Developers have clamored for a paid upgrade system since the App Store’s inception, but I’m worried Apple won’t offer this feature anytime soon. I’m far from alone – Instapaper developer Marco Arment also predicted Apple’s non-action on his latest podcast. Yet Marco and many others don’t think this is a problem, that the current à la carte system is “the future” of software publishing. They’re completely wrong.
Apple drags its feet on paid upgrades because Apple wants simplicity for their customers. A choice between a full product and paid upgrade muddles this philosophy. For now, all users get all app upgrades automatically. If you introduce optional, paid upgrades, certain updates only apply to select customers. This adds complexity for consumers and developers having to juggle and maintain multiple app versions on the store.
In addition, the lack of paid upgrades keep app purchase prices lower. This is simple economics: on the App Store, developers have to force a repurchase between major versions in the form of a new app. App prices will be driven lower to offset the much larger sticker shock between versions and to account for boosts in upgrade revenue (100% of the product cost instead of some smaller fraction.)
I don’t think this is a good economic model, especially for more expensive, professional level software, such as Omnifocus and Photoshop. But remember, Apple is not, at its core, a software company; they make money from selling iPhone and iPads. The cheaper the software, the greater the incentive we have to keep on buying Apple’s hardware.
Also it’s professionals and power users – both niche Apple consumers – that demand paid upgrades, not the core audience. Given how rarely Apple updates its pro products (e.g. Mac Pros, Aperture, Final Cut) in the last few years, we’re in for a serious wait before Apple takes any action here.
Marco defended Apple’s inaction on this week’s Build and Analyze podcast; I disagree with him. A lack of paid upgrades causes two main problems:
The absence of an incremental purchase kills a huge source of revenue for developers and publishers. Without both new software and upgrade streams, many publishers, from high end publishers like Adobe to independent studios like Delicious Monster and Panic have a hard time staying afloat. In economic terms, a lack of price discrimination between more receptive, existing customers and newcomers is a major problem.
Customers get angry and frustrated when they can’t upgrade. Naturally for apps under a few dollars, this issue doesn’t apply, given the small investment. Yet if you’re considering $40, $80, or hundreds more for a purchase, an upgrade by way of repurchasing the software is costly. It’s especially aggravating if the next big upgrade is launched soon after you buy the old version. This keeps potential buyers on the sidelines between major releases, exactly the time when many developers need revenue to keep going.
Apple needs a change in course here. A lack of paid upgrades is killing both developer revenue and consumer interest in a lot of great apps. It’s also watering down app quality, especially in the iPhone and iPad markets where there’s no App Store alternative.
In the iOS and Mac app stores, newcomer generalist apps are dead. Long live the new wave of hyperfocused apps.
This point was inevitable given both stores have reached a saturation point. There are so many calendars, text editors, todo lists, weather forecasts and photo editors – to name just a few categories – that it’s increasingly rare for any newcomer to stand out. Several success stories emerge early (e.g. Omnifocus and Todo for todo lists, Camera+ in the photo department) receive positive coverage, gain a user base and iterate. Meanwhile most competitors flounder and struggle.
Yet developers are opting out of this Darwinian cycle by going very deep, singular and focused with their app functionality. I wouldn’t use the term “minimal” because some are loaded with options and customization for power users. “Hyperfocused” fits better as each app’s direction is simple and straightforward. Where a generalist app might have ten features, a hyperfocused app has one, but executes that one feature with depth, polish, and well thought out design.
Not every app of this style can be a winner – their very focus makes them divisive – but a few have clicked well with my workflow: Drafts and Dark Sky for iOS and Take Five for the Mac.
Drafts
Unlike other more generalist text editors that expect a setup process for new documents, Drafts presents you with a blank document and keyboard on every launch. There’s no required taps for a new document location or file type; open the app and you’re ready to type with little lag. Drafts at its core feels like the default Notes app with a serious speed and UI upgrade and that alone should appeal to many.
But speed is only a fraction of Drafts full functionality. A tap of an icon below the document reveals a full action list. You can copy to the clipboard, email, send to a Dropbox folder, tweet the content and send the text to other iOS apps. I use it almost every day for ideas capture, drafting Tweets, sending interesting links to Dropbox and writing extended emails.
Dark Sky
For weather I’ve had the My-Cast app on my home screen for over a year. Its got plenty of information and accurate, but generally a bit sluggish and the visuals need serious work. Also before heading outside I have to tap through several screens just to determine if there’s rain in the immediate future.
Enter Dark Sky, an app that’s singular purpose is to tell you if it’s going to rain in the next hour. After starting the app you get a graph and text description that measures the severity and chance of rain. The app excels in its detail – the graph can convey at a glance when an incoming storm will peak or when short gaps in the rain will emerge. Text descriptions are highly descriptive (e.g. “light rain for 14 min”). If you want something more visual, a great looking radar is a tap away. The whole package is fast, accurate and reliable. It’s found a nice home on my second iPhone screen.
Take Five
I’m a heavy iTunes and Spotify user on my Mac, yet the UI of each app is cumbersome and bulky. The row based, options everywhere design works well for heavy lifting but 95% of the time I just want to know the details on what’s currently playing.
To address this UI bloat, several iTunes and Spotify mini player apps have popped up. I tried both Simplify and Bowtie, two popular options. Yet while both did the job, I wasn’t crazy about their memory footprint and occasionally rough visuals.
That led me to Take Five, an option by Iconfactory, the design shop responsible for Twitterrific, xScope, and Flare. It’s a now playing visualizer pared down to the essentials: album art, song, album, and artist. Yet in targeting such a simple feature set, IconFactory delivers a really well thought out experience. Its got best in class visuals with a cool blue and black color palette. Keyboard support extends to a show/hide hotkey for the music app you’re using, be it iTunes, Spotify, Rdio, or five other players. You can turn on a Growl-like auto notification that pops up the mini player briefly when the track changes (with Spotify’s often shoddy Growl integration this is an especially useful feature.) Take Five’s main ‘hook’ is in its pause functionality; with a keyboard shortcut or icon click you can pause your music and have it auto fade in after a set period (hence ‘Take Five’). It’s a cool perk for quick breaks.
“Save for later” apps – apps like Instapaper that capture and cleanly format text articles for later consumption – are essential to my workflow. I rely on them to read long form content for my job, for blog posts and just for fun almost every night. But last week there was a serious shakeup: Popular app Read It Later reinvented itself as Pocket. It aims to be a save later service for not just text articles but almost anything online, from videos to photos and mp3 clips. That’s ambitious, something I had to investigate further.
Thirty articles and a few days later with the app I’m hooked. Overall Pocket is an awesome app, albeit with a few rough patches. It’s a tool I’d recommend to almost anyone, especially to iOS newbies given its straightforward setup process. There are several things that Pocket does especially well:
A consistent experience across multiple platforms. A lot of other media apps provide a smart UI on both the iPhone and iPad. Yet it’s rare to see an app ecosystem work so consistently on the iPhone, iPad and the desktop. With Pocket there’s a uniform, drop down based navigation on each device that’s easy to use. Its grid based, Flipboard-esque layout works especially well on the iPad or web while remaining fully usable on the iPhone.
Visual design. Many apps dedicated to browsing or media discovery have a color scheme and layout that is heavy on contrast or overly skeuomorphic. It makes a strong first impression but can get a bit boring or distracting when you’re trying to browse through or read individual articles. Pocket avoids these problems by leveraging a light palette with subtle contrast and few gradients to maximize readability. This minimalist design looks borderline “non-native” to the iOS platform, but I think for Pocket it’s a smart move. The look feels fresh and distinctive, much in the same way the Twitter client Tweetbot distinguished itself visually with a chrome, metal and gradient heavy design.
App integration. This is where a lot of competition falls short; you can have an awesome reading experience, but that becomes meaningless if you can’t move articles in and out of your save for later app easily. That’s not a problem with Pocket. It uses the same API as Read it Later which has been around for years and consequently there’s huge app support.
Video integration. I’m a big film nerd, so naturally I capture a lot of clips, video essays and trailers. Pocket has native support for Youtube and Vimeo, which gives each saved video article a proper headline and thumbnail. With two taps I’m watching a video full screen on my device. Instapaper, Readability and other choices either can’t play video at all or add a lot of cruft around the video itself.
The setup process. Pocket goes out its way to make capturing content as easy as possible. On iOS devices it identifies other apps that are Pocket compatible and provides custom setup instructions for each. To add a web bookmarklet, an essential capture tool, its step by step tutorial is best in class.
Nevertheless Pocket isn’t perfect. The app’s filter for switching between text articles, images and videos is occasionally inaccurate; usually the articles view accidentally pulls in a few videos or vice versa. Also Pocket’s web site needs a bit for work on typography; its body text color is too light and it doesn’t offer the same font choices available on its iOS app. Finally while the default sans serif and serif options look nice, text customization (i.e. font choice, line height, margin size) lags behind what Instapaper provides.
So is Pocket better than Instapaper? Yes and no. If you trend toward content that’s graphic heavy, video based, or anything that strays from pure text, Pocket should be your first choice. For die hard readers of news articles, blog posts and other text-heavy content, stick with Instapaper.
I plan on using both: Instapaper for reading, Pocket for videos and everything else. I’ll detail in a future post exactly how I integrate both apps into my daily workflow.
What qualifies as a ‘great’ iOS app over the long run? For me it’s simple: It saves me time. It doesn’t have to have a great icon, a great design, sexy graphics or get lots of praise from tech bloggers. If any of those traits add to saving time (and they often do) great, but time and efficiency outweigh everything else.
I use Drafts because its simplicity and raw speed saves me a few seconds every time I have to capture an idea or reminder. IA Writer’s clean typography and lack of customization focuses my mind for longer form writing. Marsedit’s quick WordPress and browser integration saves me a few minutes for every linked list post I make. Omnifocus syncs effortlessly and reliably between my Macs and mobile devices; I spend little time worried about lost contacts or todos. With Reeder I can scroll through and consume a day’s worth of tech, design and film news on my subway commute home.
Paring down your app set to mostly those that increase efficiency or save time isn’t a groundbreaking idea, but it is easier said than done. Like many in the tech industry, I get a regular share of recommendations via Twitter and RSS. I use to always download what had buzz with the tech bloggers, what was ‘innovative’ and what just looked cool. Yet after playing with a hot new app for a few days, 95 percent of the time I’d delete it or move it to some back folder, never to be touched again.
Don’t let this be you. Make hard decisions on the apps and tools you use. Granted there’s always edge cases: Gaming apps by their very nature should be arguably something that takes more, not less of your time if it’s a fun experience. There’s also something powerful with occasional experimentation: I downloaded Clear knowing full well it wasn’t a tool for me. Yet just playing with the app for a buck and hour of my time gave me design inspirations for my day job. Not everyone has the same priorities either. With my mobile workflow, saving time is paramount; I want to get in, get my work done and get out as efficiently as possible. You might instead favor aesthetic beauty, or great icons, or other traits.
Whatever that app goal is, stay focused. Is that new app that’s new and noteworthy on the App Store really going to integrate well with your workflow? Is it really better than what you already have? Ask those questions before you download.
Video game consoles are still putting up great numbers seven years into their current generation. But why have their user interfaces remained so bad? I was reminded of this on a popular Giant Bombcast (gaming podcast) from two weeks ago; the hosts talked at length about the sad state of Microsoft’s latest XBox Live UI refresh. Microsoft largely sidelined avatar functionality, one of the rare bits of personalization and whimsy from an otherwise business-like UI. The Netflix interface was overhauled so poorly that the hosts had moved their film streaming needs to other platforms. Common actions now required more taps of the controller than in earlier XBox Live iterations.
Ironically, XBox Live is generally regarded as the premier console gaming network. It costs $50 a year and generates a lot of revenue for Microsoft, a cool billion two years ago. So why isn’t some of that money being plowed back into great UI design?
The XMB, Sony’s navigation interface for the PS3, doesn’t fare well in the UI department either. Among the Roku, Apple TV, Mac, iPhone, and Boxee, all of which I own or have played with heavily, PS3 has the worst user experience. There’s too many actions and layered menus to get more complex actions done. Software updates, large in size and not skippable, pop up frequently before gameplay. (Sony apparently never got the memo on auto background updates.)
Yet UI may be beside the point: clearly the healthy state of console gaming’s market derives from the games themselves. But that market is changing, growing up and moving more mainstream. XBox 360s are being used now more for streaming media than gaming. A “one box media center” for the living room could just as easily be an XBox as a Roku or an Apple TV. Media partners clearly see this; content providers from Amazon to ESPN and HBO are supporting consoles in full, often adding their services to the XBox and PS3 just as fast as other set top devices.
In addition, while a Xbox 360 or PS3 costs $150 more than an Apple TV, that a premium price tag delivers far more capable hardware. It’s hardware that powers more immersive games, along with more responsive and novel interfaces (e.g. the Kinect) than their cheaper counterparts. Beefier hardware also means getting cool tech features (e.g. Dolby Digital 5.1, 1080p) before the competition.
Yet as we’ve seen before, muscular tech, lots of money and media partners will only get you so far without a solid user experience; just ask RIM. Competition is heating up: Apple and the rest of the portable market is on one side, chipping away at consoles’ casual gaming segment. Smaller, cheaper and simpler boxes from the likes of Roku form the other wing, attacking consoles’ non-gaming features. Without a adjustment in UI and other consumer-friendly maneuvers, I fear gaming consoles could be effectively squeezed out in the middle.
I was initially as shocked as everyone else when I heard the news of Facebook spent $1 billion to pick up Instagram. But then it came together in my mind: Instagram’s purchase is major win for great mobile design, for online products with high engagement and fast code.
None of those things make sense at first glance. Instagram is a photo sharing app without a proven business model or positive revenue. A company of thirteen (!) employees who’s majority share of 30 million plus users are already Facebook members. A social network that at its core isn’t groundbreaking; sharing photos on a wide scale has been done with Flickr since 2004.
But the deal did happen, and there’s there’s several standout lessons here for designers and developers rooted in the reason why:
If you create a product with very high levels of engagement, you can be a threat to some of the largest tech companies out there. Facebook saw people shifting their mobile time away from Facebook to apps like Instagram and wanted in. Instagram just has a certain cache, or ‘stickiness’ with their app. For now at least, when influencers want a ‘cool’ way of sharing photos, Facebook and Flickr often aren’t their first choice. Instagram is.
Why? Great engagement derives from unique, emotionally driven design. Granted, part of Instagram’s engagement comes from marketing and sheer luck. Nevertheless, Instagram’s design is standout. For example, other apps use photo filters, but not with the same range, fun naming convention or ease of use to jump between them; Instagram makes post processing fun. It’s been one and a half years since Instagram’s debut, yet how many other apps can make that same claim?
High engagement levels can be retained with simple, straightforward design. Instagram was one of the first apps to make sharing to multiple social services so damn easy. It doesn’t take many more steps than your typical iPhone shot: capture, pick a filter, pick who to share to and you’re done in three taps.
High engagement is generally maintained only from a quick, responsive app or platform. This is where blazing performance at the development level comes in. Earlier in the year an in-house engineering blog post revealed some of Instagram’s tech under the hood. It’s straightforward, well thought out and was able to scale to 14 million in a year.
Nevertheless, a lot of positives about this deal can’t stop skepticism on my part. I don’t trust Zuckerberg’s claim that sharing with non-Facebook social networks will remain unaffected. Facebook has reneged on its promises repeatedly and likes a tightly controlled ecosystem. I’m thus also worried about the app’s “independent” future long term.
Overall, for a company of Facebook’s size, 1 billion isn’t crazy, “we’re in a bubble” cash. It’s a buy with stellar engagement levels and a mobile weapon to keep away from the other big tech players: Google, Amazon, and Apple. Given the stakes in this arms race between these four companies, don’t be surprised if the numbers only ramp up in upcoming years.
Sometimes the smallest of changes can deliver a huge impact. In my case, several weeks ago I switched away from the popular Mac text editor TextMate to Sublime Text 2, giving a significant boost to my productivity in the process.
Some context is necessary here: any web developer can tell you that, with the profession’s focus on coding, text editors are generally the most important tool at hand. For the last five years, TextMate was my editor of choice and I depended on it for effectively everything at work: HTML, CSS, Ruby on Rails, even extended emails and design notes. I loved the speed, the keyboard shortcuts and its slick bundling system – macros, commands and templates came together in a unified package.
But ever since Mac OS X Lion came out, TextMate’s lost a lot of its luster. The downward spiral began with the program’s incompatibilities with much of Lion’s new functionality; this only amplified the nagging suspicion TextMate would never receive a significant upgrade (its last significant release was version 1.5 back in January 2006.) The TextMate team unexpectedly announced a public alpha of TextMate 2 for release before the end of 2011, but I think it’s a case of too little, too late.
I’m a relative newcomer to Apple and Steve Jobs. Apple computers were an influence growing up, but I didn’t really start using Apple products until my post college days. Yet, after getting my hands on my first Mac, a Powerbook G4, I was hooked. The accessibility and design of Apple’s products always matches the tech power under the hood – a rarity in our industry. I don’t think any of it would have been possible without, as Jobs often said, Apple at the intersection of technology and liberal arts. It makes the company’s work unique and often amazing.
Jobs was a genius, yet still fallible; his flare ups of anger and arrogance were well documented. Paradoxically, the weaker sides of Jobs made him even more approachable, human, and charismatic. That made news of his passing all the more tragic. Not every Apple launch under Jobs has been a success, yet his successes fundamentally changed the technology industry at least three times with the Mac, iPod, and iPhone. Who other single person has had that kind of impact?
Jobs also had an incredible gift at intertwining technology and design, along with a relentless drive. Both are traits that I strive to achieve every day in my career.
Rest in peace Steve and thanks for always moving the ball forward.
With HP cutting its losses on the Touchpad and PC markets, the “post-PC” era begins with Apple firmly in the lead. The evidence is overwhelming: HP’s exit leaves Apple as the only major consumer PC maker with increasing profitability. Yet even with its leg up on the competition, Apple is shifting away from PCs. Over 70% of its previous quarter’s revenue was wrapped up in iPhones and iPads; it’s fair to expect that percentage to only increase in the future.
However, I disagree that the failures of HP, RIM and other companies in the tablet sphere make a long run Apple victory inevitable. In actuality, the post-PC era is divisible into two eras. The first is a transitional, “tablet as entertainment” era that we’re in now in which Apple clearly dominates. Yet there’s a final, longer term “tablet as PC” era of the future where I doubt a single company will control the market.
We’re in a fast changing digital landscape; innovation has worked into almost every device I use regularly and with cloud syncing my content is accessible from anywhere. So why do higher end cameras, most notably digital single-lens reflex (DSLR) ones, feel absent from the picture? Big manufacturers like Canon and Nikon push out better sensors and higher end lenses year after year, but the core design remains unchanged. Contrast that progress with the rapid evolution in smaller, camera-equipped devices like the iPhone and Canon S95, both in terms of technology advances and mainstream adoption. If the trend continues the DSLR will be relegated to a niche device for professionals only.
The problem starts with perception: compared to hot new consumer tech like the Kinect and iPhone, DSLRs, at least to a mainstream audience, are run of the mill if not behind the curve. First and foremost, in a golden age of lightweight, powerful technology, DSLRs remain heavy and bulky. They also have a higher learning curve than most other tech devices: even using its simplest operation, the most basic DSLR offers a bewildering array of dial settings, zoom levels and menu adjustments.