Tuesday, October 21, 2008

I never thought I'd see this

Hodgman doing a Ted.com video. I received his new book today, yay for Amazon.

Sunday, October 19, 2008

Don't downsize the innovators

Bob Sutton has a great post about downsizing. Do you layoff employees, and if so, who goes? Or do you hang on and, since they aren't as busy, have the employees do more professional development in anticipation of an upturn? He also restates the common HR caution about not just hiring people who are like you, and especially, don't lay off the innovators
A downturn can be an opportunity to get rid of incompetent people and, of course, destructive assholes. But beware of the evils of using layoffs as a reason to expel everyone in your organization who does not act, think, and look like everyone else -- beware that most of us are prone to hold an overly narrow image of a "good employee." As I show in Weird Ideas that Work, since we human-beings have powerful and positive emotional reactions to people who are "just like us," and equally powerful negative reactions to people who are "different," the hiring process in most organizations acts to "bring in the clones."
Anyway, maybe it's not innovative, but I though the new PC-Mac "bean counter" ad was pretty good, but probably not a impressive as Hodgman's singing in a previous ad.

Saturday, October 11, 2008

A clean cube is a happy cube?

A Money article is about "Audi's clean desk fetish". The clean desk versus clutter desk argument has been going on for a long time.
Some researchers, however, dispute the benefits of a spotless workplace. When Herman Miller (MLHR), an office furniture supplier, conducted an observational study of workplace organizational habits, they found that "filers" actually stored more useless information than their unkempt counterparts. The company identified a group of "work masters," or efficient employees, and reported that those staffers were more inclined towards piling than filing.

"When people place things on their desks, they're encoding information in the spatial connections and layers," says David Kirsh, a professor of cognitive science at UCSD. Kirsh, who studies workers in their natural settings, says many workers prefer to use a two dimensional surface. "If you disrupt that and force them to stack or file, you lose information."
Kirsch, that name sounds familiar doesn't it? Yes, he is the distributed cognition guy, along with Hutchins, whom I've talked about before. The famous paper (Fresno State people click here, everyone else here) is:
James Hollan, Edwin Hutchins, & David Kirsch (2000). Distributed Cognition: Toward a New Foundation for Human-Computer Interaction Research. ACM Transactions on Computer-Human Interaction, 7(2), pp. 174-196.
Sometimes the reason for a clean desk is a more about security (IBM used to be big on that) or safety, which is hard to argue against. But from a cognitive or productivity point of view, does a cluttered desk implies a cluttered mind?

Maybe not :)
"A clean desk isn't always the sign of a productive employee."


"In fact, a clean desk can hinder worker efficiency."

I love this guy.

The premise is that people use their desk space as an extension of their minds.

"The human mind, specifically short-term memory, has a limited capacity," Brand said. "It has seven, plus or minus two, 'chunks' available for storing things.

Friday, October 10, 2008

Deconstructing the airport

Pretty interesting talk by Paco Underhill about "how to remake air travel for the twenty-first century". Malcolm Gladwell hosts -- his final question to the speaker is whether the airport experience is inherently unpleasant :)

More Malc videos here.

Monday, October 06, 2008

When to do what

Jakob Nielsen's alertbox column does it again: a really nice description of "When to use which user experience research methods". It complements something I posted earlier about software engineering experimentation.

Sunday, October 05, 2008

Forgot one

Another thing that got me thinking about evidence-based-medicine (and evidence-based-software-engineering) was Bob Sutton's post about "... Ten Commandments for Minimizing Medical Errors":
He [Michael Guiliano] also cited some more general research showing that diagnostic error occurs in 10 to 25 percent cases in medicine in general... He went through many causes of these errors, but one I found especially interesting (in light of our emphasis on parents earlier in the morning) was research cited in Jerome Groopman's How Doctors Think, which discussed how badly doctors interact with and listen to patients -- including one study that found that the average doctor only waits 18 SECONDS before interrupting a patient who has begin to describe his or her symptoms.

It not only costs a lot and is inconvenient, it's not as good as it should be :(

A recent interview with Evan Handler (the actor) about his health care experiences
In his books "It's Only Temporary," and "Time on Fire," Handler wrote that during his months in the hospital, he was given intravenous drugs that were supposed to go to another patient, that nurses tried to give him medications his doctors had forbidden for him and that staff members refused to follow the hospital's posted hygiene precautions for immunosuppressed patients like himself.
and an October article in FastCompany
For patients, of course, getting what's been proven to work is nothing more than we expect from a Jiffy Lube. But according to a 2003 New England Journal of Medicine paper, only 55% of American patients get all the treatment that is generally accepted as necessary for their problems. To make sure the number at Geisinger is near 100%, surgeons, pre- and post-operation, face a computer screen that asks a set of questions: Is the patient on a beta blocker? A statin? Were antibiotics given at least 60 minutes before surgery and discontinued after 48 hours? A staffer sends an email query if there's no response to any of the 40 steps.
and Harold Thimbleby's article in interactions "Ignorance of Interaction Programming is Killing People" (watch the YouTube video of a medical interface) reminded me of the National Safety Council's "Odds of dying from injury" data.

What is your lifetime probability of fatality caused by "Complications of medical and surgical care and sequelae"? Take a moment to think about that. It's probably much worse than you think: 1 in 1,308. Pretty unbelievable. Although not as bad as lifetime odds for a transportation accident: 1 in 80. At NASA KSC I remember a briefing where we were told that the best guess of a fatal space shuttle accident (at that time) was about 1 in 84. More than you ever wanted to know about the PRAN (probability risk analysis number) here, and a quote:
In fact, the Shuttle’s real-world track record provides a bleaker assessment: The number of shuttle catastrophes (2) over the number of shuttle launches (114), places the statistical, historical catastrophe odds around one in 57, or 1.7%. This number is misleading, say NASA officials, because it doesn’t take into account lessons learned, and safety measures implemented to prevent similar accidents in the future
Changing the subject, computer security expert Bruce Schneier has a very interesting column on "The Seven Habits of Highly Ineffective Terrorists" in Wired:
Terrorists, he [Max Abrahms] writes, (1) attack civilians, a policy that has a lousy track record of convincing those civilians to give the terrorists what they want; (2) treat terrorism as a first resort, not a last resort, failing to embrace nonviolent alternatives like elections; (3) don't compromise with their target country, even when those compromises are in their best interest politically; (4) have protean political platforms, which regularly, and sometimes radically, change; (5) often engage in anonymous attacks, which precludes the target countries making political concessions to them; (6) regularly attack other terrorist groups with the same political platform; and (7) resist disbanding, even when they consistently fail to achieve their political objectives or when their stated political objectives have been achieved.

Saturday, October 04, 2008

What is that?

Finally, a google maps application to answer the question "What can I see from here?" It actually accounts for the curvature of the earth, and refraction. Be sure to play around with the "visibility cloak" button. If you want to see how it's done, the FAQ is interesting.

Changing the subject, here are a couple of places for your next Fresno County vacation you can compare east side versus west side soaks: Mercey Hot Springs, and Mono Hot Springs. You might want to base the visits on which water appeals to you more: this or that.

Thursday, October 02, 2008

McConnell on agility, and far-out pictures

Earlier this week Steve McConnell had a webinar about agile software engineering, sponsored by the IEEE Computer Society. I usually can't stand webinars, but this one is good. McConnell knows his stuff (I've talked about him several times). The archive of the recent webinar is here.

Someone at extremetech.com is fascinated with pixels and took close-up pictures of HDTV screens. See the slide show here.

Finally, I am thinking about how to use Google's real-time analytics for YouTube videos. It might be useful for usability tests, but I'm not sure yet.