noise to signal:
Design

This may only be funny to designers

But it had me falling off my chair. What if Saul Bass had designed the titles for Star Wars?

Time to show off something from the day job

As many of you know, my day job is as a software designer/interaction designer/product designer (so many bad names, so few good ones) at Microsoft. Part of my job for the last several years has been doing map design for the Virtual Earth team.

I'm not a trained cartographer, but it turned out that I have a pretty good set of skills for it — a strong sense of visual hierarchy, a head for complexity, and good background in the technical aspects of image display technologies. I've also been fortunate to work with a terrific set of programmers who are great at coming up with innovative solutions to the design problems I pose for them.

One of those design problems, which I've been driving for a couple of years now, has been to try to come as close as possible to the aesthetic qualities of paper maps, while still maintaining the usefulness and readability necessarily for online mapping tasks. Which is a nerdy way of saying "Make them look awesome but don't screw things up."

Today we finally released one of the big milestones we've been working towards, which means I get to toot our collective horn.

Here's the competition's map of Mt. Rainier national park. It's fine, but a little low on detail.

Here's our standard road map style, which was all that was available until today. We add the peaks, so you know where the mountain the park is named after actually is, and our labels have a bit more visual hierarchy applied.

And here's the version that went live on our site today. Yay! Hill-shaded maps!

Here are some other highlights to get you started looking around:

• The lava dome at Mt. St. Helens


• Sand Hill country in Nebraska


• Mt. Kilimanjaro


• The island of Viti Levu in Fiji

The top three mistakes Americans make about Graphic Design

Couldn't have said it better myself.

The problem with using cliché ideas

Every designer I know has a "swipe file" of brochures and posters and business cards and web sites and whatnot that they browse through when they're trying to solve a particularly tricky design problem. The idea isn't that they're going to steal something wholesale (although goodness knows that happens often enough, like the infinite thefts of Rodchenko's shouting woman design). The point is more that looking at interesting ideas sparks more interesting ideas that you then take in a new and different direction.

There's another kind of swipe file, though, that's more dangerous; it's the one that's in our heads all the time. It's where clichés live, which makes it both dangerous and easy to avoid. Easy to avoid, because one can simply look at a design and say "no, I've seen that somewhere before."

Dangerous, because if a designer goes ahead and uses a cliché anyway, unfortunate parallels may turn up. The movie on the left appeared on my Netflix home page and, man, I was sure I had seen it before. After thinking about it for a few days, I remembered the poster for the somewhat different movie on the right.

Milestones of information design

Which sounds sarcastic, but really, I do think this is a pretty damn clever solution.

Designer creates wall of breasts

A Dutch designer has created a wall of fake breasts to help male shoppers buy bras that fit their wives or girlfriends.

Wendy Rameckers works at the Piet Zwart Institute for Retail and Design in Rotterdam, reports Het Nieuwsblad.

"Most men have a selective memory," she explained. "They know all about their car, but never seem to know their wife's bra size.

"When trying to buy a sexy bra for their wife or girlfriend, usually they point to other women in the shop or, when asked about size, they say a 'handful'."

The wall consists of rows of silicon breasts in all sizes. By look and touch, male shoppers can work out the right size, she says.

(Via Jim Leftwich)

How about an editor?

I spent some time today -- over the course of a few breaks -- watching films in the Amazon.com/Tribeca Film Festival Short Film Competition. It's an interesting idea: filmmakers submit short films, Amazon customers watch and rate them, and they get winnowed down that way for the second round of competition.

The four films I watched over the course of the day were all basically entertaining, and exhibited a pretty good level of craft -- and mostly shot on film, which surprised me.

They all had the same problem, though: They all needed a damn editor. At roughly six minutes each, they all felt long.

I understand that these are films by people at the beginning of their careers, but it still seems as though they should have learned a better sense of pacing by the time they're producing films that look this good.

To me, a good short film is like a good short story: there's absolutely no time for extraneous material. Characters need to be defined in only the most essential ways. Ideas must be crystal clear, and they need to communicate immediately. If an idea needs to be shown more than once to communicate, there's something wrong. Over the course of six minutes, you've got to nail your concepts and sell them.

Great ads are good examples to follow -- they show just enough to communicate the idea, and no more; pacing has to be perfect, because they generally only have 30 seconds to tell the story. A good example is the shipping company ad I linked to back in July. Watch it; it's a great example of spare, perfect characterization and story. Within the first four seconds we know the character is bored, spending his days in a dark warehouse doing a repetitive job, and probably eats poorly -- and then, whoosh, we're on to the action of the story. This in contrast to two of the films I saw where the characters never developed any clear personality of their own.

I'm reminded of Brian Eno's story about the creation of the Windows sound for Windows 95:

"The idea came up at the time when I was completely bereft of ideas. I'd been working on my own music for a while and was quite lost, actually. And I really appreciated someone coming along and saying, 'Here's a specific problem -- solve it.' The thing from the agency said, 'We want a piece of music that is inspiring, universal, blah-blah, da-da-da, optimistic, futuristic, sentimental, emotional,' this whole list of adjectives, and then at the bottom it said 'and it must be 3 1/4 seconds long.' I thought this was so funny and an amazing thought to actually try to make a little piece of music. It's like making a tiny little jewel. In fact, I made 84 pieces. I got completely into this world of tiny, tiny little pieces of music. I was so sensitive to microseconds at the end of this that it really broke a logjam in my own work. Then when I'd finished that and I went back to working with pieces that were like three minutes long, it seemed like oceans of time."

Watch some for yourself -- I'm curious to hear whether a) this is a general thing, or I was just lucky, and b) whether I'm just a crabby old man before my time, and the films are just fine, thank you.

Of course, the next thing is for me to put my money where my mouth is and make my own damn 30 second film. (Even my silly little sketch clocks in at 44 seconds)

(and no, the irony that this is my longest post in months is not lost on me)

Edgar demonstrates the design expression "graphically strong"

We're really just tool-using creatures set loose in the modern jungle

(I should really be finishing writing a spec right now... it's the last thing I have to do before I head out for a long weekend... but I have to get this out of my head or I won't be able to focus any better than I have been for the last half hour.)

Yesterday, as my bus was sitting at a stoplight, having finally made it downtown through the rainy traffic, I watched as the bus driver (probably 5' 3" or so) pulled a two foot long stick out from behind her chair. She leaned waaaay over to the right and poked at the small fan sitting at the base of her front windshield, adjusting its angle so it blew air from the defrost more efficiently across the steamed-up window. She was pretty good with it, too, making subtle adjustments until it was just right.

As we neared my stop, I walked up to the front of the bus and leaned near her. "So, is that a Metro-issued stick, or is that your own invention?"

She grinned, a little embarassed. "Oh, it's mine. I made it from a duster. I can use it to change my route numbers, too!" She gestured to the keypad and display, high above the windshield. "That way I don't have to get out of my seat when it's not safe or I don't have time to stop completely, like when I'm late coming off of an earlier run."

We pulled up to my stop. I complimented her on her ingenuity, and got off the bus.

As I walked up to the gym, I of course started thinking about her solution from a design perspective. My first thought, as a designer, was to start redesigning her "cockpit". Surely there's some way to integrate the sign controls into the space around her? Surely there's some way to make the defrost mechanism more efficient, so it doesn't require the Metro mechanics to install little aftermarket fans at the base of the windows? And I walked along, thinking about new console designs and arrays of micro-jets around the edge of the windshield.

But then the anthropologist side of my obsessive personality took over. I love it when people have little tricks like that. To me it speaks directly to the fact that we are a tool-using species. The bus, one could argue, isn't a tool for her any more -- it's part of her environment, as much as the veldt or a watering hole would be to earlier hominids. The stick is a tool she invented to address her challenges in dealing with her environment.

I think of the myriad ways people adapt in their offices in the same way. On my main monitor right now, I have three pieces of paper taped to the edges that have information I always need to reference -- the ascii codes for certain special characters, the web-safe color values and their associate hexadecimal codes -- as well as three yellow stickies with information that hasn't quite graduated to a typed sheet (although it probably should since we're moving offices in a couple of weeks).

I'm also reminded of Miz Becky's mom, who had her kitchen re-done a couple of years ago. Being of modest height, she told the builder that she needed some way of getting to the top shelves. He crabbed a bit but eventually built her a very short, two-step staircase style stool -- just tall enough to let her get to the upper cabinets, short enough to be very stable, and nice enough to keep in the kitchen instead of needing to be put away. When he delivered it to their house, he admitted that while it was sitting in their shop, he sold three more to people who stopped by and said (all three, apparently) "Oh my, I've been looking for something exactly like that!"

So, fellow tool-using hominids, I'm curious: What clever solutions have you used that grossly enlarged forebrain of yours to come up with? How do you use tools to interact with your environment?

Sneak preview

To some extent, you can blame this blog on my Web Portfolio students eighteen months ago. You see, one night we were scheduled to start critiquing the site designs they had come up with, but nobody wanted to go first. To get things going, I offered to let them warm up by critiqueing my site. And warm up they did.

I was proud, really, that they had learned enough in the first five weeks of the class to lay a really solid and well-argued critique on me. But it did make me confront several problems that I had been trying to ignore for a long time.

The short version of the thrust of their critique is that my site had become this sprawling monstrosity of a site, very chatty and expansive. The problem was that it didn't do anything very well: it wasn't really a good day-to-day site, because that stuff was hidden away, but it wasn't a good portfolio site, either, because there was just way too much content.

All of this being a long-winded way of saying that I started this blog as the first part of that equation: separating out the chatty daily updates from the main portfolio site. And it's worked well at that task.

Unfortunately, I didn't follow through with the second part of the task: Making the other site more like a good portfolio site: A few galleries, 15-20 carefully selected images in each one, nothing extraneous. In fact, after starting this blog, I haven't updated the other site at all. The shoemakers kids etc. etc.

It's still getting tons of hits a day, but I'm always a little embarassed to send a potential client there because it's the virtual equivalent of a shoebox full of photos: "Here are some snapshots I took. Do ya like 'em? I'm real proffsh -- proshfes -- profsession -- uh, I'm good."

So this weekend I'm finally making steps towards building me a portfolio site. And there's a preview version online, if you want to check it out. (And for those of you reading this after the new site's gone live, I've maintained an archived version of the old site)

So now it's your turn. Granted there's only one gallery, so you can't really see how the site will feel more filled out, but take a look and let me know what you think. Becky was having trouble reading the intro text in the Naked Proof gallery, so I made it a bit thicker and less contrasty. The site's built entirely with CSS, so if you're looking at it on an older browser it may look a little weird. I'd like to know that, too -- how many of you are using those older browsers?

And hopefully this won't sit for another 18 months before I finish the job.

A form of validation

At the beginning of this year I taught the first, and so far last, term of my Web Portfolio class at the Photographic Center Northwest. In March, I wrote about some of the challenges of the class.

I later heard from the Executive Director that there had been some pretty harsh criticism of the class that unfortunately didn't make it into the evaluations -- primarily that some of the students thought there was too much focus on design, and not enough time spent working with the tools.

Probably a valid criticism, but when I was planning that first version of the class I decided that the design sections were hugely important. A portfolio site has to support the work. If it doesn't, that disconnect gets in the way of the work. So I spent a lot of time working with the students on refining their sites, pushing them to refine their designs as much as they could.

And it worked, from my point of view: Every site design not only reflected the individual photgraphers' work; it also reflected their design sense, not mine.

The point of all of this is that I got a beautiful note from one of the students from that class today:

Hello David, I wanted to let you know that my website is the featured site of the month at Photo Techniques Magazine newsletter. It was just posted today. I'm sending a link if you want to take a look, or it's under "portfolio" at www.phototechmag.com

Thank you again for all your help and design expertise in the challenge of getting the site to compliment the photos. I wish you very happy holidays. Best Regards, Connie

Ain't that the best?

In case you're reading this after this month, here's the direct link to Connie's site.

I am a depressed designer

Miz Becky's delightful sister stayed with us for Thanksgiving last week, and brought along her shiny new Apple Powerbook. I'd seen it a couple of times before, but this was the first time I'd spent much time around it.

One night as I was heading to the kitchen to get a drink of water, I noticed she had left the Powerbook on the dining room table, recharging.

As I walked by, a small white light near the rear center of the closed case caught my eye. I stopped to try to get a better look at it and noticed that it was pulsing. Gently. Slightly faster than, say, a breathing human. Whooooosh... Whoooooosh... Whoooooosh...

I guessed, and she later confirmed, that it was the indicator that the laptop was asleep (as opposed to off). You know. It's breathing.

That one, tiny, stupid little thing is more beautiful and elegant and intuitive and charming than anything, anything at all that we make here at Microsoft. And that's why, though people may use our products, they'll never love our products.

Customers don't customize

There's an ongoing debate in the software industry, which can be broadly summarized this way: One camp believes that most customers, given the option, will customize every aspect of their computer as a form of personal self-expression: color themes, desktop images, program skins, custom MyWay/My Yahoo/My MSN home page, and so forth.

The other camp (which your author is in) thinks that, although there is definitely a group of people who will be obsessive about customization, and despite the fact that some of us customize our environments (I can't stand the standard windows XP blue color scheme), that behavior isn't generalizable to the population at large. Most people seem to think of computers as useful tools at best, and any time spent fussing with customizing their computer is time taken away from the tasks they need to complete on the computer so they can go do whatever it is they really enjoy doing.

Anecdotal case in point (and the inspiration for this morning's rant): The Disney Tattoo Guy (one of this week's Yahoo Picks).

Mr. Reiger claims to be Disney's number one fan, and judging from his site I'd guess that's a pretty good claim. Not only does he have 1500+ tattoos of Disney characters covering 85% of his body, but his custom-built house is filled with Disney artifacts. He claims to have spent over US$900,000 on "his Disney love".

Imagine my delight, then, to find this piece of anecdotal evidence among the photos of his Disney-fied living room:

This is a guy who has more Disney knick-knacks in his house, and on his body, than Walt himself.

And yet... his computer, where he undoubtedly spends a lot of time (I'm guessing he's a big eBay user, as most collectors are these days) uses the default windows desktop and color scheme.

Thank you, your Honor, no further arguments.

A design success!

About a month ago, one of our marketers came to me and said "Dave! I've got a really fun project for you!"

Naturally all of the hair stood straight up on the back of my neck. Any experienced designer knows that the correct response to that statement is to point the requester to another designer. "Oh, I'm really busy right now, but I think Ann has a few minutes to spare."

"Fun", you see, is marketer-speak for "enormous time sucker with little or no payoff".

Unfortunately, I'm the only designer on our team, so the only person I could fob this task off on is... me. I sighed. "What is it, Steve?"

Steve explained to me that our group was going to sponsor a 10' x 10' carpet square at the Professional Developer's Conference, in LA this year, and that we could print anything we wanted on it.

After conferring with the vendor who was actually going to print the carpet, I found out that Steve wasn't kidding -- we really could print anything we wanted: 200 ppi, 24-bit color, any file format I wanted.

After some discussion, we settled on a map of LA with the conference center pointed out -- the best way to communicate the value of our products.

We went through some tribulations in getting it out -- and it did turn out to be a pretty big time sink -- but Steve's report from the first day of the PDC made it all worth it:

The full MapPoint carpet was uncovered today, and as I suspected yesterday, it looked great!

A number of attendees were seen using it as a paper map – standing on it pointing with their toes while planning where to meet or where a restaurant was. A new game was invented as well – Spatial Twister. “Left Foot Figeroa st.”. “Right hand 45th”.

Here's a large mockup of the carpet square (131k), as well as some action photos:

Theoretically the conference people are going to take up the carpet and send it back to us on a truck once the PDC is over. I'm looking forward to seeing it in person!

Sell your stock in News Gothic

The New York Times is standardizing on Cheltenham:

Starting today, the front page and main news sections of The New York Times are receiving a gentle typographical face-lift.

In place of a miscellany of headline typefaces that have accumulated in its columns over the last century, the newspaper is settling on a single family, Cheltenham, in roman and italic versions and various light and bold weights. A narrow variation will be used for The Times's signature one-column headline, which often appears at the top right of Page A1 on the main article of the day.

[ . . . ]

Before today's change, at least six headline typefaces commonly appeared on the front page. That kind of variety was customary for newspapers in the early 20th century, possibly because metal type was too costly and scarce for printers to stock full ranges of size within a family.

You can read the full story here (free registration blah blah blah). Don't miss the helpful graphic demonstrating the changes.

One typeface I'm going to particularly miss is Bookman Antique, which I enjoyed as a headline face on their inside pages:

(graphic shamelessly stolen from their article)

I understand and believe in the value of standardizing on one type family, but I always felt that Bookman Antique gave a nice, sturdy, trustworthy feel to those multi-column heads that Cheltenham doesn't convey.

great logo

Is this not one of the cutest logos you've ever seen?

It's for a company that sells organic foods, which makes the ladybugs particularly appropriate.

Designers with a sense of humor

Rebecca sent me to Chopping Block's Oranges site.

The song is great.

An ending and a beginning

Yesterday was the end of an era in the design world at MS: Our beloved design training manager, Scott Berkun, gave his last talk. He'll be cleaning out his office over the next week, but he's effectively gone.

There had been other talented people in that job, but none of them took to it with the same energy that Scott did. He was always excited about the possibilities of getting Design deeper into the product cycle, and he helped us get excited about it when we felt like we were faced with impossible odds. (And the odds ain't good, folks -- 50,000+ employees worldwide... roughly 200 designers... and you wonder why some of our products don't make any damn sense to human beings)

As a last lecture, it was a doozy. It had a classic Berkun title: "Ten Golden Rules: Thoughts on Software from Voltaire, Willy Wonka, and Microsoft Postmortem archives." It was also clearly a "whaddya gonna do, fire me?" talk: He had a beer, he said f*ck a couple of times, he sang a song about the product cycle, and he (jokingly) yelled at the training coordinator who asked him to repeat questions for the tape.

If you're involved in UI design at all, I'd check out the essays on his site, and bookmark it. He's going to do as little as possible for a couple of months, and then is planning to do a bunch of writing. I'm looking forward to it.

How design works

In this great interview with Tim Brown, CEO of IDEO, he talks about the process of design, and explains it much more succinctly than I ever have been able to. This one is going on my wall at work.

"The naïve view of designing is that it’s purely an additive process, about adding more and more and more. Actually, design is a funnel-shaped thing. It becomes an editing process: What is appropriate? What can be stripped away? So design is a holistic way of thinking. It's about being able to create the whole of something, and in such a way that somebody who’s using that product, whether for the first time or the tenth time, understands it can interact with it as seamlessly as possible."

Design for extreme situations

Here's a nice example of useful, usable information design, found in the trunk of our rental car in Detroit (a Ford, naturally -- do they rent Japanese cars in Detroit?)

It is, of course, made of glow in the dark material, so that when you need it, you can see it. Unless you're blindfolded. But does that really happen outside of the movies?

Behind the Typeface

OK, this is probably one of those things that's not only only funny to designers, but it won't make any damn sense to anyone else.

Nevertheless, there's a wonderful combination of type history and goofyness here: Behind the Typeface: Cooper Black

Low end media rulez!

There's a great article today over at Jakob Nielsen's site: Low-End Media for User Empowerment.

The summary is that on web sites, low-end media (text and static images) is, on the whole, much more usable and useful than high-end media (swoopy animated graphics, complex Flash navigation models, video, 3-d models).

Nielsen lists some technical impediments to high-end media, like download time, searchability, and accessibility, but rightly points out that those limitations will pass with time.

The greater problem, as Nielsen points out, is the issue of user control. As soon as something starts spinning, or animating, or making noise unasked, the user has lost control. As soon as content is streamed to the user, instead of allowing them to read or skim at their own pace, the user has lost control. The more a site tends towards fancy swoopy interface models and away from "click this clearly labeled link to see that relevant information", the less sure the user is of what they're going to see, and the less control they have over what they see.

And the dai-ichiban number one main thing the user wants, as I've said elsewhere, is to maintain control over their experience.

There are, of course, exceptions to the rule. Showing video of the Berlin Wall falling, or Rodney King getting beaten, is much more affective than still photos could ever be. But in cases where video would be useful, showing a still or two from the video as a preview gives control back to the user: "This is what you're going to see. Will it be worth it to you?"

It's constantly amusing to me that so many of the bad decisions people make about what to do with their sites could be solved asking a question I learned in the first term of my design education: "Is the design weaker without it? If not, then take it away." Is someone's experience really enhanced by that spinning golden arrow? If not, a simpler solution will probably be better.

More customer service critique

On the pretty good side: I called Canon customer support today to inquire about my Powershot G2, which we all know I shipped to them last week for some repairs. My expectation, from my original conversation with their phone support people, was that they'd send me an e-mail saying that it was going to cost me a ton of money to have it repaired, and did I want to go ahead with it? But I didn't hear anything from them. Silencio.

When I called them today I talked to a very chatty support guy (apparently he's learning programming, but types really badly, so it's slow going) who told me that they had already done the repair, under warranty, and it should be shipped out in the next couple of days.

Now, while I was pleased as punch that they decided it was a warranty repair -- I've owned the thing since last January, which by my math makes it over a year -- they could have improved my experience by dropping me a quick e-mail letting me know what the status was, instead of forcing me to call them to find out that they're doing right by me.

It wouldn't be that hard, either. Presumably they have some kind of tracking system in place. It wouldn't be that hard to add a field for "owner's e-mail" and set up the system so status changes trigger e-mails to me, like
"We've recieved your camera"
"We've classified this as a warranty repair"
"Repairs are done and the camera should be shipped in a couple of days"
"The camera has been shipped, here's the tracking number"

Heck, Moveable Type alerts me any time someone comments on this site, or tracks back to it -- and it's shareware.

The trick here is to make the customer feel as though they're in charge of the experience, or at least that they feel like they understand what's happening, instead of being in the dark. It's a small thing, but it has a huge effect on the customer experience. And on the ROI side, I have no idea how much my fifteen minute conversation with their support guy cost them, but it has to be dramatically more expensive than sending me a few automated e-mails triggered by their existing tracking system.

All of this of course is infinitely better than my current experience dealing with Tiffen, the manufacturers of my Domke camera bag.

The summary of the experience goes like this:

• I sent them my bag on December 23.


• I got busy and didn't realize time had passed.


• I called them at 2:00 on March 17 (I take good notes) to inquire as to the status of my bag. The support guy found my record in their computer, even though their new computer system "might not have it yet". It was after "Rochester" had closed for the day, so he printed out my record and said he'd contact me the next day.


• I called them again today, at 11:00 so I knew "rochester" would still be open. After he pulled up my record, whoops, his computer froze and he couldn't do anything. He took my phone number and promised to follow up with me today. Obviously they're closed for the day now.

The upshot of this is that now I'm going to have to call them back, and that I'm not convinced they'll have a straight answer for me. They're great bags... but now I'm not sure I'd ever buy another one. A lifetime guarantee is only worth as much as it's easy to get that guarantee fulfilled.

And clearly I'm irked enough that I'm telling you, gentle reader, my experience, so you have this knowledge to help make future decisions about what camera bag you might want to buy.

A designer's eye

I picked up a brochure on the bus the other day called "keeping our transit system safe". Here's one of the headings from the brochure:

I was showing this brochure to Miz Becky that evening and laughing at how sloppy the designer and printer had been. "What do you mean?" she asked.

I pointed out the tell-tale hairline of white on the left edge of the letters:

I explained that that meant that the designer didn't "trap" the type, or make sure it overlapped the yellow background slightly. Worse, the printer didn't know to do it either.

The problem is that if you depend on precise registration, when the plate inevitably shifts or stretches a little bit on the press, it will leave these little hairlines of white -- or sometimes larger white gaps that are visible to a normal human being.

That exhange got me thinking about how important it is for a designer to have a good eye and attention to detail.

A story on the subject of having a good eye: Several years ago I was flipping through a design magazine -- I think it was Publish or Print -- and I opened to a two page spread. On the left side there were two tiny dots. "That's odd," I thought, "I wonder why one of them is larger than the other."

Then I looked at the other page of the spread, where I saw this headline: "Introducing the monitor for those of you who can tell that the dot on the right is .000765 inches larger than the dot on the left."

"Oh crap," I thought. "I'm a target market."

I've since discussed this event with other designers, and the consensus is that this isn't that odd of an experience for people in our weird profession. Part of the reason we do this job is because we can't help but see that the fourth "e" over is kerned badly, and we can't stand it. We're compelled to fix it.

A related part of being a good designer is attention to detail. Again, an illustrative story: About 9 years ago I needed to hire a contractor to do some art production work for me. I gave both people the agency sent over the same test: I asked them to reproduce a complex piece of art for me from a printed example -- something I had done a couple of weeks earlier because an electronic version of this piece of art no longer existed.

I also made sure to give them far too little time to complete the task. Not because I'm evil -- I am -- but because I wasn't really interested in seeing them complete the task. I wanted to see how they approached the task.

One of the people had put many more boxes down on the page, but they were all "roughed in": different sizes and in approximate locations.

The other person had only put down a few boxes, but they had a careful system of grid lines laid down, all measured off of the graphic (I had made sure to leave a pica ruler casually lying around on the other desk in the office).

The second person had clearly thought through the graphic, figured out what was needed to reproduce it accurately, measured it out, and was proceeding in a careful and precise manner. They got the job.

It's that obsessive attitude towards visual issues that I think is one of the defining characteristics of a good designer. In many ways, we don't "do design", we are designers. We can't help it.

Update on UPS

OK, after much discussion in the comments on Going through the motions doesn't equal good customer experience, I wrote to the UPS customer support line asking if the e-mails were answered by a human or a computer. Their response:

Dear David Adam Edelstein:

Thank you for your response. All inquiries submitted to customer.service@ups.com are answered by a Human Being.

We hope this information proves useful. If we may be of future assistance, please feel free to contact us.

Thank you for using UPS Internet Services.

So there you go.

Going through the motions doesn't equal good customer experience

Yesterday I stopped off at my local UPS Store to send my digital camera off to Canon factory service. It's the first time I've interacted with them since they switched from being Mailboxes Etc., and as a hardened customer experience junkie I was curious to see how their planned "streamlining" had gone.

I walked in with my package and headed for the counter. The pleasant looking young woman behind the counter chirped something unintelligible at me.

"Excuse me?"

"U P S Label?" she chirped again, enunciating for me.

"Uh, no, it's my label."

"OK let's just go over to the terminals here an we'll just click guest and then you fill all this out and click done and a label will print and then you can come back to the counter ok?" She leads me to the computers that I mistakenly strolled past. I fill out what's actually not a bad form: my address, e-mail, and phone, their address and phone, insurance, etc. I click print.

The smallest label I have ever seen slides out of the (I now know) grossly oversized label printer next to the terminal. It has a bit of information and a bar code on it. Surely this can't be the label?

I take it back to the counter. Chirpy has been replaced with a guy about my age. While he takes my package the final step of the way (including finding out how fast I want it to get there, getting my money, and printing out a MUCH larger label for the package -- I miss what happens to the little one) he fills me in on UPS' plans for throwing out the whole brown/blue/red label thing, updating their branding, etc. "They're keeping 'Brown', though," he says. "I wish they'd change... I'd like a brighter color."

Once I had been trained, it all seemed to work well. I left figuring that was the end of it.

That evening, while I was cleaning out the day's crop of spam, I nearly deleted an e-mail with the return address of iShip_Services@iship.com. I didn't recognize the address, so I was about to delete it when the cat made an odd noise. As I looked away and looked back, I noticed the subject line: "Shipping label generated for package to Canon Factory Service."

"Huh," I thought, "I sent something to Canon today."

The e-mail was, of course, from UPS; but the only place in the e-mail that acronym showed up was at the bottom in the middle of a long package tracking URL.

I of course can't resist a "contact us" link and send them a message suggesting that they should brand their e-mail as UPS so I can tell that it's them. The next day, I get this e-mail in response:

Dear David Adam Edelstein:

Thank you for your inquiry. We apologize for any inconvenience caused in this matter. Please provide the package tracking number so we may investigate this matter.

Thank you for using UPS Internet Services.

I write back, include the tracking number, and explain that the point wasn't the package, the point was that the original e-mail didn't clearly refer to UPS.

The same person replied:

Dear David Adam Edelstein:

Thank you for your reply. Our system indicates this package was last scanned transit on March 25, 2003, and is scheduled for delivery by the end of the business day March 28, 2003.

Ship Notification is an information-based service only supported in UPS OnLine® Solutions shipping systems. This convenient, timesaving option allows the shipper to automatically transmit an e-mail message or fax to their customer or a third party informing them of details associated with a shipment.

Our records indicate the shipper for this package is THE UPS STORE. If you have further questions or need more details, you may contact the shipper of record, THE UPS STORE, at 206-254- 0346.

Thank you for using UPS Internet Services.

So what's my point here? Simply this: UPS has all the right pieces in place for great customer experience. They have a customer-controlled entry system, to minimize transcription errors. They have convenient locations. They send an e-mail when the package is shipped and provide easy tracking. And they have e-mail support that replies promptly to customer questions.

The problem is, there's no coordination or care behind any of this. Better signage in the store would have told me I needed to start at the computers. Better UI on their web site, where I originally confirmed the store location, would have told me it was possible for me to generate a shipping label at home, and even have it picked up. A better branded e-mail would have lowered the risk of me deleting the tracking information. And better trained e-mail support would have actually READ the content of my e-mail, instead of just looking for a missing tracking number.

The net result of all of these things is that, while I'll probably use UPS again because their location is convenient, I'm not terribly attached to them. If FedEx or DHL demonstrated they had even marginally better service, I'd switch with no qualms.

Teaching web design

Last term, I taught a class I hadn't tried before, called "putting your portfolio on the web". The idea was to teach just enough design for students to come up with a great site design, and just enough Adobe GoLive that they could build the site.

I was, frankly, terrified at the prospect. Neither of those subjects are particularly easy to cover by themselves. Cramming both of them into ten weeks was, I felt, going to be risky.

I made a few mistakes over the course of the term. On the first night, I gave out a handout explaining the structure of a web page, and talked people through it, but I didn't really explain it in terms that made any sense. I also probably relied too much on the book teaching them what they needed to know for the first few weeks. What I'll probably do next term is start on the first night with a more coherent lecture on how web pages are put together, and demonstrate that in GoLive.

I should probably also have spent more time over the course of the term demonstrating what they were going to do in homework that night, and making sure the next week that people understood what was going on. It was clear by week 10 that asking "did anyone have trouble with the homework" didn't really get people to discuss the troubles they were having. To some extent they're all adults, and if they're having trouble they should talk to me; on the other hand, it's my classroom, and if they're not learning, then it's at least partially my fault.

I also did some good things. My best idea was to have them bring in a portfolio web site every week for us to critique. This worked better than I had hoped. The process of looking at other people's sites, really looking at them, and tearing apart the graphic design, navigation, flow, usability, etc., taught them about design in a deeper and more holistic way than I possibly could have otherwise in 10 weeks.

Another good idea was for me to think of the class as though I was taking them through the same process I would take them through if they had hired me to design a site for them. We looked at their work and talked about their audience, the feel of the work, and how to show that message most clearly. I explained about gathering materials for inspiration and had them bring in some materials. That proved to be a great exercise -- it helped to show whether they were on track in their understanding of the work and the direction they wanted to take it, or if they had to go back and think harder about what their work was saying. By guiding them through the process of design discovery, especially in a group critique setting, they were able to see things in their work that they wouldn't have on their own.

Was it all a success? I think so. One comment from a student during the last class was that "all of these sites seem really personal, like they really fit the work." Another student, who came into class with a clear idea of what she wanted, and just wanted to learn how to build the site, left the class with an entirely different design, and was delighted.

Not all of the sites are up yet in their final form, but here's three out of five final sites. I'm delighted with them, and happy to see that the combination of great students and some good ideas on my part worked so well:

www.andreajwalker.com
www.nicholasduers.com
www.noellephoto.com

Now, on to next term and not repeating my mistakes!

A trifecta of design books

This might be of general interest to people thinking about design issues.

My friend Robert sent me an e-mail this afternoon:

> If you had to recommend a single book on good web design,
> what would it be?

My answer:

Hmmmmmmmmmmmmmmmmmmmm

I'm not sure I *can* recommend a single book.

I'm also not sure I can recommend a book specifically on web design. Most seem to focus on L33T HTML skillz, or the latest bad flash ideas, etc.

I guess I'd treat these three books as three volumes of my perfect book:

Designing visual interfaces is the best book I know about interaction design. If you were only going to read one, I'd read this one. It's focused on GUI design, but all of the concepts are applicable to designing a web page.

Design Basics is a great intro to graphic design issues, whether you're designing a poster or a web page.

MTIV is a great book on the process of designing an interactive site, finding inspiration, interviewing the client, etc. And it's funny.

The single best thing I can recommend, besides reading these books, is that you spend a lot of time looking at other similar web sites, and think critically about them. What works? Why? What looks cool, but doesn't work well?

I should add that the last thing -- critiqueing existing designs -- is the single best tool I know of for improving one's design skills.

At least 90% of the work of big-d Design is in thinking about the design problem. Coming up with visuals that fit with the solution you've come up with is the last 10% -- last meaning, it's the last thing you do.

Haven't figured out how the content is going to be structured, or even what content is going to be there? Don't know how much space you have to work with? Don't know who the audience is? The design will never support the content unless those questions are answered. Design and content will go their separate ways, only vaguely related.

You should know that already!

If you're reading this (and it's not clear why you are) you probably know that my day job is at a Large Software Company Near Seattle. While you might expect that this would make me some kind of apologist for our allegedly monopolistic business practices, I'm not. I don't know whether we've competed unfairly, or whether that warrants the proposed penalties (and, frankly, I don't care). Nor do I know whether those practices are worth all the hate we seem to be on the receiving end of, these days (I doubt it).

What I do know, however, is that there are plenty of things we do badly that are worth getting on our case about, and those things seem to be ignored in the storm and fury of the legal proceedings. Most of what we do is what Alan Cooper calls "dancing bearware" -- the impressive thing is not that the bear dances well, but that it dances at all.

What's annoying me this morning, and what set off this particular rant, is the password entry in Windows 2000. If you mistype your password, as I did when logging on this morning, you get the following message:

The password is incorrect. Please retype your password. Letters in passwords must be typed using the correct case. Make sure that Caps Lock is not accidentally on.

Now, this is certainly good advice. My keyboard is designed so that it's easy for me to hit the caps lock key instead of shift, and there's no feedback other than a little LED approximately 2 feet from where my focus is (which causes me all kinds of grief when I'm using Photoshop, but that's another rant). My current password is case sensitive, and sometimes I can't log in because caps lock is on.

Let's think this through further, though. Why is the caps lock a problem? 1) Because I don't notice when I turn it on. 2) Because it means I type the wrong password into the computer.

Hey, the computer knows what I typed... and the computer knows caps lock is on. Why doesn't the freakin' computer tell me caps lock is on, instead of making me figure it out for myself? It's the computer! It knows already!

Now, this is a tiny point, to be sure. The difference between "Caps lock is on, maybe you should turn it off" and "Make sure that Caps Lock is not accidentally on" is marginal. But it is indicative of larger problems in how we build software.

Ever tried to troubleshoot printer problems on a Windows system? Plug and play works great when everything works, but when it doesn't... whoa nelly. I've seen people reduced to tears by the process. Most of the time, the trouble has to do with settings on the computer that don't match features on the printer, like how much RAM the printer has -- if the computer assumes the printer has more than it does, the printer can puke if it's sent a file that's too large (I used to work in a print shop where we had an old TI laser printer; it needed to be rebooted every afternoon because of a memory leak, which we'd notice when documents started coming out missing the lower case "i". Everywhere in the document.)

I should never, ever have to tell the computer what the printer has. I should never, ever have to run back and forth between the printer and my computer to make sure that all the settings match. The computer knows what settings it needs to know about. The printer knows what those settings are. The computer should handle all of this.