I ended up writing my reflection on the future of user-centered design rather like a blog post. Since I have to use most of my writing energies for school these days, posting my homework is probably my best chance for having new blog content!

My Apple bias is about to show through again, but it’s because I really think Apple is one of very few companies getting it really right on a really large scale. iPhone was the first successful device that really, truly put the user before the technology. Plenty of stuff that was possible (copy and paste, third-party development, &c.) wasn’t done, because it would have compromised the user experience. Heaps of legacy tech that people thought was crucial (hardware keyboard, stylus, removable battery) was instead completely jettisoned to make way for a better experience. The point of all this Apple fanboying is that it seems obvious to me that iPhone and iPad are the beginning of a new generation of experiences that eschew traditionprioritize the experience, and exhibit good taste.

Eschewing tradition is interesting to consider in light of the Sears & Jacko chapter. It’s pretty fascinating to see people pontificating about the future of software design in the couple of years just before the iPhone was unveiled. Much of the speculation in these interviews focused on improvements in existing, familiar devices like cameras, music players, or PDAs — things that fewer and fewer people need to carry around anymore, thanks to iPhone-style smartphones. Of course we will have fewer devices, and of course they will get smaller. Stephanidis mentions “specialized devices designed specifically for everyday activities”, but it seems to me that more and more of your interaction with technology will happen through a single device. Not touchscreens and dedicated, disparate UIs on every appliance, but instead apps that talk to those appliances, all on one phone, all ahdering to the same UI guidelines. Of course, this doesn’t mean that iPhones are our future — iPhones are great for now, but a future of “pictures under glass” (as described in Bret Victor’s marvelous rant) is a short-sighted one. Hopefully something much more tactile, more visceral, and more present is our future.

Prioritizing the experience is harder than it sounds, because it deals with cooperation across disciplines that is still pretty rare to find. In the case of iPhone, the software and hardware engineering went out of its way to avoid taxing the system so that gestural interactions would never, ever skip or bog down  — thinking about which led me to compose this tweet: “Technology that does not constantly push its limits will not constantly remind you of its limits. Instead it feels magic. Restraint wins.” Companies that understand prioritizing the experience will get more common. And the ACM article makes it clear that it will be harder and harder to get away with bad UX. Customer satisfaction depends on it; people no longer have to settle for crappy experiences. Coming from a company that prioritizes good design, someone still trying to argue against HCDE practices seems to be saying “wait a minute, are you trying to tell me that we need to make our products good?”

Exhibiting good taste is my interpretation of the thrust of the Norman article. (I should mention here that I think Don Norman is one of the coolest people of all time.) Don’t mindlessly implement whatever users ask for. Don’t assume that the user has a far better idea than your team does of how the software ought to behave. Instead, put your expertise and experience to work and offer them something great. Let them adapt to it a bit, rather than trying to create something that will adapt to them every time. As I have put it in my talks on designing for iOS, “sometimes it is more thoughtful to ask people to learn how your app works than ask them to decide how your app works.”

I got into a Twitter debate with Lukas Mathis, my favorite writer about software design. It is really hard to have a proper debate on Twitter! I thought I should lay out my thoughts, and my understanding of Lukas’s position, here.

Mac OS X Lion introduced a new type of trackpad scrolling that works the way iOS does: you drag as if the content is on a piece of paper and you are sliding it along. But you can still switch back to the old way, which I can barely believe anyone ever thought felt natural: you drag in the direction you want the viewport to move — even though the viewport doesn’t actually visibly move on the screen! And really, if you need to use the word “viewport” to describe how something works, it can’t be very natural!

Apple, too, felt strong enough about this basic interaction to solidify the consistency across platforms and change to content-scrolling on the Mac. In my case, it took about a day or two to get used to the new way. I even installed Scrollvetica, by my coworker Jim Correia, to make Snow Leopard work the same way. But I am of course an Apple fan and a software designer. So I was eager to switch to the future of scrolling and do things “right”.

But there are likely to be many confused and frustrated people out there, discovering that the viewport-scrolling they’ve gotten used to over the years has been reversed. Shouldn’t they be able to go back? (I do love the way Apple makes you turn off a setting labeled “Natural” in order to go back.)

And here is where Lukas and I disagree. He says that scrolling direction should not be a preference setting at all. We’ve figured out the right way of doing things, and we should encourage people to learn the new way. Just like how, for example, we make people learn to use Mission Control instead of Exposé.

In general, I agree that software creators should find the best way of doing something and ask people to learn, rather than finding several decent ways and asking people to decide. But there are two cases where I think preferences make sense:

One, dealing with technical differences in people’s setups. Sometimes you just have to work differently because you are interoperating with different types of servers, running on different hardware, and so on.

And two, dealing with abstracted input devices. Any input, other than something direct like a touch screen, has some sort of mapping going on. A keyboard layout is the way you want the keys to map to symbols on the screen. Keyboard shortcuts are how you want combinations of keys to map to commands. And scroll direction is how you want trackpad drags to map to moving content. These mappings are to some degree arbitrary, and people build up their own understanding of each mapping over time. The more time they have spent, the more painful the eventual change is going to be.

In the case of most changes, like onscreen controls moving or Mission Control replacing Exposé, people can see the change and adjust their interaction accordingly. But having a fundamental input change is frustrating; there is no change to notice until you have already made the mistake. The way you give input feels more personal, because it’s about how the computer reacts to you, not just about how the computer behaves. When something goes wrong, it seems like the computer thinks it’s your fault.

Of course, scrolling direction is easier to relearn than a keyboard layout. But if you are keen on taking advantage of Lion improvements today, I don’t think you should be required to go through that input frustration today. Encouraged, yeah. But not required.

I pasted a section of the marvelous Chicago Manual of Style to someone, to defend my use of a conjunction at the beginning of a sentence. It denounced the commonly-taught rule that forbids such use, and included this line:

In fact, a substantial percentage (often as many as 10 percent) of the sentences in first-rate writing begin with conjunctions.

Understandably, the answer that came back included this:

This is really confusing to me … because it seems to contradict the idea that if a word is frequently misused over time it becomes correct […]

Also, is there a definition for “first rate writing”?

It was a thrill to try writing a compact but reasonably thorough and accurate response:

I think we are stuck in the common conflation of language-related disciplines. Grammar, orthography, lexicography, usage, and style are different things, and they result in different kinds of advice.

Lexicography is a research discipline, so it tends to be almost totally descriptive — people who compile dictionaries just study how people use language and document it. So yes, the commonly recognized meanings of words in everyday use can and do change over time.

But usage and style are closer to art, and are more prescriptive — people who write usage and style guides try to explain how to use language as effectively as possible. Lexicography can help, but it can’t be the only factor. As with any (functional) art, one of the most reliable ways to know the right thing to do is to study respected artists (“first-rate writers”) as precedents: renowned novelists, esteemed journalists, cultural icons…

Of course, judging what is a good or bad precedent is fuzzy and difficult and controversial, and people can spend their whole lives trying to figure it out. That’s why we have all these disciplines! And reference books, for when we can’t spend our whole lives doing the research ourselves.

Here’s a software-design analogy. You could go out and describe how most software is designed, but your results would have nothing to do with how the best software is designed. Almost everyone uses a floppy disk icon to mean “save”, so it is widely recognized — but that doesn’t mean it’s good design. Likewise, Apple might do something unprecedented, like replacing OK/Cancel dialogs with single-button popovers on iPad — but that doesn’t mean it’s bad design.

…Wow, that was fun to write. LANGUAGE!

Popular blog reaction: “Yes, that’s right, a non-insignificant number of teenagers in America do not know who Osama bin Laden is.”

Typical tweet: “Amazing. Many American teenagers apparently have no idea who Osama bin Laden is.”

Correct reaction: According to a blog post (which is questionably-written enough to repeatedly use the figure “100,00%”), among people who use Yahoo search, who searched for something about Osama bin Laden on May 1st 2011, who phrased their search in the form of a question, the sixth most common question they searched for was “who is osama bin laden”, which amounts to an unknown number of searches, and which is unrelated to whether they have any idea who Osama bin Laden is; and two-thirds of people who searched for that question, who searched while logged in to Yahoo, who might be in the USA or maybe anywhere else in the world, are supposedly teenagers, according to their Yahoo demographic data, which may or may not be correct.

WOW SO DUM AMERIKAR

I did another one of these random album cover designs, because the meme is going around again.

Orchestras in Romania.jpg

Create your own band and debut album cover randomly

To Do This:

1 - http://en.wikipedia.org/wiki/Special:Random – The first random Wikipedia article that comes up is the name of your band.

2 – http://www.quotationspage.com/random.php3 – The last four or five words of the very LAST quote on the page is the title of your first album.

3 – http://www.flickr.com/explore/interesting/7days – The third picture in the top row, no matter what it is, is your album cover.

4 – Use Photoshop or similar to put it all together.

5 – Post it to your preferred online outlet with this text in the “caption” or “comment” and TAG the friends you want to join in.

My year in 24 words, again. Writing it made me realize that 2010 was probably the greatest year yet. I can’t believe this all happened, let alone in such a short span of time.

Endless home improvement. iPad development scramble. Secrets. Silent Saturdays. At3. Personas. Science reading & listening. Job offer; raise. Briefcaseget. ScLl. Speaking engagement. Chaos;Head. Warmachine.