Hipmunk on Daring Fireball
It’s no exaggeration to say I’ve been trying to get my work on Daring Fireball since I started my iPhone development and user experience adventures three years ago. Thrilled to see it finally go down:
…[T]heir custom date picker is simply brilliant. Hipmunk has a good web interface too, but it has nothing on the iPhone interface. This is why native apps matter.
Thanks for the love, John!
I’m especially pleased he calls out the date picker. That was a lot of work to get right and it’s not the sort of thing you can get away with being wrong in something like flight search and still have happy users.
Memories of System 7.5
Almost everything good that has ever happened in my life can be traced back to my early experience with a Mac. The first family computer that ever lived in my house was a Performa 6116CD. 
I absolutely loved that thing, especially by contrast with the rest of my life. School was typically dull: I spent very little time learning about anything that was important to me. I think I could count the number of friends I had with half of one hand – and they were certainly outnumbered by people who disliked me but couldn’t find constructive ways to express those feelings. My home life was no picnic, either.
Yet none of that mattered when I was at the keyboard of my Mac. It was, all at once, a second school, a conduit to another world, an infinitely deep toolbox and a magic wand of indescribable power – running at 60 MHz.
I thought it would be fun to venture down memory lane and revisit my Mac of 1995. Of course, the hardware itself is long gone. But through the magic of Sheepshaver, I’ve been cobbling together the scraps of my favorite childhood memories. Other kids had sports, comic books or Jesus. But the thing I believed in was my Mac.
System 7.5
My childhood experience with the Mac spanned System 6 through Mac OS X 10.2 but System 7.5 was easily the golden age. That would be the first time I had long-term access to a machine I could customize any way I wanted.
Once installed in Sheepshaver, even through an emulated PowerPC processor, System 7.5 is extremely performant compared to 16 years ago. On a Late 2010 MacBook Pro, loading from an SSD, boot time is about two seconds, compared to about 30 seconds in 1995.
The cheerful parade of Extensions and Control Panels marches at the bottom edge of the screen. Performance be damned, I loved collecting these.
Of course, the System 7.5 era was extremely long – an interminable wait for Copland, the next generation operating system that would make unicorns fly from your 4x CD-ROM drive. As time went on, the UI started to look pretty stale.
So it’s important to install one of my favorite extensions from the period, called Aaron, to spruce things up a little:
That’s better. Aaron adds a little flair and dimension to the otherwise flat and bland System 7-era UI and I liked it a lot better. Even at 10, I was starting to be curious about the nuances in UI design.
AOL
AOL was my very first taste of the internet. I believe our first bill came out to $80. So that didn’t last long. Luckily, their unlimited dialup service showed up about a year later, so I would be back in action. Sadly, but unsurprisingly, you can’t actually use the AOL client anymore. Still, I got to poke around with the modem configuration panel that was a frequent source of frustration once upon a time.
ClarisWorks
This little suite was bundled with the Performa. Very little to be excited about here but I spent so many hours cranking out school reports and other projects in its Word Processing, Paint and Vector Art modules.
Hotline
The gravest of my youthful indiscretions was easily my voracious appetite for pirated software. Enter Hotline. Before Napster, before Gnutella, before BitTorrent, there was Hotline. Hotline let anyone set up a file server on their home computer. It included chat, BBS and persistent user accounts, too. Vibrant communities sprung up around these little amateur servers. They dedicated themselves to everything from religious evangelism to technical support to sharing anarchist/conspiracy text files. Of course, being the internet, there would be plenty of pirated software in the mix.
To my utter delight, the mid-90′s version of Hotline I got started with so many years ago not only still works, there’s even a handful of servers still in operation. Back then, I was lucky to pull down files at 2.8 KB/sec via dialup. A limitation of either Sheepshaver or Open Transport, the aged TCP/IP stack Hotline uses, now caps me at 60 KB/sec, but that’s a big improvement I’d have killed for as a kid.
Hotline is a major hinge in my history. With access to so much software, I dedicated myself to learning how to use it. I rarely had access to any documentation beyond what was built into the apps so it was often an exercise in trial and error. It was also fun beyond words.
This began my life-long study of interfaces and user experience. If this hadn’t happened, I have absolutely no idea what I’d be doing with my life right now.
Photoshop
Hotline could be extended with customized icon sets. If one of the two dozen included user icons didn’t strike your fancy, you could create your own. The trouble was that only other users with your custom icon file could see your handiwork.
Of the thousands of active Hotline servers in operation during its golden age, two emerged as dominant tribes vying for the loyalty and patronage of the masses. Known as BadMoon and SoSueMe, the servers collected thousands of customized user icons and then distributed them as authoritative custom icon sets.
Of course, I wanted to get in on this. ClarisWorks’s Paint module really wasn’t up to the task, so I had to find and learn Photoshop 3.0. This little head-start on graphics tools ended up being important – years later, I’d be able to design my own UI elements thanks to this early noodling.
It meant days of downloading but it was worth it.
One striking thing about Photoshop 3.0 is how very little has changed after all this time. The color picker is identical. There are the many cluttery pallets for layers, brush diameter, colors, and channels. Later versions would introduce layer styles, which were awesome but a little rigid, and endless other bits of junk. The overall workflow, aside from crappy Save For Web, remains much the same. (This is why I now use Opacity to design UI – it’s built for how I actually work.)
ResEdit
I loved ResEdit when I was a kid. Apple’s resource editor let you poke your nose into most system files and applications, revealing image assets, icons, interface elements and plenty of other technical goodies I didn’t really grok at the time. It was surprisingly deep, including a little MacPaint-like editor for the icon files along with a drag-and-drop interface editor. At the instigation of David Pogue and Joseph Schorr, I recall using it to make the bloated trash can look filthy and overflowing.
Gaming
No exploration of Mac history would be complete without a look at some of the platform’s greater gems of gaming. PC’s may have had more games by volume but the Mac didn’t have any shortage of fun, either.
Escape Velocity
I sunk so many hours into EV, it’s not even funny. A nerd who grew up on Star Trek and other scifi, I found this game’s premise of space exploration, commodity trading, secret missions and interstellar combat extremely compelling. Entire Saturdays vanished into its gaping maw.
Marathon 2
Before Halo, Bungie made Marathon. It was a rich story of treachery and tragedy among the stars. Crazy AIs and three-eye aliens all trying to get you killed while you blast things with enormous guns. No full-motion video cinematics here, though. If you wanted story, you had to read.
SimCity 2000
I was terrible at SimCity. My budget rarely balanced, my people always complained.
I loved it anyway. SimCity 2000 is still surprisingly playable, too. Definitely a timeless piece of work.
End of an Era
The way many expected the System 7.5 era to end was pretty bleak: Apple collapses, the Mac dies, and its software and hardware begin to decay into uselessness.
Of course, history went a different way. I’m glad that Apple survived long enough to ship Mac OS 7.6 and OS 8, that in the time since Apple has rebuilt itself into the juggernaut of its industry. Mac OS X beats the hell out of anything that came before it. I still remember picking up my copy of Macworld at the supermarket and learning how Apple bought NeXT – and hoping that the future would bring brighter days for everyone’s favorite “beleaguered” company. And it did.
Still, I’ll always look back with fondness on those days of innocence before a Unix shell was a keystroke away, before every UI interaction was beautifully animated, before we measured even the tiniest of hard drives in gigabytes, before collaborative multi-tasking and protected memory. When using the computer was new and exhilarating. When the Mac was more than just tool – when it was an escape to another realm of existence. Those were the days when a little boy, without coming anywhere close to realizing it, laid the groundwork for all the wonderfully fun things he’d get to do years later as a man. I learned way more from my Mac than school ever gave me.
Thanks for the memories, Apple.
Hipmunk for iPhone
My awesome new job and how I built our awesome new app. Check it out!
The app even won Best Design at LAUNCH 2011.
Ideals are Opportunities in Disguise
Idealism gets a raw deal. At least, it gets me a raw deal. Years ago, I was sitting around a table with a bunch of people at least ten years my senior. Social media, that old chestnut, was giving our company trouble. People kept using it to complain. It hit me like a bolt of lightning:
“What if we committed to overhauling our culture so that the customer always, always, always came first in our processes and our perceptions? Then people would stop falling through the cracks and getting pissed off on the internet. And word of mouth would get even bigger for us!”
Everyone looked at me like I was an alien.
Sometimes it’s delivered with a sneer, other times exhaustion, and occasionally, there’s even contempt:
“That’s idealistic.”
Ideals, it seems, are academic contrivances that hinder How Real Business Gets Done.
I can’t escape my idealism. Sure, I’ve launched v1 before it was perfect, accepted a minor bug or two in a release, fine. But at no point have I ever sacrificed the core of the user experience to any other cause. User experience is the compass by which I judge every decision.
I configure my values this way because I’ve seen first hand how powerful it can be. Not just in software, in web applications, in innovative, industry-changing businesses…
But also: in dog grooming.
I got my idealism from my mom. Whatever town she’s in, she’s the best dog groomer there is. After years of working for stupid, short-sighted shops, she set her sights on a business of her own. With nothing more than a GED and her ideals, she renovated a space and got to work.
And thrived.
The biggest challenge to scaling her wildly successful business? Finding people who were skilled enough to match her quality of work or genuine love of animals. It was impossible.
My mom had two options: hire on people she knew weren’t up to her standards or stay the size she was. She wanted growth – who doesn’t? But she knew she couldn’t just hire crappy people. Her shop’s growing reputation was built entirely on her quality of work. People loved the idealism that inspired outrageous standards of hygiene for the facility. People whose dogs usually couldn’t stand going to the groomers suddenly lost their fear, because for the first time grooming meant being treated humanely.
Ideals had created differentiation. Bad people would destroy that progress. In the short term, yes, her bandwidth would increase and more dogs could come through the shop. In the long term? She’d be just another grooming shop with tepid business – or no business. The worst part of all, I know now: she wouldn’t be proud of her shop anymore.
So she chose secret option C: Open a grooming school.
In hindsight, of course, this is obvious. It wasn’t then. It was risky. It cost a lot of time and effort to get licensing to train. Putting together course materials and a curriculum is a very different skill set than grooming dogs. Shifting from spending all your time grooming to most of your time teaching? Very difficult.
But it worked.
The revenue from a steady stream of students smoothed out an otherwise highly cyclical business. The option to have dogs groomed by students opened the shop up to new clients who had been unable to afford the previous up-market rates. Constant oversight meant even inexperienced groomers were sweating the details and doing things right. Daily bandwidth increased dramatically with only a marginal impact on quality. Best of all, when a star pupil came through the program, they could be immediately recruited after they finished training.
There were hiccups – students could definitely botch their work at times, but the risk was baked into the price, so it didn’t harm reputation. Picky clients could opt out of the student work at the old rates, and many did. Overall, everyone was happy, including the many animal lovers who discovered how to make dogs part of their professional lives through grooming.
True to form, my mom found a way to have her cake and eat it, too: way more money without sacrificing the quality of her work.
None of this would have been possible had ideals not played a huge role in making decisions. Absent ideals, I’m not even sure she would have gone to work for herself.
I can’t escape my idealism. And I don’t want to. My ideals are a map to build trust, solve problems and, in some small way, make the world a better place. The only article of faith I have is that, with a bit of work, that map leads to success. And in the end, without my ideals, I couldn’t build software, or anything, and enjoy it.
There are limits. You can’t pay for a sandwich with a song. Idealism is not a business model. Idealism is a tool. It’s a fulcrum for making difficult decisions and your flashlight in the darkness of ambiguity. It helps you understand the success conditions for every move you make.
I won’t stop putting the user first and neither should you. Next time someone dismisses your idealism, look very hard: an opportunity could be lurking across the bridge they won’t cross.
Improve revenue by dicking your users
It’s sometimes pointed out to me that my idealism around the user experience is inherently flawed. One day, the reasoning goes, rubber will meet the road for any company and it’s going to be necessary to do something to gain revenue at the expense of making the user happy.
And I guess it’s true. I mean, consider:
There’s Blockbuster. Keeping a broad inventory is a lot of work and expense. It’s easier, and more favorable to revenue, to stock only the most popular stuff. Also, you can definitely make a ton of money by charging late fees.
Hmm. The only problem there is that Blockbuster just filed for bankruptcy.
Okay, okay, that’s fine. How about Comcast? Having installers permanently on the payroll is a pain in the ass – paid time off, benefits, training costs, ugh. Outsource that action, let someone else do the worrying instead. Sure, these techs won’t care about the company culture (such as it is), and since they get paid by the installation, they won’t care about conducting business in a way that leads to a long-term positive opinion of Comcast. There will be less oversight, so they might screw up in ways that are embarrassing. Time management could be challenging for these local outfits and people might be late for appointments… But – revenue!
I guess the wrinkle is that kind of thinking tarnished Comcast’s brand so severely, they had to change the name of their consumer service. Maybe customer perception had nothing to do with it – rebranding is fun and it can’t cost much, right? Any long-established brand would want to do it, eh? Maybe not so much.
Fine, how about Yahoo? They made a really great play – push the portal angle really hard, don’t focus too much on search. I mean, if search works too well, people won’t stay in the portal and then how can you monetize all these millions of eyeballs? Nah, display ads. That’s where it’s at. Sell banners by the boatload. Bulk up that ad sales team!
That only worked until the dot-com bust, though. Now Yahoo’s market capitalization is a tenth of its biggest competitor, Google.
So maybe dicking your users isn’t all it’s cracked up to be, assuming you want prosperity to continue more than the next few quarters.
What about being good to users? How does providing an outstanding user experience change things?
Turns out the iPad is the fastest-selling non-phone product. Despite the fact that, as shipped, the iPad couldn’t print, can’t use Flash, and doesn’t have a camera, people are buying it in droves. 4.5 million units sold in the first quarter it was available. Maybe “being hip” is suddenly important to the broad cross-section of consumers who are buying it, and they have been convinced that upwards of $500 is a fair price for the privilege. More likely, though, is that a focused, task-oriented, touch-based interaction scheme where no one nags you about software updates is more enjoyable and intuitive than a netbook.
Apple is one of the largest companies in the world. Their focus on the user is not limited to the iPad.
Then there’s Zappos. Their values include being good to everyone – customers, employees and vendors alike. Their website has been consistently great for exploring their inventory and making informed decisions about shoes before buying. High quality images, easy to use filtering, detail-packed user reviews, all of it conspires to make purchasing easy. When you get on the phone with their customer service folks, you’ll find people empowered to help without rushing you back off the phone. They’ve long refused to outsource any activity that’s core to their business, including customer service and their fulfillment operations. They want to make sure these user-facing elements of their business are air-tight. This isn’t cheap.
Neither is the billion dollars Amazon spent to buy Zappos last year.
Then there’s Google. Say what you want about their creepy ways, Google revolutionized search. They made it work extraordinarily well, made it focused and made it fast. They’ve invested huge amounts of money on infrastructure to make sure their service is as snappy as possible. Instead of display ads, which would have been the easy but user hostile approach to making money from their traffic, Google borrowed Overture’s Pay-Per-Click advertising model. Paid search ads are perhaps the only form of genuinely useful ads for the user. They can actually solve the problem of your search.
At their IPO, Google shares could be had for around $85, already a respectable price. Today, they fetch over $500.
Creating a good user experience is important. It builds goodwill between your company and your users, yes. But much more importantly, it compels you to make a better product. Constantly re-evaluating your product for the benefit of its users future proofs your business. Look at Netflix, busy obsoleting itself by pioneering living room streaming. When you care about doing things well, your business moves at pace that’s very difficult to overtake. You’re a moving target and your products become much harder to compete with.
So can you dick over your users to goose your revenues? Absolutely. There’s a lot of short term juice in alienating the people you need most. Unfortunately, money is an addicting, distracting force. Before you know it, you’ll be dependent on the cash your user hostile approach to product requires. Ask Yahoo.
Anyone playing the long term game should approach the problem differently. Do it right and you’ve got the potential for a billion dollar business. Even if you never get there, gushing praise from your users is a lot more fun, and profitable, than simmering rage.