Wednesday, May 30, 2007

Technology Is Not The Answer

Article at Popular Mechanics website talks about how the US government spent 15 years and almost $500M dollars on military tech but apparently the soldiers don't really like it! Wow.

A soldier is quoted saying "There are a lot of things I'd never use in my position. It seems like a lot of excessive stuff."

From general feature and technology bloat to details like lack of consideration in equipment lag times, this could be a good example of techno-centric development that forgets to address the critical question of the user's experience and helping the soldiers actually work better instead of just working differently (and blowing R&D money).

Tuesday, May 22, 2007

Design Documents vs. HTML Prototyping

Garret Dimon wrote a fascinating article called Just Build It: HTML Prototyping and Agile Development over at Digital Web Magazine.

In the article, he makes the point that while design documents definitely have their place, prototyping in HTML or even AJAX-y stuff is worth considering under various conditions. It's filled with interesting stuff and reminds me to really find some time to get cranking on trying to advance in CSS and respectably wield JavaScript soon. And I had been hankering to learn more about design documents after reading a bunch of books touting their importance!

Sadly enough, I don't feel like there is a lot of good literature on how to make good HTML prototypes or design documents. Looks like it will come from interacting with other designers and getting the right experience.

Side note: I'm totally digging the writing style that Dimon's article uses where every small section has a heading. I am seeing it more and more on the web and in Seth Godin's "All Marketers Are Liars" it's notably prolific and handy. It sort of reminds me of scientific papers with many headings.

Sunday, May 20, 2007


Originally uploaded by ericpan.

A funny example of error prevention via pub signage.

The first sign "DO NOT EXIT" is at about eye level, followed by "UNLESS EVERYTHING IS ON FIRE" right below it and then at waist-level "JERK!"

This is the inside view of the main entrance to UCSD's Porter's Pub (I've always entered through there at least). There is a "back entrance" that they were using as the only entrance on Sungod where they had a guy standing and carding at the door instead of the usual carding at the bar.

Some people ran up to it expecting to exit through it and then after about 1 second ran the other way.

Monday, May 14, 2007

Windows Vista: Lacking in product conceptual integrity?

From the article Facing the full horror of Windows Vista at iTWire:

So far, Transit has been using Vista Business full-time for a fortnight. And so far, we've found nothing that works better than in Windows XP, dozens of things that are annoyingly different without being a functional improvement, and several things that work at best intermittently and at worst not at all. On the whole, we wish we'd never moved.

Is this what Alan Cooper meant in his book The Inmates Are Running The Asylum when a product can lack conceptual integrity if a team of competent designers don't do their homework, develop personas, create specifications and a storyboard?

Well, I actually remember Cooper bringing up conceptual integrity in the context of not letting users directly define the product with feature requests, etc. Instead, it seems to make more sense to first have a bottom-up research approach where data about the users turns into personas and models which then gives way to a top down design approach from there.

Having random feature injections in lieu of user/persona-driven design seems like a bottom-up design approach which could lead to clunky user experience. Bottom-up research guiding top-down design lends itself to focused, coherent user experiences. That's my take on it anyway.

All that research work is to presumably understand what the user aims to achieve and accomplish and then to allow that knowledge to guide design. A storyboard would show how Vista "works better than Windows XP" from a persona perspective and how the "dozens of things" that are "(annoyingly) different" would be actual functional improvements.

As for the "several things that work at best intermittently and at worst not at all," I'm wondering if that's the job for QA and usability testing.

I doubt Microsoft would embark on the production of Vista without "design due diligence" especially with their roster of notable and brilliant designers and researchers. I'd be interested in seeing what their data looked like and how it translated into product specification, interaction design, and the usability testing results.

It would be even more interesting as a case study if Microsoft technically conducted the entire design and development process appropriately and the iTWire comments are accurate. (I've never even tried Windows Vista so I have no bearing on the accuracy of the comments.)

Friday, May 11, 2007

Shipping Notifications: Gamefly vs. Netflix

I've been a Gamefly subscriber for a good while now and I've been generally pleased with the user experience. However, recently joining Netflix demonstrated that Gamefly could easily improve their shipping notifications.

Here is what Gamefly shipping notification looks like in my inbox:

Here is what Netflix shipping notification looks like in my inbox:

Notice that Gamefly tells me what has shipped while Netflix goes one step further and tells me what day to expect it. I think it's nontrivial that Netflix uses days of the week in the e-mail subject line notification. I'm going to receive it within one week for sure (usually within days). Most people know what day it is instead of what date it is.

Supplying a date in the subject line would needlessly increase the cognitive load for customers who just want to know "What am I getting and when will I get it?" I know today is Friday and if it's arriving Tuesday, with little effort I know that's 4 days from now. Additionally, knowing it comes on Tuesday includes connotations that may mean something to me.

For example, Tuesday is a weekday/workday; it's far from the weekend; I have weekly presentations to give on Wednesdays; and so on. All this information would help me to realize I won't get to watch it until Thursday or Friday and decide to let my neighbor borrow it.

Instead, if I see it is arriving on the 15th, I have to recall today is the 11th, the difference is 4 days, 4 days from today is Tuesday, and then the rest would be the same. Maybe a minor amount of friction but I can always appreciate less cognitive friction.

Here is what the content of the Gamefly shipping notification looks like:

Here is what the content of the Netflix shipping notification looks like:

Gamefly adds a "shipped on" which is much less useful that Netflix's "arriving on or around…" note. Kudos to Netflix for repeating that info anyway. I know that Gamefly has detailed records of how quickly different customers receive their games. So does Netflix. I'm glad Netflix puts that information to use and I'm hoping that Gamefly is working on it.

Sunday, May 06, 2007

Seven habits for junior designers

Chanpory Rith wrote an article Talent Isn't Everything over at Boxes and Arrows which has 7 interesting suggestions for junior designers to improve chances at success:

  1. Work quickly. Produce a lot.
  2. Attend to details.
  3. Be versatile.
  4. Make an effort to learn.
  5. Anticipate problems.
  6. Set goals.
  7. Display a positive attitude.

Of course, he goes into much more detail for each step than my copied list does.

What I enjoyed about the article is how he focused on practical details that junior level designers could pick up on such as focusing on getting your ideas and work out for others to see as opposed to spending significant extra time on trying to get it "perfect" since it would probably be more efficient to let other designers give feedback and advise.

I now realize that junior designers and senior designers have different skill sets and are expected to produce somewhat different things. I'm going to try and keep an eye out for information that could help me to learn more about what makes more sense for junior designers to concentrate on and what junior designers should expect to pick up or refine later as they transition into senior design positions.

I find that having realistic and clear expectations for myself and knowing what is expected of me by others helps to significantly reduce stress.

Saturday, May 05, 2007

Chris Bangle On Love and Trust… In Design

This video is great and I hope many people watch it. The talk is not only eye-opening in content but also in delivery. While BMW designer Chris Bangle illustrates the way designers work and the love they have for it and each other, I found myself wondering about all the things he didn't mention that would normally be mentioned in a story like this. How did they gather data? How did they negotiate things? Who was on the team? What happened to the engineers? And so on…

In the end, I realize that his choices regarding omission and inclusion of content were challenging me to think above all the methods and technicalities of design and think about what it means to be a designer. Bangle's mental provocation will be buzzing in my head for some time to come…

Wednesday, May 02, 2007

CHI 2007: Day 2

Today, I spent all day in a course at CHI taught by Susan Dray and David Siegel from consulting firm Dray & Associates. While it was a good course overall, the first half of the course was relatively slow and as the course progressed, topics seemed to become exponentially more useful to consider. Unfortunately, time became a factor and the parts I was most interested in were rushed through.

During one of the breaks for the day-long course, I met a coursemate from Stanford who did research for Sony in Paris in her last summer break as an undergrad. She took one day off to take this course and meet with her colleagues from the summer in Paris. In another break, I got up out of my seat and turned around to see a friend and fellow designer Anshuman. He also took the day off to take this course. They both seemed to also think the course was a tad too general for their tastes.

An unofficial recurring minor theme in the conference presentations and courses thus far is the defense of qualitative research against statistical significance. Challenges from engineers and managers seem to focus on concerns regarding sample size and a bunch of other factors in any experimental method. Each speaker has a different answer to it but the general idea is that statistical significance is irrelevant when discussing user studies. Speakers seem very adamant about it and I see what they are saying, but at the same time, I imagine it being much harder to defend in real life than they make it seem.

I've been wanting to talk to some of the groups of foreign attendees to see what HCI (and the industry of HCI-related design) is like in their countries. I ignorantly assume that many countries and their corresponding industries are behind us in what still seems like early stages for the USA. I also want to ask them which research labs, design firms, or companies from their countries they most admire. I'm pretty sure that would yield a bunch of interesting stuff to look at which would normally be much more tedious for someone like me to find out.

Being at a conference is oddly draining. I'll have to try save the rest of the ideas for future posts.