Burns gives a straightforward and easy to understand exposition of how to use Selenium version 2 as a browser testing framework. What may be most impressive to some readers is early in the book, where there is a list of which browsers are able to run it. Chrome, Internet Explorer, Firefox and Opera. The only major omission is Safari. Sorry, all you Apple users. But maybe Selenium 3 will include it.
At the most basic level, you can quickly learn how to install and run it so that it can then capture your manual actions at one of those browsers, starting at some base web page. This starts with the simplest case, where you are going to that page and then going to other pages by clicking a link in the current page. But Selenium goes much further in capability. You can go to a given text element that is present, and then tell Selenium that this text should be present. Ditto for insisting that a clickable item be present. This way, Selenium can test in future versions of the page for the presence of items, rather than being limited to testing a user's actions.
Another common scenario is where multiple browser windows are present. Also testable.
More interesting to programmers is how the book goes beyond the previous. WebDriver could be the key distinguishing feature of Selenium 2. It controls the browser from outside, whereas the earlier abilities were run inside the browser. The advantage of WebDriver is that it bypasses the locks placed on javascript code that runs in a browser. Those were necessary to prevent malicious javascript code being run from a webpage visited by the browser. But in the context of automated testing, you often are running javascript testing code that you wrote yourself. So to get around the inherent and deliberate browser limitation, it was necessary to ab initio from outside. The second half of the book takes you into programming. The writing of javascript that runs inside WebDriver to access the browser.
