My app is a song database that musicians can use to assemble set lists; when finished, they can print them out and bring the hard copy on stage like musicians have always done. Some of the first feedback I heard was that the text or the buttons were too small; I didn’t get this because the app uses Apple’s standard interface elements, practically guaranteeing ease-of-use. Finally I realized that musicians were taking their iPads on stage with them and reading the set lists directly from the screen rather than using printouts. I had to add a “performance mode” with all the elements sized for viewing from a distance, and interacting with quickly between or during songs.

Since then I’ve added features that allow musicians to control their on-stage equipment from their iPads, and I’m about to add features to allow multiple band members’ iPads to interact with each other on stage. These and many other features have taken me by surprise because I never would have used the app in this way; I only envisioned it as an easy way to generate printouts. My users have taken it much further than I imagined, and I’m fortunate that they’ve waited for me to catch up so many times.

This fits perfectly with the Derek Sivers book I recently read, Anything You Want, in which he describes his customer service philosophy and quotes another author, Steve Blank: “No plan survives first contact with customers.” And it fits with another book I read this year, The Lean Startup, which advocates the idea of a “minimum viable product” to test your business ideas while saving your resources for growth after you know where it’s needed.

This has been an interesting lesson for me since I typically judge my success in building things by how closely I achieved my original goals. Instead, I’m going to start thinking about how quickly I can find out what the users of the product want, and how effectively I can adapt the product to serve them. I can’t wait to try my next project with this in mind!

I have to admit, Steve’s passing struck me in an unusual way. I never met him, he wasn’t part of my family, and yet I felt more personally connected to this loss than from any other celebrity or public figure. As someone mentioned in a tribute this morning, everyone in the technology world looked up to him, and for me that’s been the case for my entire professional life — since I first started creating multimedia in HyperCard in 1993 to yesterday, when I installed a pre-release copy of iOS 5 onto my iPhone for app testing. I’ve made my living from Apple’s products for almost 20 years.

I’ve told people that if the Mac platform went away, I’d change careers, and I’m serious. I can “do” technology, but the satisfaction of creating something well-crafted and elegant is what keeps me interested in my work. Other platforms are too focused on tech for tech’s sake. I suppose that’s Steve’s biggest contribution — melding art and technology, left-brained and right-brained — and my life is exponentially more interesting because of it.

To my delight, there have been only good things about reading a book like this electronically. Unlike a paper book, I don’t need to hold the pages open … so I have two hands free to rest my head, pet the cat or drink a cup of tea. For a book filled with nautical terminology, the built-in dictionary is great — I just touch a word to get its definition — and from there I can jump to its Wikipedia article for more info. And one of my favorite iPad apps, Google Earth, is a great complement — as the Pequod’s crew sails past landmarks like St. Helena and Cape Horn, I can “fly” there to learn more about them. The end result is a highly educational multimedia experience, using a 150-year-old text document as the source.

The screen text, brightness and contrast are just fine, and the backlight means that if I read in bed while my wife is sleeping, I don’t need an annoying clip light (none of which work very well, in my experience). An added bonus is that the iPad syncs wirelessly with my iPhone through my iTunes account. So if I want to read a few pages on the iPhone, I can open the same book and it will automatically jump to the page where I left off on the iPad.

The only thing holding me back from doing all my reading here is availability. When the iPod came out, I digitized all my music in a weekend and never looked back. It’s not so easy to scan in my favorite books and magazines. But one of my regular magazines is experimenting with electronic subscriptions. And as far as classic novels go, I’m thinking of trying Don Quixote next!

Eventually I stopped worrying about websites collecting data that isn’t personally identifiable. In other words, if one website has a list of every page a user from my city visited, and another website has a list of every page a user from my city visited, but neither website can correlate the lists and neither website knows that user is me, I’m okay with that. I know that the data can be used to improve the operation of the website, but can’t be used to target or report on me specifically.

Something has happened over the last few years, though, that changes the paradigm of online tracking, yet I haven’t heard anyone talking about it. It’s called Gmail — the best free, web-based email service available. Unfortunately, it’s made by Google — the best search engine available. And while you don’t have to be logged in to use Google, you do have to be logged in to use Gmail. And whenever you’re logged in, your data is personally identifiable by both services.

This isn’t a problem for me because I rent my own web server space with my own email accounts. But I do post content on YouTube, and after Google bought YouTube, they merged those login systems as well. Now if I log into YouTube to upload a video, and while I’m waiting for the upload I use Google to do some web searches, my searches are being tracked as part of my personal account. It’s the difference between reporting on how many books were checked out from my local library, and reporting on which books I checked out. And that’s where I draw the line.

Now Facebook is trying to get into the same game, by partnering with websites to offer a “log in with Facebook” feature, and I expect the trend to spread. The appeal is obvious: personally identifiable data is much more valuable for marketing purposes. But if you value your privacy, I’ll make these recommendations. Use these services only as long as you need to, and then log out. If I need to stay logged in for a while — waiting for that large YouTube video to upload — you can log in with an alternate browser (e.g., Firefox vs. Safari) while continuing to use your main browser anonymously. Finally, remember that if you log into one site, then use another site owned by or partnering with that site, they can correlate their data to create a more complete picture of you than you would probably want them to have.

I’ve worked with over a dozen designers and created over 100 websites, and I routinely encounter confusion about how this process works. The problem is that some designers and some clients become so focused on perfecting the conceptualizations — image files created in Photoshop or Illustrator — that they forget that those are not the ultimate product. Here are a few concepts to keep in mind if you’re working on the design phase of a web project:

1. Implementing a web design is not a process of “converting” the design files to a web format, it’s a process of building a website using the design files as a guide — just like a house must be built using blueprints as a guide.
2. The goal of any good developer is to match those depictions as closely as possible, but the more particular you are about the details, the longer that process will take. To take a simple example, if the design files show some leading between lines of text, most developers will see that and implement it. But if the developer set the leading to 10 pixels and you had your heart set on 11, the developer will have to spend time revising that.
3. It’s not important that the design files depict every page in exactly the way it will appear on the web. What’s important is that they depict every element that will appear somewhere in your site. For example, if you have a different submenu for each of five site sections, you don’t need to provide five different designs — just give the developer one design and he or she can extrapolate the other four from that. A less intuitive example is icons. If you have a half-dozen different icons that appear on a dozen different pages, it’s more convenient to give the developer one file containing only those icons, than a dozen files showing the complete pages in which they appear. Conversely, if the site contains an unusual page with unique elements, you must provide a file showing that page or at least those elements.
4. The individual elements are typically what the developer will spend time the most time implementing … combining them together to make each page is the easy part. So it’s helpful to think of your design as a collection of elements rather than a complete whole. How many elements are there — did you create four different styles of buttons, when two would suffice? Does every page use the same handful of text styles, or do you envision dozens of different text styles throughout the site? If you try making a list of every element you’ve designed and giving each element a name, you’ll have an idea of what the developer will face.
5. In most cases, the website you’re designing will grow and evolve … usually before the design phase is even complete. Therefore, it’s important to think of different ways in which the design elements can be reused. If you provide a box to present a couple sidebar blurbs, can that box expand to accommodate a longer blurb in the future? If the main menu contains five items now, does the design allow for adding two or three more? Will new text use a small number of styles, or will the person entering the text also have to perform a lot of formatting to keep the look consistent?

If you think of the design as one step in the middle of a long process, rather than the end result, you should be able to create a design that is both easier to implement and more flexible and valuable over time.