The case for consistency: native design isn’t always right

Quite a few companies are reclaiming mobile app development these days. Some are moving outsourced development in-house, and others are consolidating line-of-business development projects under a single department. In both cases, one of the most popular questions they ask Gigaom Research is whether to go native, hybrid, or HTML. It’s a perfectly valid question, but in many cases, they’re asking it for the wrong reason.

“Native or not” remains an important choice when you’re concerned with application functionality. HTML-based development (the “mobile Web”) trades performance and control for increased code reuse across devices. Native apps, coded specifically for a target platform, take longer to develop and port to other platforms, but allow developers to squeeze out every ounce of device performance or write to specific hardware features. By placing HTML and JavaScript inside a native container, hybrid apps split the difference between the two, combining a large amount of code reuse with substantial device connectivity and performance advantages.

Choosing a development strategy can have a huge impact on your application’s responsiveness and feature set, but a lot of the questions we get are about design, and with rare exception, there are three reasons why native design (making an app look and feel like it matches the operating system’s UI) often isn’t worth the time.

1. Good design is app-centric

I’m not claiming that design is unimportant. Quite the opposite. Mobile design is critical, and in some cases, as Gigaom Research Analyst Paul Pangaro outlined in Designing for mobility: directions for mobile UX, design can actually define your app – like Tinder’s left-right swipe. That’s precisely why your choices need to reflect the best possible user experience within your app. No one is suggesting that you toss decades of UI on a whim, but if custom icons, fonts, gestures, or menus work best for your app and the context in which its users will interact with it, that’s far more important than staying in sync with the OS running it.

It’s important to remember that mobile apps are rarely used in conjunction with other apps beyond a standard set of export functions such as email, social sharing or copy-and-paste. And given the size of even most tablets, it’s rare to have more than one app active on the screen at one time. This frees developers to focus on building the richest, most appropriate context for one specific app.

2. Developers are busy, and customers are impatient

As another Gigaom Research Analyst, Rich Morrow, noted in his Sector Roadmap on cross-platform mobile development, the “talent crunch” is the biggest challenge development organizations face. There just aren’t enough qualified developers and designers on the market, and enterprises and ISVs alike need to focus the limited resources they do have on tasks with the greatest payoff. At the same time, mobile users are notoriously impatient and more than happy to find another app to meet their needs. Both situations call for a faster release schedule, and hybrid apps with a single, shared interface across platforms are often the middle ground to get you there.

3. Consistency is king

We’ve all experienced the annoyance of “That isn’t where it was in my version of Word” when using a friend’s computer. Imagine that experience repeated across an iPad, an iPhone and an Android handset as you access the same app from different devices. Now imagine supporting that application or training users in a BYOD environment. A consistent interface across platforms is far less likely to frustrate users, and it will make your business more efficient, too. If you’re mobilizing an existing Web or desktop app, there’s a good chance that your mobile UI will be substantially different from the original — optimized for touch, smaller screens and focused use cases — but consistency among mobile platforms is good for everyone.

There are always of exceptions. For example, if you’re building an app that extends the operating system or another core application, matching the default UI is core to your value proposition. But if you’re building B2C apps that can stand alone or developing any sort of enterprise app, consistent design is much more important.

Amazon’s Echo is a good listener but a wretched assistant

Never has the gap between a flawless technology experience and a closed ecosystem loomed as large as the gap between the Amazon Echo and the Ubi personal computer. While Amazon’s Echo works beautifully and is a gorgeous cylinder that is ready to hear and (attempt to) obey my every command from pretty much anywhere in the room, it fails because its abilities to connect with a variety of web services are very limited.

Meanwhile, the Ubi, a voice-activated computer that is older and, yes, much more painful to use, wants to do the same thing. Like a teenager, though, it isn’t adept at listening to my commands, sometimes awkwardly interrupting my conversations, and its music playback is not nearly as graceful as the Echo’s.

But what the Ubi lacks in technical ability it makes up in a democratic willingness to try to control a variety of web services via If This Then That, SmartThings and others. If you combined Ubi’s openness with Amazon’s grace and technical acumen — provided by the powerful speakers inside the Echo and the seven-mic array that pics up your voice from across the room — you’d have the perfect voice-activated digital assistant.

Instead, I paid the [company]Amazon[/company] Prime member price of $99 (it’s $199 for non-Prime members) for what is basically a voice-activated timer, task list and way to access my Amazon Prime music library. The Echo also answers questions via a Bing search about 70 percent of the time it’s asked, although some basics — such as my requests to convert a temperature from Fahrenheit to Celsius — proved unintelligible to the Echo (you can see that in the screenshot below).

Echo recognized requests from half a dozen people, including two children, although my daughter is having a hard time with Echo because she can’t always say “Alexa,” the wake word we use for the device. (Sadly, you only get two options for your wake word: Alexa or Amazon, but a spokesperson from Amazon says it will add more wake words over time.) You can’t change the search engine, so you’d better love Bing.

How it works


Before I get too deeply into my review of the Echo, let me pause to explain how it works and what it can do. The device is a little under nine inches tall and is about the diameter of a wine bottle. It has a ring of lights at the top that acts as an indicator, showing it has heard your command, and can be turned to raise or lower the volume. As electronics go, it’s elegant enough to sit in a visible place in your home. Mine’s on my kitchen counter and I can talk to it from just about anywhere in my downstairs living room, dining or kitchen area.

The Echo also comes with a remote control that you can stick to a surface via a magnet or double-sided tape. The remote reportedly comes in handy for communicating with the Echo in noisy environments, but I’ve found it’s most helpful for fast-forwarding songs that I dislike since my home is rarely noisy enough that the Echo can’t hear us. Using the remote for this is faster than saying “Alexa, skip this song.”

When you open the package, setup takes less than 10 minutes and requires you to go to Amazon’s site to download the Amazon Echo app to your Android phone. The app lets you customize things such as your zip code (so Echo can get weather and news for your area), plus specify important elements such as purchasing preferences.

The app uses the card format you may recognize from Google Now.

The app uses the card format you may recognize from Google Now.

For example, the automatic purchasing was turned on when I got the device, which meant that I could just tell the Echo to buy Taylor Swift’s album and could have it in my music library immediately. Thankfully, I don’t have one-click ordering turned on, and as added protection against my eight-year-old’s love of Top 40 songs and instant gratification, I also added a PIN number I need to enter before making a purchase.

You can use the app to see what the Echo heard when it misses your verbal request, and to check your Shopping or To Do list. It’s also where you can go to create multiple profiles so you or your partner can share a single Echo but have multiple To Do lists. I’ve found the app to be a good place to troubleshoot when the Echo gets things wrong. It is also popular in our house since we occasionally ask the device for the Spanish translation of an English word, and the translation goes into the app because Echo doesn’t know how to speak it.

What works, what doesn’t

Like any new tool, the Echo takes a bit of getting used to. I imagine in a few weeks it will have completely supplanted a few items in my home, such as the egg timer. I like the timer setting, although the alarm is a bit quiet. I use it for cooking, but also for keeping track of time — “Alexa, set a timer for 4 pm.” I also like asking it to tell me the time since we can’t see the clock from our living room couch.

We’ve been adding things to a shopping list though Echo, although we don’t use that as our master list yet for the weekly run to the grocery store. We also haven’t tried out the feature to let our daughter bring up webpages on her Kindle Fire tablet for schoolwork, but I’m looking forward to that. The idea is that she can say “Alexa, search tornadoes on Wikipedia,” and the page will come up on her Fire Tablet.

We dislike the Echo’s Amazon-centric worldview. My family spends plenty of cash on Amazon each month, but our lives are managed via Google calendars and our entertainment is on Netflix, Spotify (mostly consumed over a Sonos) and Hulu. Our home automation runs the gamut from Philips Hue bulbs to a Nest and Chamberlain garage door opener integrated with SmartThings and IFTTT. I’d like to bring those devices and services into the Echo. Amazon does have integrations with TuneIn and iHeartRadio, and I would expect to see more integrations coming since the brains of the Echo are hosted in the cloud and can be updated over time. Amazon’s spokesman does note that other Amazon products have an SDK and that Amazon does want to hear from developers about what they want to do with the Echo, so there’s some hope out there, although Amazon is a company that is known to build proprietary versions of existing open source platforms.

Is the Echo a Jambox rival?

Is the Echo a Jambox rival?

The Echo also acts as a Bluetooth speaker playing music via Spotify, Pandora and other services via your phone, so people considering a Jawbone Jambox might consider Echo instead. It sounds as good as the mini Jambox, although it’s not as adept with the bass. You have to control playback via the phone, not via your voice.

Why not add…

Since Amazon is clearly thinking about ways to build devices that push them deeper into the home and gather more data, there are ways to make the Echo more robust and yet unique enough that Amazon can truly make it an essential gadget in many homes. Here are some things I’d like to see, beyond the ability to control more web services:

  • Expansion packs for languages, so I can query for vocabulary or questions in other languages.
  • Home automation control so I can integrate it with my Nest, Hue lights or SmartThings.
  • A safe mode for my kid, so she can’t purchase anything, but also so she can’t inadvertently play music or send websites to her Kindle Fire that she shouldn’t. Ideally, this would be based on voice recognition.
  • A way to link two Echoes together and sync them, the way you can with a Sonos music player in party mode. As speakers go, these sound pretty good, so some people might use them all over their house instead of investing in multiple Bluetooth or Sonos speakers.

The end game


As a consumer and Amazon Prime member already, I’d pay $100 for the Echo because I’m old and having voice-controlled access to Amazon’s Prime Music streaming service fits my musical tastes (long live the 90s!). Plus, I like the list features and have a chunk of my music on Amazon’s cloud already. If I weren’t a Prime member this would be a much tougher sell.

If I hated the existing Prime Music offerings, I wouldn’t buy the Echo unless I already bought my music on Amazon and kept it in Amazon’s cloud, or if I were a big user of TuneIn or iHeartRadio. But I do see the device as a unique, well-done offering from a technological perspective that could put Amazon deeper into people’s homes if it manages to open it up a bit more.

I like the Echo and find myself looking for more ways to interact with it, and hope those come along in the near future. If Amazon is really a company hell-bent on getting more data to sell me more physical and digital goods, then the Echo can gather a lot more data about a lot more devices if Amazon wants to let it control more things. So both from a selling-more-Echoes perspective and a business-strategy perspective, my hope is that we see Amazon open Alexa up.

And yes, I did say that again.

We need computer sight to make smart homes smart

The smart home won’t be built using apps and connected devices. To truly embed computing into our home environment we need better computer vision, projectors and a new understanding of computing.