Nine Lies We Tell Ourselves About Mobile
As someone who spends a lot of time looking up mobile stats, poring over research reports, and who was around for the early days of mobile, I can barely stand to read half the articles that try to wax philosophical about mobile’s history, or draw some broader conclusion about the industry and it’s direction. They are almost always based on some mistaken assumption, and we perpetuate lies about mobile by putting them down in writing and spreading them around.
I’ve been doing mobile design since 1999. I was there for the dawn of location services. I worked on the first US mobile data network of any note, and designed stuff like the first MP3 music service for mobile, a very early app store, and the first Google mobile search.
I saw a lot of this first hand, working with others to help come up with some basic truths about our narrow fields of interaction and interface design. And from this I have to say: if your first experience with mobile was designing for the iPhone 3G, you’ve been lied to. And this is a bad thing.
In an effort to begin setting the record straight, let’s look at some of the most common lies we’ve been telling ourselves about mobile.
1: 90% of our Mobile Users are on the iPhone
This myth isn’t specific to Apple products. The point is that the world is too complex for me to believe any number that’s 90, or 95, or 99%. For any number that radical you better validate.
I have worked with several clients who have been given an analytics report with this type of information, and who have been working off this kind of information for years. When I first saw this particular figure, I was working for a global company and Symbian was still the dominant smartphone platform. Remember, Nokia sold more smartphones than anyone else for five years after the iPhone came out.
Drilling in to the numbers, the most common error is bad analytics, and not changing with the times. More than once I have found that this is the result of zeroing out all the “unknowns.” Remember that from the old days? No one wanted to see that messy 1.7% of various Unix users, so your analytics guys just erased them from the data. Well, when they started tracking mobile, they did it the same way. The analytics package got an iOS plugin, which allocates every mobile Safari user to “iPhone” and all the other mobile users are unchanged, or “other.” Re-run the numbers properly and suddenly you find that you have four times the mobile users you expected, 60% are Android, 20% are iPads, and it’s time to write a new mobile strategy.
You can also get there by cherry-picking the data, or by not using valid gathering methods. I have seen information from a mall-intercept in San Jose used as a statistically valid sample years later. Think back to high school math class. Estimate the answer first. If your numbers are way, way off the global or regional standards, you better dig in a little more and find out why.
2: Before the iPhone, We All Carried Flip Phones
Untrue. The phones we carried were all different shapes, and a flip-phone was just one of them. Some even had large screens and carried excellent cameras, played media very well, and had decent browsers. Many of us also carried smartphones. Oh, sure, many fewer, but some. These even included touchscreens, though almost all used resistive technology and offered a stylus, for what seemed like good reasons then.
RIM had around 2 million Blackberry users in 2005, almost all in the U.S., where, admittedly, 208 million phones were in use. Including Palm, Windows and a few others, around 2% of the U.S. population carried a smartphone.
And remember smartphone subscriptions in the US only passed 50% about a year ago. In other parts of the world it’s higher, or much, much lower. You and all your friends have a smartphone, but half of your potential customers do not—even today. There are lots of featurephones still lurking around.
3: Where Dumbphones are Still Popular …
You can pretend the “dumbphone” doesn’t exist. There are a few dozen million out there, but in a market of seven billion phones, who cares? The device that’s a step down from a smartphone is a “featurephone.” And well over half of the devices in the world are still featurephones.
Products I have worked on in Africa and the Middle East—popular, revenue-generating products—have 90% featurephone visitors. For mass market services in the U.S. like banks, more than 20% of your visitors will be on these “low end” devices, even though the website targeted for those phones is insultingly awful. You are simply missing out on 30% of your user base by not targeting their mobile devices.
Featurephones have proprietary operating systems, but can install apps and have a web browser, and usually a camera and everything else you are used to on a smartphone. People use them for mobile banking, for Facebook, to buy music online—for everything you’d expect a smartphone to be used for. Usually less so, but with the scale of users we’re talking about, that’s a lot of use.
Featurephones have pretty good browsers these days. Most are Webkit, and many are the same browsers available on smartphones. Opera Mini for example ships with some featurephones, so you barely need to address constrained memory or processing or browser support. I am still seeing designs for low-end phones assume some are grayscale. This is not a thing. Featurephones are low-end or different, but not actually old. Be dismissive of the non-smartphone only if you want to leave 4 billion customers on the table.
4: The iPhone Changed Everything
Sure, this is true, but mostly due to mindshare. Now that every designer has an iPhone they think it’s important to design for them. The technology we use has been around for years. And this is not trivial or pedantic. I don’t mean that Apple didn’t invent touchscreens. I'm talking about technologies we use every day to make our work better.
WURFL, is a good example. It’s the most popular and widespread of the technologies used for “device detection” or figuring out on the server side which mobile phone and browser is contacting your site. That seems really useful with all this fragmented Android stuff right? Well, Luca Passani started working on it in 2001, because the same issue came up right away with the variation among all those featurephones.
A lot of the core technology we use when designing or developing for smartphones goes back years, or decades. That also means the next big technology shift—or even just small shifts such as platform changes like Android taking over today—are not to be feared or dismissed. Everything changes, but slowly and only on the surface. The fundamentals don’t change so quickly.
5: WAP Kept the Mobile User in a Walled Garden
First, some definitions: WAP is the Wireless Application Prototcol. It was (very approximately) a subset of HTML that allowed web-like content to display on very low-end mobile devices. This limited web marked a sad time, but did not last long. Anyone who is designing for WAP now (and they do) is wrong, or using the wrong word when accounting for low-end phones.
A walled-garden is a contained area that is difficult or impossible to escape from. The term is most often thrown about today in discussions of things like Apple’s App Store. But in the bad old days before the iPhone, mobile network operators kept you to their content only. The mobile web was no good because you could only see what Verizon, or T-Mobile, or Vodafone wanted you to see.
Except that’s not really true. Sure, for that brief time when WAP or i-Mode (a similar technology only used in Japan) was the way onto most phones, there was no point in trying to view the rest of the web because it wouldn’t work. But even that was not universal. The Nokia Communicators way back in the 1990s, for one, had pretty fully-featured browsers.
But if you talk about the U.S. market, I was there when it launched. And I admit we tried—for misguided reasons, which I disagreed with!—to push the walled garden. But even at launch day, you could go wander away and browse or search the entire web as you wished.
The coming of the iPhone—as far as I can recall or tell by researching—changed nothing about this. Everyone had opened up much more completely years before this. Even the Apple choice of always trying to load the desktop version of a website was not new. Other smartphones thought that was the way to go or offered it as an easy-to-find option.
6: Before GPS, Location Came from Triangulation
Um, no. To be really, really picky, it’s not GPS and it’s not triangulation but GNSS and multilateration, but that’s not the biggest issue.
This is another area where I worked directly. I promise this is true. Sure, the phone network could pretty much automatically tell where you were based on cell-ID, sector or, if lucky, multilateration (a 3D version of triangulation). But mobile network operators are risk-averse and lack vision. All of them decided not to build services to do anything with this but to wait until GPS made its way onto phones. So, very literally, phones did not have access to the telemetry data the network had on them.
When GPS (then WAAS, and more recently GLONASS as well) was installed, the first phones with it were almost useless. They could only use GPS directly, which therefore took forever to determine your position. Eventually operators added the capabilities into the network to send the information on basic location (AGPS) to the phones and it got better, but it took two or three years.
Due not to technology, but entirely market choices, location services did not move from good, to better, to best, but emerged pretty much fully formed, years later than they could have, and then the services—the apps and websites using location to make the experience better—took a decade to catch up.
7: In Places Without High Speed, 4G Data …
Outside a big city, in the U.S. or most parts of Western Europe, 4G is unheard of. Unheard of. Seriously. I have been at conferences where the line is changed to “for users without 3G data…” Much of the world has very poor data networks, and 4G LTE is a dream they heard about once. It’s not coming soon.
Also true, much of the world, and many Americans, also can’t go down to the corner coffee shop and get free high speed WiFi either. You have to design your mobile services for everyone. The point here is that the exception should be made for places with 4G.
8: The Reduced Speed of Mobile Networks Means …
Yeah, sorta. But in some ways, mobile networks are about as fast as most available wireline networks. The problems occur in latency. If you build a website that makes 42 calls for files (images, js, css) then each of those has a significant overhead in time while the network negotiates to get it, and figures out how to slot it in to bring it back. The same website, with everything embedded (yeah, you can embed images) can load 2 or 3 or 20 times faster just by reducing the number of things you ask for.
This is not being nitpicky, mobile cellular networks are not long-range Ethernet and have very different ways of communicating. This sort of detail matters, especially when the developers don’t bother to use their phones or their “data minutes,” and do all the testing over WiFi or in emulators. That won’t reveal this problem, and they will dismiss complaints by real users about performance.
9: Text Messages Are Just Data, So …
The misconception arises from thinking that typing letters into one end of a system, and having them come out the other end as letters must equal “data.”
Without getting into too much detail, SMS (the Short Messaging Service, what text messaging is really called) travels over the “signaling” part of the network, the one used to make the phone ring and so forth. This is a part separate from the payload portions where voice and data travel. That means you can send SMS with non-data phones, over non-data networks, and since this is the most robust part of the network, you can send these messages when your network coverage is so poor that nothing else on the phone works.
Want to be more confused? MMS, the version of SMS where you attach a picture, is no longer like this. The whole message goes over the data network. If you are building a service for an area with poor network coverage, you have to avoid sending MMS if you want to keep the user’s costs down, and make sure the message actually gets there.
The Danger of Lies and Forgetting History
A lot of this information came from simply reviewing literature from the past and figuring out from newer research how to apply it to current situations. I think reason it often gets ignored is that no one else seems to bother to look. The lie that “everything is different now that [technology]” and the increasing speed of innovation means we all become comfortable dismissing the past.
Because of this, we run the risk of spending a lot of time basing our deeply thoughtful design effort on improper foundations—on stereotypes, exaggerations, gut feelings, and personal bias.
The danger with this should be apparent to UXers. We have to get out of our comfort zones to discover and understand the truth about technology, systems, organizations, networks, and markets, not just people. People are not, fundamentally, changing. We have eyes, fingers, voices, houses, and jobs. Our lives, goals, fears, and loves are much the same whether we have a kitchen phone with a cord, a featurephone, or a 4G LTE smartphone.
- Tomi Ahonen: Tomi Ahonen Phone Book 2012 (TomiAhonen Consulting, Hong Kong, 2012)
- Cathy De Rosa, et. al.: Perceptions of Libraries, 2010: Context and Community (OCLC, Dublin Ohio 2011)
- Canalys: 64 million smart phones shipped worldwide in 2006 (February 12, 2007)
- Horace Dediu: “When will the US reach smartphone saturation?” (Asymco, October 7, 2013)
- Dan Nystedt: “U.S. marks new cell phone record in 2005.” (InfoWorld, April 7, 2006)
- The World Bank “Mobile cellular subscriptions (per 100 people).”
- Steven Hoober and Eric Berkman: Designing Mobile Interfaces (O'Reilly Media, Sebastopol California, 2011)
- Ilya Grigorik: "Mobile Performance from the Radio Up: Battery, Latency and Bandwidth Optimization." (Google I/O, May 17, 2013)
- Kevin Fitchard: “Mapping out the world’s LTE coverage (It’s in fewer places than you think).” (GigaOm, September 20, 2013)
- Gail Frederick and Rajesh Lal: Beginning Smartphone Web Development (Apress, New York, 2010).
Image of hanging mobile phone courtesy Shutterstock