I recently attended O’Reilly’s Open Source Conference in Portland and was very excited this year to see a tremendous focus on testing. There were sessions on unit testing and test-driven development; but, I was most excited to see people tackling one of the most complicated and least-advanced testing environments: web testing.

Testing web applications has been a bit of a black art. You’ve got a minimum of three languages to deal with (HTML, CSS, and Javascript), perhaps a fourth language for your tests, and a cornucopia of environments in which to test. Each web browser can be seen as a virtual machine for your application and, currently, there’s a bare minimum of four different virtual machines your application will have to run in (Internet Explorer 6, IE 7, Firefox 1.5, and Firefox 2.) That’s the fewest you can get away with, and you also have to worry about variations in browsers running on different operating systems. Yahoo!’s supported browser list has 22 separate browser/operating system combinations. It is no wonder to me why developers throw up their hands at browser compatibility issues.

Luckily, the team at the Open Source Applications Foundation has been hard at work. Their new testing platform, Windmill, is still rough around the edges but seems to be a great competitor to Selenium. Windmill’s focus is making it easier to debug tests and author them in any programming language, as well as fitting into continuous integration. It provides you with a nice test authoring environment and console with any of its supported browsers to help in the traditional test, fix, and refactor development cycle.

It is easy to forget that we are in the early days of web applications. The coming of Web 2.0 and the heavy use of Ajax is roughly analogous to the leap from text-based command-line applications to graphical applications on the desktop. I am very excited to see our testing tools begin to mature to meet the new challenges of this platform.

I finished eating a tasty morsel at Chipotle (three crispy tacos) and my buddy and I headed back to our cars. He was parked near me and had a little flyer on his windshield. I didn’t. The flyer — an advertisement from a local vendor for Sirius Satellite Radio — was only for Toyota Prius cars. Talk about targeting.

In advertising, there’s nothing worse than spending ad dollars on a group of people that simply don’t care about your product or service. Putting that flyer on my car (or every car in the parking garage), for example, would have been a complete waste of money.

The good news is that today, you can “toyota prius target” your customers online. Tools like AdWords allow you to advertise to those searching for answers (e.g., a Google query for “brooklyn dodgers baseball cap”). You can also focus those ad dollars on specific geographic regions, cities, or areas, otherwise know as geo-targeting. That means if you only do business and have customers in Fort Myers, FL, you can advertise solely within that city.

The story goes on too. You can drill down into analytics to see what paid visitors looked at after they came to your site. You can see if they took any specific actions, like downloading a brochure or filling out a contact form. And by the way, hopefully when potential customers visit your site, you have both kinds of SEO working for you.

I don’t think my friend actually acted on the flyer he received but he definitely looked at it. It is tough to get people to take action. But it is even harder to get someone to act on something that is totally irrelevant to them or worse, something they intensely dislike. Note to advertisers: I went to school in Boston but if you ever send me anything related to the Red Sox, I will boycott you forever.

January 9th must have been a scary day for the Mobile Web. Steve Jobs stood on a stage at MacWorld and told the entire mobile industry not only that it would be competing with its much-rumored iPhone, but that it finally lets users browse the “real web,” and not just those plain, boring “mobile” sites. Every company, consultant, and developer who had spent time or money working on a great mobile site to give their users a better mobile experience must have felt like they’d wasted their time. And here we are, just over a month after the coveted iPhone hit the hands of consumers, and they are all quickly coming to find their reliance on none other than the Mobile Web.

Just days after the launch, and even before the launch, scores of iPhone-centric sites were popping up, including the iPhone version of Digg and the slick suite of Leaflets from the guys at Blue Flavor. Now, a month later, there seems to be an iPhone site for almost everything, from reading various newspapers to tracking your fuel mileage.

Why would iPhone users, with a mobile web browser that can parse and cleanly display full-size web pages, gravitate toward these special sites, which often have a more limited feature set and a more ordinary appearance?

The answer is context. Content is still important, but context is king on a mobile device. If you have a device, like the iPhone, with a small screen, a limited ability to enter lots of text, and a slow or latent connection, the last thing you want is to have to zoom around different parts of a page, type in a load of text, or wait while twenty-odd connections finish loading your one page.

Instead, you typically want to complete a specific task, and don’t want extraneous “features” or “information” to get in the way of making it happen. Let’s pretend you live in the DC area, even if you don’t, and you want to take the Metro. On an iPhone, you have to wade past news headlines, advertisements, and links to information about the Board of Directors before finding the route planner. Then, you have to type in your origin and destination with the on-screen keyboard. Once you submit, assuming you typed correctly, you wade past that other info again to find the next train.

Compare this with their mobile site (which doesn’t work on the iPhone, since it’s old-school WAP, but should work on other mobile phones) or the iPhone-centric Meenster. Within three finger- or key-presses, and no input or scrolling, you can find the same information. These sites recognize the limitations of the mobile platform — even the iPhone — and provide a user experience that helps you do what you came to the site to do as easily as possible.

Rather than make the Mobile Web irrelevant, the iPhone has instead done just the opposite: mobile applications are more relevant than ever, and iPhone users are quickly choosing to use services that have chosen to offer sites that provide them with a better user experience.

When considering whether or not the investment in a mobile-centric site is worth it for your company or project, consider two quick questions:

• Would someone using a mobile device have a reason to need your service immediately?
• Does the information needed require a form to access, or is it found more than one click into the site?

If so, you probably have a significant audience that would be better served by a site tailored to mobile, and when you serve users’ needs quickly and effectively, they’re not only going to come back, but they’ll do your evangelizing for you.

The Mobile Web isn’t dead: it’s just getting started.