ARTS & FARCES internet

I’ve been working in my spare time over the past year or so to migrate major parts of this website from ExpressionEngine to a combination of WordPress and Apple’s OS X Server Wiki module. In the process, I’ve been actively making as many of the site’s templates as possible responsive.

After watching the server logs and analytics over the past year, I’ve seen the number of diverse methods of accessing our content steadily increase. And it’s all over the map: Desktops, notebooks, large tablets, small tablets, iPhones, and Android mobile devices of all sizes.

I subscribe to Ethan Marcotte‘s three key functions of responsive design.

1. Flexible grid
2. Flexible images
3. Different views for different contexts driven by media queries

It’s a lot harder than it looks, at least for me, mostly because I’m much more comfortable with content than I am with code (hence the move from ExpressionEngine to WordPress). I can do the coding, but I’m not the right tool for the job, as Mr. Natural would say. It takes me three to five times longer to do something than someone who dreams in code.

So, I’ve been plodding along trying to learn the innards of WordPress while coding a set of responsive templates. Then, out of seemingly nowhere, comes Josh Chan, writing for Six Revisions, telling us that responsive web design is not the future. Wait. What? Not the future? Hell, according to Chan, responsive design is just about as futuristic as Flash was a decade ago.

Chan even tips his hand when he writes, “I believe we should not write off other mobile development techniques and alternatives just yet.”

While I’m not sure I buy his server-side scripting argument, Chan is spot-on in his three basic criticisms of responsive design:

1. Content isn’t fully optimized for diverse devices (unless you use a mobile-first approach).
2. Performance isn’t optimized because it’s the same content with the same assets being pushed up the airwaves and down the wire.
3. Navigation usability issues abound unless a separate navigation structure is deployed for each device class.

Dave Winer nails it (yet again) with “iTunes is an outliner.” He also provides an excellent analysis of how iTunes got to be so bad.

I’m absolutely dumbfounded with all of the tech press fawning over iTunes 11—except for Farhad Manjoo writing for Slate, who says it’s time to take it out back and shoot it.

I have an incredibly large iTunes library on our internal server. It’s mostly music—of which I have subsets on my MacBook Pro and iPhone and of which my wife/business partner has subsets on her MacBook Pro, iPhone, and iPad. But also a lot of movies and TV shows for Apple TV. And books. And audio books. The entire library lives on the server because that’s where the disk space is.

When I first installed iTunes 11 I mistakenly clicked the Podcasts icon button and stumbled into the podcasts section—of which I only have a few but they’re important. I was ecstatic when I clicked the List scope button and saw a most reasonable outline:

• Title
• Episode

Imagine my disappointment when I eagerly clicked the Music icon button and wait, what? No List scope button. How can Apple’s user experience people possibly get the simplicity of using an outline in the podcasts section but not in the music section. Did the natural hierarchy just evade everyone?

• Artist
• Album
• Song

Really? And the Songs scope button is just a simple, non-hierarchical list, serving mainly to demonstrate how putting hierarchical content in a non-hierarchical container is um, problematic.

So, again, you’re right Dave. iTunes is an outline. Apple just doesn’t know it or desperately wants to suppress it. How else to explain that the company could/should have put an outliner in the Mac Toolbox in the late 1980s (he writes, double-checking to make sure that it’s really been that long; it has).

Update: Sunday, 2 December 2012 11:33AM CST: I’ve realized another brain-dead usability issue with iTunes 11. Music is a central part of my workflow, and while I’m sure it’s a generational thing, I listen to albums—not songs. Actually, I carefully curate my music collection and tend to listen to an artist’s entire catalog. Well, usually not an entire catalog because most artists release some real honkers—either to fulfill a contract obligation or an expriment gone bad or because it seemed like a good idea at the time but didn’t withstand the test of time.

I never, ever use shuffle because the artist and producer set the tunes in a specific order usually for very good reasons.

Back to my workflow. When I get into my office each morning, the first thing I do is decide which artist I want to listen to first. I then find the first album under the artist—which always defaults to alphabetical; a pity, but I don’t have date information for a lot of my music and the collection is so large it would be a monumental task to re-tag it. Most often, the artist’s catalog plays through and I’m jolted by the transition to the next artist in iTunes’ alphabetical order. Today’s Tab Benoit > Taj Mahal is tolerable; enjoyable even. Tomorrow’s Gillian Welch > Gov’t Mule is enough for conniptions.

The key to my workflow is that I almost always have music playing while I’m working and I don’t like being interrupted to have to tend to the music. Set it and forget it and I’m happy. This morning I discovered that iTunes no longer automatically plays the next album. Suddenly the music stops and I have to tend to it, manually adding the next album. Yes, I know I could add the next album(s) to the Up Next queue, but it’s still forcing me to do something manually that should be automatic.

Am I missing something, or is iTunes really that pathetic?

Update: Monday, 10 December 2012 11:33AM CST: It seems that every single day I’m confronted with another inexcusable major flaw with iTunes 11. The latest one: In order for the music tracks of an album to play in the correct order, they must be tagged in the correct order; having track numbers—like most reasonable people—just isn’t enough. This is almost certainly a play for Apple to increase its iTunes Store music sales. Ain’t going to happen here—Apple’s .m4a (MPEG-4 Part 14) file container format is lossy (when it could easily be lossless).

But that would require Apple actually caring about music—something it clearly doesn’t. If it did iTunes Match—another Apple abomination—wouldn’t try to overwrite my lossless music with its crappy 256KB lossy .m4a files.

When Tim Cook canned Scott Forstall and announced that Jonathan Ive would “provide leadership and direction for Human Interface (HI) across the company in addition to his longtime role as the leader of Industrial Design,” more than a few of us heaved a huge sigh of relief. Gone would be Apple’s non-functional skeuomorphic eye candy that existed solely for cosmetic purposes.

But all that is—at least so far—nothing more than wishful thinking according to Kontra, who has written the definitive article on Apple’s current design malaise. “Apple’s software problems aren’t dark linen, Corinthian leather, or torn paper,” writes Kontra. In fact, Apple’s software problems aren’t much about aesthetics at all… they are mostly about experience. To paraphrase Ive’s former boss, Apple’s software problems aren’t about how they look, but how they work.”

Ive is unquestionably the world’s best current industrial designer. But there’s no guarantee that industrial design expertise translates into user interface design, human factors, or usability. Moreover, as Kontra notes, whether or not Ive has the time to take on these additional duties is a great big fat unknown.

Kontra accurately observes that Apple’s vaunted Human Interface Guidelines (HIG) were more widely respected in the pre-web days:

“Today, not so much. A radio button or a checkbox can initiate a webpage navigation. Navigational menus now come in circular and triangular popups. There are now purely gestural UIs otherwise betraying none of their functionality to the user. Layers of sliding panels cascade on each other. List items slide left and right to initiate drill-down actions, up and down to reveal interactive media. Some UIs are beveled in 3D, some flat with no shadows, most a loose melange of many styles. And with each such ‘innovation’ they bury the notion of a once powerful HIG a foot deeper.”

This is a much deeper and more complex problem than most people—even human factors and usability practitioners—realize. For starters, there’s a separate, and not always complementary, set of human interface guidelines for iOS.

Here’s hoping Apple takes a really deep collective breath and remembers Dieter Rams’s ten principles of good design (which heavily influenced Ive):

1. Innovative
2. Useful
3. Aesthetic
4. Understandable
5. Unobtrusive
6. Honest
7. Long-lasting
8. Thorough
9. Environmentally friendly
10. Essential

There’s much hand-wringing happening currently on the intertubes about Apple’s skeuomorphic (attempts to replicate physical world objects’ ornamental design cues in a virtual form; in user interface designs, think trash can icons and torn paper calendar pages). Nicholas Gessler writes in “Skeuomorphs and Cultural Algorithms” that these ornamental design cues are almost always unnecessary and serve solely to make new designs “look comfortably old and familiar.” Austin Carr, writing for Fast Company‘s Co.Design, offers a clear definition: “Skeuomorphism is a catch-all term for when objects retain ornamental elements of past, derivative iterations—elements that are no longer necessary to the current objects’ functions.”

Skeuomorphic design is anything but new for Apple—its user interfaces have always tended toward skeuomorphic, and not just starting with the desktop metaphor of the first version of the Macintosh OS. Late versions of Apple ][ software programs, like AppleWorks, had skeuomorphic user-interfaces—something quite difficult and painful to accomplish on a non-graphical display. Apple’s past skeuomorphic interface designs existed in the service of existing cognitive models and making the user interface more intuitive and approachable. Although most users found the original Mac user interface confusing; after all, the trash can icon was used to both delete files and eject floppy disks. Over time, Apple’s skeuomorphism devolved into gratuitous theatrics of overly rendered textures and animations. “In many ways, the iPad book app feels like it was designed with the intention to look simply like a book, whereas it appears the Kindle was intended to feel like a book for those who love to read and want to,” writes Tom Hobbs for Fast Company‘s Co.Design.

Hobbs clearly identifies the pitfall inherent in relying on overly-skeuomorphic designs: “Ultimately, it encourages designers to become less critical and less inventive, which is detrimental to evolving new and improved solutions.” Carr cites one of Apple’s former senior user interface designers who worked closely with Steve Jobs as saying, “It’s visual masturbation. It’s like the designers are flexing their muscles to show you how good of a visual rendering they can do of a physical object. Who cares?”

Carr clearly articulates the problem with skeuomorphism from the perspective of users: “The issue is two-fold: First, that traditional visual metaphors no longer translate to modern users; and second, that excessive digital imitation of real-world objects creates confusion among users.”

What’s most surprising is that Carr reports that Jonathan Ive, Apple’s senior vice president of industrial design—who exudes more gravitas within the company than anyone other than Jobs’s ghost (Jobs was also said to be an enthusiastic supporter of skeuomorphism)—is apparently losing the skeuomorphism battle to Scott Forstall, Apple’s senior vice president of iOS software and NeXT alum.

Related links:

• Austin Carr’s interview for Fast Company‘s Co.Design with an unnamed former iPhone user interface designer at Apple

When I first started working on the web, I had a Sun workstation connected to the internet’s backbone through MRNet. Everyone thought I had blazing internet connectivity, but that wasn’t the case. In my home office I had dial-up, then ISDN, and then DSL. Later today I’ll investigate the cost of converting to cable. But the internet services I offered (FTP, Gopher, and web) were blazingly fast.

Back in the day, the rule of thumb among those of us that cared about usability (we were still called human factors practitioners back then) was that any resource (FTP site, Gopher hole, or web page) had to load in five seconds or less or the user would bail. In the 1960s time-share computers had to respond within 10 seconds to maintain the illusion of having one’s own computer.

Now Google engineers are saying that 400 milliseconds is the limit. “People will visit a web site less often if it is slower than a close competitor by more than 250 milliseconds (a millisecond is a thousandth of a second)” reports Steve Lohr, writing for the New York Times. Lohr cites Microsoft’s Harry Shum as saying, “Two hundred fifty milliseconds, either slower or faster, is close to the magic number now for competitive advantage on the web.”