iPhone are good examples of simple and intuitive user interfaces that have captured consumers. The iPhone has many features yet its user interface allows basic phone usage with ease. Any user interface that has a “user manual” or needs training is poor.
– Piyush Patel
I think Jitterbug has found a niche. While my high school kids love the bells and whistles, my parents have a disdain for the small buttons on current cell phones.
– Gene Field
Cell phone? What's that?
– Matt Staben
I work with a lot of older folks (not you, Jack) and they don't even want a micro size phone. Just something the size of the old Western Electric “Princess” line would be great. With regular old AA batteries to power it.
Why does everything have to be so small? Plastic doesn't cost THAT much!
– Andy Kunz
John is right. As has been noted in Embedded Systems mag before (sorry, I forget to whom I need give credit for this paraphrased quote): “Keep it as simple as possible, but not too simple”.
Sachin is also right. There are two issues: UI and feature creep. Features can creep as long as the UI remains Usable. Absent a good UI however, even basic features are cumbersome.
I for one welcome the Jitterbug and its ilk. Simple, non-feature laden appliances have a niche, at least for me. But they need to work. It should come as no surprise that I, like most today, am surrounded by digital appliances, feature rich and not. I take issue with most of them. My DVD player is notorious. It's actually a Polaroid brand but notorious would be a good name too. If I press the eject button too soon after power up I can expect to never see my disk. I must cycle power and try again with a modified “timing sequence”. And where's the power switch? It's that cord that plugs into the power strip inconveniently located behind the entertainment center. The power button is software assisted and of course, doesn't work when the thing is locked up trying to figure out why I want to eject the disk so soon after power up.
Simplicity AND function AND good UI = TRUE.
– Steve Ciricillo
Even if a gadget has an intuitive interface, it can still be difficult to use if there are too many features in the device. While the “swiss army knife” approach sounds good, this tool does a lot of things, but doesn't do a lot of things great. If i'm in the woods, and need to cut a tree for firewood, I'd have a tough task doing it with my trusty swiss army knife – bu my ax would have that campfire burninig pretty quick.
What I'd like in a phone would be a GOOD PHONE – how many times have you lost a call, and you can see the cell tower! I think sometimes these phone squeeze so much non-phone functionality in, that the product is not very good as a phone – which is the application I bought the bloody thing for in the first place.
Certainly in designing products, we need to listen to the marketing and sales guys, but we need to resist features just because the competitor's products have them. How much use do wome of these features get, and how to balance that usage with the increased cost of the product (hardware and software development) to develop is what we should be focusing on, IMHO.
– Tom Mazowiesky
Great article. Nothing really new here, other than the complexity of the menu systems, except that more and more people are encountering it. Back in the 70s my employer decided to develop an embedded-systems pyrometer. In no time at all it had an HP-style RPN calculator included in the specs.
A few years later I was leading a team developing a processor emulator aimed at production instead of R&D. A few words in the wrong ear and suddenly voice feedback was in the requirements spec.
– Bill Swan
I agree with the prevailing argument that there is a difference between number of features and usability. In fact, I think product designers and managers are to blame for not vesting the adequate time and proper people/skill-sets, during design cycle. Human-computer interaction and “human factors” should be considered a subdiscipline in itself not to be easily relegated to the miscellaneous pile. Not all software engineers are created equal with respect to UI design (I admit that it's not my strongest suite); embedded engineers, especially EEs, are, in my opinion, the worst in this category as we've made careers and focused our talents at solving design trade-offs – i.e., maximizing “real estate” – that, while related in some way, lead to choices orthogonal to the ones at hand when talking about human interaction. Determining that the pin-to-pin layout for components on a PCB should flow in a certain direction, or what the order of bit fields should be for a device register, is a different matter than determining which way the menus of a GUI should branch for people raised in English speaking cultures.
On the other hand, I do agree with Jack that the feature-bloat of some products is counterproductive. While working in defense and aerospace industry, I came to the (once again, personal) opinion that many product managers and market/biz development folk choose to vest in inconsequential features because they don't have a plan to solve the “hard problems” the customer really wants solved. In defense and aerospace, a some of these are: selective signal targeting in dense co-channel environments like urban areas, dealing with non-monotonic frequency dilation as part of the “moving emitter” problem, or modeling space craft rotation with higher precision from ground segment. For the telecom industry, signal quality and reliability (“dropped calls”) would be one. It's easier for a vendor to add ring-tones to a cell phone than figure out how to lower call drop probability.
– Shawn Price
I don't mind a new feature as long as 10 of the old, useless features are removed first. Features should evolve as their usefulness or lack thereof are proven.
— side note: check out the toyota yaris for a straightfoward, only-what-you-need interface. Haven't read the manual once.
– Susan McCord