- On the Menu at Reader’s Digestion, a Literary Café (co-author: Jay Fanelli)
- Bars in Pittsburgh Whose Names Are Also Sex Acts
- Things That Are More Fun Now That My Cracked Rib Has Healed
- On The Menu at The Big One, a WWII-themed Restaurant (co-author: Jay Fanelli)
- Lists I’ve Failed to Place at McSweeney’s
Web
Intuitive Interfaces? No Such Thing
The way interface designers use the word “intuitive” has never set well with me. It’s a good way to get people to know why an interface works well, but it’s inaccurate. Over on Johnny Holland, Vicky Tenacki writes:
Digital devices can never be inherently ‘intuitive’, as the fact that they deal in abstraction automatically means that actions must be arbitrary. (An aside: for those who argue that much of gestural and time based interactions are intuitive, remember that this assumes a Western way of looking at space and time. Anthropologists would tell you that there are other ways.) In other words, interfaces aren’t ‘intuitive’, they’re ‘intuited’: before that, there’s nothing ‘intuitive’ about them at all.
Agreed. I don’t think we should stop using the word, but it’s worth pausing once in a while to recognize that the most “intuitive” interfaces only seem so because they adhere to (or successfully predict) a set of conventions and expectations and behaviors.
When you get as close to cognitive psychology as interface designers should, “intuition” takes on a technical meaning: immediate knowledge that comes from someplace besides conscious reasoning, however much the process might seem conscious to the subject. (That conscious / unconscious contradiction helps explain, among other things, the most effective American politics of the last half-century and the human compulsion to believe in the supernatural.)
When we call an interface “intuitive,” though, we really mean that it’s easy to figure out. That’s a very different claim about the brains and bodies we’re discussing.
Changing the Car Clock: Reflections for the User-Minded
Thanks to networked time servers, I only change one clock when daylight savings rolls around: the one in the car. Changing car stereo clocks usually works one of two ways; each one presents the human interactor with a different kind of challenge.
The “Find a Pushpin” model (FP, for short) requires the human to locate a pushpin or other narrow object with which to depress small, recessed buttons marked “H” and “M” (for reasons unrelated to my favorite place to buy clothes). We might think of the problem the FP model presents as requiring a primarily physical solution, insofar as we consider the buttons’ labels instantly intelligible to a human looking to change the displayed time.
The alternative model requires the human to determine a precise combination of holds and presses to apply to readily fingerable buttons. This model, which I’ll call the “Finish Him!” model (FH!), after Mortal Kombat match’s daunting final button-press sequences, presents a problem requiring a primarily cognitive solution, insofar as most humans in the target group come equipped with the requisite physical equipment.
In designing interactions for the Web, we don’t usually have the choice between physical and cognitive solutions. However, we do have the capacity to provide multiple physical mechanisms for accomplishing the same task, allowing for a similar increase in convenience for this or that group of users.
For example, Google Labs has an experiment that allows one to use keyboard shortcuts to move from entry to entry on a Google search results page. The result: Having just typed in a search query, I needn’t move my hands from the keyboard in order to find and open my preferred result. For a certain group of users (the Japanese of the “Google user” set, perhaps), this method—not pointing and clicking—will provide the greatest convenience.
Even limiting ourselves to a single physical mechanism (e.g., the mouse or trackpad), we can often provide multiple points of cognitive interaction with a particular kind of content. Most commonly, we tend to want to provide searchers with a reliable, easy-to-use search function and browsers with a clean, functional navigation.
Other, more complex examples still focus on the human user at the other end of the line. Syndication of a site’s content allows a certain set of users more control over the times and mechanisms of their interaction with that content. Likewise, the prevalence of the API for web services reflects the growing desire of a certain set of users to create their own points of interaction with the service.
Anecdotally, it has seemed to me that American cars favor FP, while Japanese cars favor FH!. (I can’t speak to German or Korean tastes.) This comes as no surprise; Japanese culture’s greater engagement with technological gadgets likely leads designers to think that the most convenient solution will be the one that requires a few moments of cogitation, rather than an open-ended search for a kind of object one isn’t likely to keep in one’s car.
In the States, though, our impatience with two-handed interactions with technology—an impatience that makes the iPhone wildly popular here but a dud in Japan—means that the greatest convenience will lie with a quick search for something small and pointy. Also, as a whole, we probably keep our cars a little messier over here.
These are generalizations of the broadest kind, but that’s how culture works: It makes generalizations true, mostly by seducing or tricking its members. (Maybe this works differently outside free-market capitalism.) And really, the automakers only can address their users on these levels: as humans, first, and, most narrowly, as members of a culture with demonstrable preferences and habits.
We in the Web often have an advantage over the automakers, in this respect. We produce sites and services specialized enough to allow us, through conjecture and research, to identify with some accuracy the salient characteristics and preferences of our user groups.
In the U.S., that is, the set of humans who have cars is impossibly diverse. Not so for set of humans who might visit, say, a regionally successful business school’s website, or the online banking section of a consumer lender. (Not so, also, for the set of humans who have luxury cars, one overlooked reason that so many interface enhancements happen in those vehicles first. [That is, it’s not just about money…])
To make the most of this advantage requires solid research of the kind I’m sure all automakers do. Happily, many of our clients have already hired firms with serious expertise in these areas. However, this is no excuse to treat our armchair speculations about the user base as gospel. Instead, we have to reconcile our intuitions about who these users are and what they want with a serious look at any research we’ve been given or can find or can create ourselves.
With great power (relative to the automakers) for knowledge of our users comes great responsibility. User groups are unruly, surprising things, to which we are, nonetheless, beholden. It’s up to us to respond to them as they exist, not as we’d prefer them to be (at worst, for selfish or idiosyncratic or idealistic reasons).
Where Do I Make this Joke?: Facebook-Twitter Integration & a Problem of Social Media
Last night, Jay had a few of us over to play cards. He tweeted:
Jay is waiting for the gentlemen to show up.
This morning, I @replied:
@thebristolkid Gentlemen? I guess you waited all night.
The joke’s a little predictable, a little modally antiquated, sure. But that’s not the point. I saw Jay’s status on Twitter, but I also saw it on Facebook. There, Jay has the Twitter app set his status when he tweets (as do I). So, his tweet about waiting for us to arrive updated his Facebook status more or less immediately.
The point, then, is this: The appearance of Jay’s message in more than one place—and the ways that such multiplicity positions me in relation to Jay and others—reveals one of the peculiarities of online social media, their simultaneously collaborative and competitive aspects.
We’ll get there by looking quickly at my reaction to seeing this message in many places and at the different roles of Twitter and Facebook status updates (in my life, at least). We’ll then consider the differences between online and offline mediators of our social relationships, and finally the two aspects of social media that motivate this post, perhaps best conceived as intersecting pressures on individual online services or platforms.
Perhaps an hour after I @tweeted Jay, when I got around to Facebooking this morning, I saw his status, and, in it, a missed opportunity. See, I meant my joke mostly for Jay, but like so much online comedy, I also meant it to have a public or semi-public aspect. That is, it might have been nice for someone else involved with our card game to see it and to respond. None of those people is on Twitter, though some are on Facebook.
Moreover, I might have liked for some friends who aren’t involved with our game, who may not even know Jay, to see the joke. I’m surely not the only one who’s identified a performative aspect to his social media behaviors; it seems to me that, among other ways, jokes like these function to complement or reinforce my meatspace persona. In a way, I want people who aren’t in on the joke to become in on the joke so that they can better understand who I am and how I think about the world. (Really.)
Given that I have far more friends on Facebook than I do followers on Twitter, should I have held off on making the joke on Twitter, and waited until I made my way to Facebook? Should I have re-posted it on Facebook, violating that strange but compelling prohibition on repeating a joke before the same audience (if not on repeating a joke at all, at least without acknowledging its travels)?
It might be nice if there were a way to tell the Twitter app on Facebook that, hey, that dude I follow on Twitter is this dude on Facebook. Then, it could ask that dude for permission to post my Twitter @replies as comments to his status messages, and we’d be off. (Such a configuration would no doubt require like nine databases and three APIs that don’t and maybe can’t exist.)
Then again, it might not be nice at all. The more I think about it, the more I think my uses for Twitter and for my Facebook status differ—or at least, should differ. Twitter “feels” more directed towards two-way communication—this despite the @reply’s absence from the service’s original conception (Twitter-blog post and napkin-sketch). One’s Facebook status, though it may draw replies, seems more about one-to-many transmission, less about soliciting a response.
My perception of this difference may derive from just how few Twitterers I follow, and how few follow me, as compared to my Facebook friends. That is, on Facebook, it would be as unlikely for others to comment regularly on my status as it is that I will comment on theirs; my Twitter circle is more intimate.
Still, I won’t remove the Twitter app; I continue to want my tweets reflected in my Facebook status. It’s not just that I’m lazy. It’s also that some of the value of these new-ish media forms derives from the knowledge that one has configured one’s digital life in ways that are compatible with those digital lives it intersects, at least for me. The upshot of my decision, though, is a lingering dissatisfaction with every tweet on the terms I’ve sketched here: Each is either too transmissional for Twitter or too solicitational (of a response) for Facebook.
However unique this situation, or idiosyncratic my response to it, it does belie a larger problem of social media. And it’s important that I don’t say “problem with”: I’m not criticizing, just critiquing.
For comparison’s sake, our offline interactions with each other are structured by lots of different institutions, among them school, work, coffee shop, club, public transportation, and movie theatre, depending on your inclinations.
The online institutions that mediate our relationships, if no less powerful, are far more transparent. I know more or less exactly how Facebook will shape my contact with Jay before I sign up, because Facebook has been designed with that purpose made explicit. This transparency creates a marketplace: We choose among competing platforms based on our conscious ideas and expectations about each.
This may be less true for very early adopters, to be fair; that a group of early Twitterers created the @reply without regard for the built-in purposes of the service—and that the service then assimilated the @reply—shows just how much power can come from the ground up. Still, at the moment in which one creates a Twitter account, I suspect the rule is that one expects to participate within the bounds of prescription.
An armchair Foucauldian analysis might tell you that each of the offline institutions works just the same, that spatial configurations and designations of authority have been implemented—perhaps no less by design—in order to shape social relations. Even if this is so, though, these functions are so obscured that it takes a Foucault to expose them in the first place. We participate in these institutions for reasons far more complex, entangled, and at times unconscious than those that motivate our signups online.
The problem, then, is this: The free-market competition among social media platforms and their underlying collaborational ideology—best demonstrated by the obligatority of the API—may be fundamentally incompatible, or at least, only partially resolved. The partial integration of Twitter and Facebook illustrates the conflict. The existence of a Twitter Facebook app is a foregone conclusion. Yet, in order to remain different, to remain competitive, the two can’t be integrated completely.
If my suggestion for a total (or more total) integration were implemented, the differences between tweeting and setting one’s Facebook status, between @replying and commenting, would be two: first, default privacy settings, which we can read as the developers’ ideas about who should see one platform’s version of the same message, and about how that behavioral aspect of the platform should shape our interactions with one another. Still, a user can (and many users do) override these settings on Twitter and, with considerably granular control, on Facebook.
Second, with more significant interplay between Twitter and Facebook, the choice of interface would remain as a difference. But even within either site, one already has dozens of options for tweeting or updating one’s status—not just from the computer, but from the phone, too. A list of applications for interacting with Twitter makes the head spin.
Finally, even in this hypothetical circumstance, brand recognition would still play. One benefit of affiliating oneself with a particular platform is that the affiliation is public. That is, for a variety of social reasons, one may want it known that one is a Twitterer. One may also be motivated professionally; I originally signed up for Facebook, for example, so as to be better able to advise higher education clients on its potential utility in their marketing efforts.
This last factor seems the most crucial difference that would remain, to be sure. But if Twitter and Facebook’s status feature were made virtually identical, I doubt that Twitter could remain important. That is, I think that although Twitter would continue to exist, it could not do so as more than a badge, an obligatory statement of one’s partaking in a certain broad movement.
The point I want to make here is just this: There are non-parallel forces—collaboration and competition—that press against social media platforms. Under certain circumstances, these forces can also come to press against us. In this context, we must look closely at the intertwining of our relationships and platforms. That is, if we’re better off after careful critiques of our classrooms, prisons, and cube farms, then we ought to derive similar benefits from thorough examination of our online institutions. I hope to have begun to provide a small example of how this might work.
Flashback to 1994
One of the early developers working on Netscape Navigator has put the 1994 version of the company’s website online. [via kottke.org]
One interesting surprise comes on the Netscape 1.0 product page:
IT’S CONSISTENT: Netscape is available – and is functionally identical – on Windows, Mac, and Unix X Window systems. This common look, feel, and behavior is a major advantage in mixed computing environments, where training and support are minimized.
This “feature”—an iron-fisted cross-platform consistency—dates the browser’s development considerably, in a more subtle (and more interesting) way than, say, the <b> tags in the original content or the Netscape products logo. To see how, compare it to Mozilla UE designer Alex Faaborg’s position on Firefox 3 from late last year:
Visual integration with Windows and OS X is our primary objective for the Firefox 3 refresh.
Faaborg goes on to detail their efforts on both operating systems—the post is more than 1700 words long—and to point to related discussions of Linux integration (significantly more complicated due to the variety of distributions available). The post has prompted 105 comments in the months between its appearance and this writing.
More to the point, taking just my home OS as an example, there are more integration-related bugs filed than I care to count, ranging from the minutest of interface details to heart-wrenching cries for help.
In short, the Firefox dev team has taken great pains to work towards seamless integration with a variety of operating systems, and the user community has shown great interest in this movement.
What has changed since Netscape’s early days to propel these integration efforts? My guess, an obvious one, is that we no longer live in a computer culture in which the biggest decision-makers about browsers are administrators of large-scale computing systems. Though many of us do use systems on which we have limited control over software, at work and in computer labs, we also use our home machines, on which we have free reign. And even on those limited-access computers, we may be presented with several browsers to choose from.
Evidence of this older model may be even more striking, in the pitch for Netscape 1.0 above, than the cross-platform consistency itself; the audience for the pitch is clearly a set of people in a position to impose use of a single browser on many other people. By contrast, one need only look at the background image on the Firefox website to determine the target audience: empowered single users participating in a consumer economy. 14 years is a long time, but my, how things have changed.
Don't Look Now 
