In Minnesota, they are currently recounting the ballots in the Senate race between Norm Coleman and Al Franken. As part of the process, they are re-analyzing thousands of ballots that were discarded during the initial optical-scan machine count due to problems with the ballots themselves — usually stray marks and incompletely-filled-in dots. Usually, a human being can figure out the “voter intent” when the machine failed in the initial count.

Sometimes, however, the voter intent isn’t all that clear. Minnesota Public Radio has put up photographs of some of the more interesting examples of disputed ballots. And, in a weird kind of democratic recursiveness, is asking visitors to vote on the votes!

This is especially fascinating to me as an interaction designer because it emphasizes yet again how our users are not at all the rational, predictable beings we sometimes assume they are. They’ll click on anything and everything, and will fill in forms with the weirdest stuff you can think of. These ballots remind me of the kind of weird stuff you see in usability testing sessions and when analyzing the actual information users input into, say, e-commerce order forms (putting their name in the zip code field, for example, or writing angry profane messages in the address field).

My favorite is this guy who seems to have voted for Al Franken, and then risked negating his vote by writing in “Lizard People” as a write-in candidate (thinking, presumably, that it won’t count because he didn’t also fill in the circle next to it). Mischievousness, civic service, and abject nihilism are clearly fighting for this guy’s very soul — and the outcome is clearly too close to call.

(Hopefully, this is one of the last election-themed posts I make for a while.)

• Leave a Note (3)

John Gruber invites public ridicule to the UI design of this file renaming application. And yes, it certainly looks terrible, like a stereotypical case study in haphazard, bloated, bad UI design.

But is the UI design really that bad? I’m not so sure.

As someone who has had to do all manner of complex and esoteric batch file renaming over my career, this tool looks pretty darn powerful to me. I’ve had to do some absurdly massive batch file renaming (hundreds or even thousands of files) about twenty times in my life, and every single time it’s been a unique and bizarre challenge.

This is one of those tools where, at a fundamental level, more features is better than less. Fewer features make this tool less likely to serve its core function: saving your ass during a freak once-in-a-lifetime file renaming emergency. Even the wacky features like renumbering the files using roman numerals doesn’t seem absurd to me. This tool is intended for people without ninja-like mastery of the black art of regex construction (and, for that matter, for office production workers who may not know how to program anything at all, for example a photo editor).

So if it’s not the number of features that’s bad, let’s focus on the UI itself. The signature quality of this design is that every single feature is shown on one page. There are no layers of dialog boxes, no multistep wizards. The fields are numbered, too, which seems to suggest the order the transformations are processed. The entire transformation is right there for the user to see, no surprises. What’s more, showing all the possible transformations on this one screen educates the user on what the application can actually do: No poking around through menus and manuals to find out what this app is capable of. It’s all right there.

My only substantive critique, in fact, would be aesthetic: the grid of the page and the layout of the form elements could be more pleasing and better organized from an information design perspective.

Is that the nut of Gruber’s critique? Or is he being a little hasty in his judgement here?

• Leave a Note (65)

If you’re in NYC this Thursday, you should come to see From Sketching to Experience, the first of Liz Danzico’s Dot Dot Dot series of small, informal (and free) lectures. These lectures are the ramp-up to SVA’s new MFA in Interaction Design program. I’ll be one of the speakers this Thursday, and am also going to be on the SVA faculty as well when the program starts in 2009.

This lecture should be fun — four speakers from different backgrounds in a short-and-sweet presentation format. Please join us!

From the official site:

Crossword constructors sketch in pencil before ever laying a grid down, filmmakers rely on storyboards before ever picking up a camera, and cartoonists go pane by pane before designing a strip. Before interaction designers go high fidelity, sketches provide the method by which they communicate process, flow, layout — and importantly, story. Learn from four practitioners of divergent backgrounds how they practice sketching to get to their final product.

Speakers
Join us for the first lecture to hear from:

• Tom Bodkin, design director, New York Times
• Jake Barton, founder and principal, Local Projects
• Andrew Sloat, graphic designer and videomaker
• Christopher Fahey, founding partner, Behavior; forthcoming faculty, MFA Interaction Design

Details
Thursday, November 13
6-8PM (RSVP)
White Rabbit
145 E Houston Street (View Map)
New York

• Leave a Note (1)

This is a website Behavior made for the Smithsonian’s American Museum of National History during the 2004 Presidential election campaign. It is the web companion for Vote: The Machinery of Democracy, an exhibition of artifacts from America’s long and colorful history of voting technologies.

It was a fascinating physical exhibition. And I’m still proud of our interactive exhibition, too (both the Flash and HTML versions). It’s just as relevant today as it was four years ago in our first post-chad Presidential election (although the interactive map is a little out of date by now — to see a more up-to-date but less-detailed view of how voting technologies are distributed today, use the SciFi Channel’s new map from their surprisingly-good coverage of this topic).

As tomorrow’s historic election approaches, and many Americans will find this year’s voting user experience to be different than what they might be used to, I think it’s helpful to reflect upon the long and complex political and technological paths that got us here. For some colorful tales from yesteryear, see this New Yorker article from just last week. And it’s probably a good idea to learn where we are today: this New York Times article from January 2008 is a great place to start.

The Politics of User Experience

I just want to add one more thing, speaking as a user experience design professional:

It is a profound embarrassment to our profession that touch-screen voting exists at all in our country. It is an inferior system in every thinkable way, including — right up there with reliability, security, and cost — pure empirical usability. User experience experts should be up in arms over the very existence of these machines. Except perhaps for assisting with certain special needs voters, there is no excuse for a state government to have purchased these machines except for either old-fashioned corruption or a sad, abject gullibility for slick marketing presentations by election machine company salespeople. We Americans who call ourselves usability advocates should make it a goal to rid the USA of these machines by the 2012 elections. Who’s with me?

• Leave a Note (6)

Last night was the second U.S. Presidential Debate. One of my favorite new election-based interactive user experiences (in addition to CNN’s approval graphs and CNN’s magic wall) is Twitter’s election.twitter.com. Here’s a sample of what you would have seen during last night’s debate:

TechCrunch reviewed this site last week. Here’s their review in its entirety:

It sounds like a decent idea on paper: take every tweet about the candidates and stream them on a single, constantly updated site. Unfortunately, while it may be fun to look at for a few minutes, election.twitter.com is far too noisy to be worthwhile. There are no cohesive threads of arguments, and every quote that raises an eyebrow gets repeated ad nauseum. Verdict: Vetoed.

I completely disagree with TC’s analysis. It seems focused on extracting actionable, accurate, or even just coherent information. They’re disappointed in the quality of individual posts and the lack of consistent dialogue. It’s just “noise” to them.

Interestingly, the critique is identical to the initial criticism many people have of Twitter, before they actually try it and, hopefully, “get it”. They say it’s “just noise.” They say “who cares about everyone’s mundane, idle thoughts?”

Similarly, the nature and mechanism of election.twitter.com’s social function is, like Twitter itself, ‘ambient‘: It’s about getting an informal, general sense of what’s happening, not about following specific threads and individual thoughts. Just like with ambient music, you’re not supposed to actually listen to it in an attempt to extract something specific (for example a catchy melody, or a telling quote).

The “hot topics” keywords at the top of the page, extracted from the totality of the current twitterstream, are also extremely revealing. If you only look at that part of the page you’re already getting a real feel for the memes currently in circulation. Bite-sized but potent.

Bottom line: the site is not about finding content with immediate, actionable value. It’s about visiting repeatedly, or watching it flow by in your peripheral vision, without paying close attention. The goal is to synchronize with a certain public pulse.

(This is, coincidentally, what the New York Times home page does for me, albeit at a slower pace. I want to see what everyone is seeing and talking about today, and I trust that the Times home page will show that in a single page view, even if I don’t actually click through to any articles. It’s not about diving deep — it’s just about the broad overview of the zeitgeist.)

This new Twitter feature — this “topic-focused channel”, or whatever they call it — is a great new idea that I’d love to see extended to other areas and topics. It’s ideal for live events like debates, election night, live TV, sporting events, etc. Also for conferences, or even for private groups (a much-requested feature Twitter hasn’t yet delivered on). The idea is that you are really paying attention to something else — Twitter is just the back channel, the pulse of the topic. I can’t wait to see future implementations of this simple but powerful view.

• Leave a Note (1)

One thing that’s been fun about watching the Presidential and Vice Presidential debates on CNN is that you get to also watch a scrolling EKG-like graph of how viewers are actually responding to what is being shown.

The methodology appears to involve a live audience, selected by CNN, to manipulate some sort of control that indicates how much they agree with what they are seeing and hearing. When the audience agrees, the meters go up; when they disagree, they go down. It can get a little confusing when, for example, one candidate says “My opponent wants to eat babies!” and causes the meters to go up — does that men the audience approves of eating babies, or that they agree with the the candidate’s indignation? Generally, I think, it’s the latter.

I also can’t attest to the scientific validity of the graph — there are far too many variables in play to put much stock in them (for example, How did CNN pick the audience? How big is the audience sample?). Still, it’s fun to try to find patterns in the graphs, in particular to examine places where different audiences responded differently to the same statements — or when audiences didn’t respond at all.

For the Presidential debate, three different graph lines tracked Democratic, Republican, and Independent voters. For the most part, the red and blue party lines reflected which candidate was speaking, but there were many times when they converged indicating broader agreement with what the candidate was saying.

• Whenever McCain spoke in generic patriotic platitudes (America is the greatest country on earth, etc.) the Republican line shot through the roof and the Democratic graph barely notched above a flatline. Either Democrats aren’t patriotic and Republicans are, or Democrats don’t fall for that kind of rhetoric (and Republicans do). I think it’s a little of both — part of the appeal of Democratic and liberal politics is a sense of independence from systems of authority. Tellingly, Independents did not respond to overt appeals to patriotism.
• When McCain spoke of victory in Iraq, and criticized Obama for what he thinks is a plan for precipitous withdrawal from Iraq, the Independents in the audience barely responded at all. But when Obama spoke of leaving Iraq, the Independent numbers shot up. Republicans and Democrats generally followed their candidates on this question, but I think I remember observing that even the Republican line failed to reflect much enthusiasm for McCain’s emphatic case that we needed to pursue victory in Iraq. People want the next President to get America out of Iraq.
• Whenever either candidate attacked the other, particularly for some obscure or feeble inside-baseball political gaffe, the lines would split along party lines… but the Independent line would tumble. Independents do not want to hear political attacks.

The Vice Presidential debate was even more fun, where CNN selected an audience of undecided voters in Ohio. In this case, the graph plotted only two lines: men and women. Presumably, because they are undecided, most of this audience was not partisan D or R. My observations from this round:

• Anytime either candidate mentioned the word “nuclear” (or, in Palin’s case, “nukular”), the line for men shot upwards — literally at the very moment the word was uttered, like some kind of magic button was pressed in their brains. Apparently men love them some nukes. Perhaps the graph is also something of a Geiger counter.
• Whenever Sarah Palin spoke about her track record and accomplishments in Alaska, both of the plot lines were flat. Apparently nobody gives a shit about her accomplishments in Alaska. Nobody gives a shit about Alaska. If this was intended to bolster her executive credentials, the CNN chart shows that it wasn’t working.
• When Palin spoke of representing “regular Joe Six Pack” Americans, the meters did go up. The back-slapping stuff works.
• When Palin spoke about Iran, she spoke forcefully and strongly, and had some scathing critique for Obama’s (frankly baffling to me) “no preconditions” position. But, interestingly, the audience meters didn’t go up for her. People don’t want to hear belligerence towards Iran.
• When the critical question was asked (one of Ifil’s only really good questions, I think) about how the candidates would handle ascending to the Oval Office if the President were to die, Palin’s chart was nearly flatline. This is one of the few moments in the debate where the audience was asked to seriously and realistically contemplate the possibility that Palin could be our President. Despite her likability, nobody wants to even think of a President Palin.
• When Biden spoke about the death of his wife and child, the graph surged. When he choked up, however, they dropped suddenly but rose back up again. It’s as if peoples’ bullshit detectors went off immediately, only to then declare a false alarm, that indeed his pain was real.
• When Biden went on the offensive against Palin, the numbers did not respond. Fortunately for him he rarely attacked Palin in the debate. Pundits on both sides criticized him for being too gentle, but the numbers, both on the graph and in the post-debate polls, speak for themselves. Biden’s no-attack “aikido” strategy was completely correct, and his self-discipline paid off.
• That said, when Biden went off on Dick Cheney, the women’s line on the meter went through the roof.
• When Palin said “Drill, baby, drill”, the women’s line dropped. Oops!
• Palin’s winking and cute you betchas and goshdarnits didn’t work, either. When she said she wasn’t going to answer the moderator’s questions, and when she joked that she was new to this campaign and hadn’t made many promises, both meter lines dropped. People may like her and may even strongly agree with some of her positions, but they don’t like Palin not taking things deadly seriously.
• When the candidates discussed kitchen-table economic issues, a clear gender gap emerged: Men gravitated very strongly to Palin’s championing of Republican platforms, tax cuts in particular; while women responded to Biden’s empathy for the effects of the economy on everyday budgetary challenges. My theory on this is simple: Men are idiots when it comes to practical money issues. Men have no idea how much consumer prices have gone up in the last year or two (grocery costs are up 10% this year); women know. Men think that they’re going to make over $250,000 next year; women know this is a delusion. Believe me, I fit both of these patterns.
• Overall, Biden seemed to have greater appeal to women than Palin did. His appeal to women often exceeded his appeal to men, too.

For both debates, the pundits rated the candidates’ performances roughly equally. And in both cases, the post-debate polls revealed commanding victory margins for the Democratic candidates. And, as should be clear from my analysis above, the CNN studio audiences were similarly less moved by the Republican arguments. Why the discrepancy between the pundits and the audience?

The GOP strategy seems to be to run the same campaign they’ve run for the last six elections. Normally this would work. But I think things are different now. The American people are different now.

The conclusion I get from these debates is the same as Bill Kristol’s: Americans simply aren’t buying what McCain is selling. It’s a different electoral zeitgeist than it was in 2000 (when America was complacent) or 2004 (when America was scared but cocky). Obama and Biden did not have to be circumspect about their positions (as, say, John Kerry felt obliged to be) because their positions are what the American people want right now. They don’t want America to be belligerent around the world. They’re okay with taxing the rich. They want something to be done about healthcare.

The pundits will continue to equivocate about the candidates’ performances in the debates, and will give McCain’s platform credence from a rhetorical perspective. But the American people are already giving plenty of indicators that they’re just not buying it.

• Leave a Note (6)