The answer lies in the use of the Structural/Algorithmic Testing testing approach. Instead of basing test on properties of the page, when a particular feature is likely to move around, the test is based on (pivots on) that feature. The result (as the table outlines) is a little bit more work in exchange for a much higher level of durability -- without any loss in test quality (no increase in false-negative results).

A simple technique that we recommend, a Manual Script Creation Process, takes the mystery out of converting a "from life" recording into a highly durable validation test. Once the key page elements are identified, you modify the script using the Structural Testing Commands.

P.S. If this structural testing approach still isn't strong enough for your needs, with some additional programming effort you can exploit the eValid Programmatic Interface [EPI] to create a C++ program that runs your test at any level of durability you want.

• What's The Impact Of This On The "Reality" Of The Playbacks? -- The cache is a unitary resource, and eValid instances must share it during multiple-BU playbacks, but, this turns out to be a conservative assumption.
• Making A Recording Versus Scripting A Test -- The eValid approach makes creating the script so easy, our recommended approach is to "record first, script later".
• Why is having a lot of BUs running such a big deal? -- For many modern web applications -- especially those which use AJAX methods -- having an actual Browser User do the test playback assures 100% realism. This means that your tests will be exactly what an end-user experiences.
• Structural Testing For Web Apps -- There are special eValid "structural testing" commands that allow a script to be completely independent of inconsequential page changes.
• A Sequence of HTTP References -- There is a special facility in eValid to generate a sequence of HTTP Get's that correspond to the actual sequence of page components that comprise a full page download.
• Microsoft SkyDrive Applications -- The Microsoft SkyDrive based applications -- common office utilities like a word processor or a spreadsheet -- actually test quite well with eValid, but they are among the more challenging applications to test.
• Link Checking -- There's a special eValid dynamic link check command that automatically visits every link on a page, even when the page is generated dynamically.
• Constant ID Tags -- When a web page has permanently-assigned ID tags eValid tests can automatically "pivot" on those internal identifiers, but even when the ID tags are dynamic it is still possible to obtain reliable tests.
• Running A Very Large Scan -- When the eValid site analysis "spider" is scanning a website you have to remember that to do its work properly eValid has to download every page. If the page it scans to is "bad" for some reason then that may stop the scan.

Please Contact Us for additional information about any of these topics.

• 1st International Workshop on Evolution Support for Model-Based Development & Testing 16 September 2010, Ilmenau University of Technology, Erfurt, Germany, EU.
• 7th International Workshop on Web Services and Formal Methods, 16-17 September, Stevens Institute of Technology, Hoboken, New Jersey, USA
• 25th IEEE/ACM International Conference on Automated Software Engineering (ASE 2010), 20-24 September 2010, Antwerp, Belgium, EU.
• 4th International Workshop on Testing, Analysis, and Verification of Web Software (TAV-WEB 2010), 21 September 2010, Antwerp, Belgium, EU.
• Model-Based Testing and Test Automation (MoTes2010), 27 September 2010 - 2 October 2010, Leipzig, Germany.
• Workshop on Trustworthy Self-Organizing Systems, TSOS 2010, 28 September 2010, Budapest, Hungary, EU.
• Testing: Academic and Industrial Conference -- Practice and Research Techniques (TAIC PART 2010), 3-5 September 2010, Windsor, UK.

These events all deserve consideration: please take a look!

• Watch some of our Annotated/Animated Movies. You'll see how eValid works from the GUI level, and see how eValid handles specific testing situations.
• Review the eValid Frequently Asked Questions (FAQs).
• If you run into a tricky problem testing a website, you might consider checking out the 40+ entries in our Worked Solutions Resource. You'll find a wide range of already-worked examples showing how eValid solves typical testing situations.
• For general eValid usage you may wish to check into the set of our "top 99" Power User Tips & Tricks. They cover everything from simple functional testing to detailed LoadTest experiment development.