A side benefit for installing IE 7 on your machine, when you are testing or analyzing your web application with eValid, is that eValid would be using the newest IE 7 libraries. You get the benefit of all of the security and reliability enhancements that Microsoft made in IE 7 compared with the latest version of IE 6.

• Recommended screen layouts for optimal eValid test recording development.
• How to test file upload and download operations.
• Discussion of screen access issues.
• How to collect information about page response times.
• Suggested PASS/FAIL criteria for site analysis runs.
• How eValid scripts are expanded at playback time.
• Use of the SyncOnURL command for AJAX applications.
• The operation of the eValid Adaptive Playback feature with InputValue commands.
• How to test PDF objects.
• Recommendations on validating image data.

In the Engineering Psychology course, under Prof. Thatcher's direction, one of the student assignments is to evaluate a website. Students first write a detailed website usability proposal and then they conduct the website usability evaluation according to their proposal, based in part on use of the eValid engine. eValid is pleased to be able to provide eValid licenses for use in the analytic part of the student projects.

Previous versions of eValid used a 3.0 sec timed hover to decide when to re-navigate the display to describe the properties of the pointed-to element. With the new facility the navigation done in real time without waiting.

Here is the revised PageMap GUI Documentation that tells all about the changes.

Mainly developed under the auspices of the Office of the Deputy Under Secretary of Defense for Acquisition and Technology, in cooperation with NDIA's Systems Engineering Division, this event a number of topic areas specifically oriented to software, software engineering, and the Internet.

From time to time we apply a similar methodology to determine the "best online experience" of public websites here in the USA. The definition we use for this is a measure of the Ease of Navigation (based in part on after page download speed and broken/unavailable page counts) and Quality of Maintenance (based in part on on relative age of pages in the website).

The latest survey we have done is of the following 10 Top E-Tail sites: Circuit City, Eddie Bauer, JC Penny, L.L. Bean, Lowes, Niketown, Payless, Staples, Walgreens, and Walmart. These sites all involve E-commerce, have significant complexity and size, and are well-known brands.

You can review the Top 10 E-Tailers Best Online Experience Summary Report and access the eValid scans (the raw data) for yourself. Here's the scoresheet:

Please let us know if you have any questions about this survey. If you would like to schedule a detailed competitive analysis between your own website and those of your competitors. We have a special Comparative Analysis Request form for you to use.

• The Third International Workshop on Quality Assurance and Testing of Web-Based Applications, Chicago. Illinois, USA, 17-21 September 2006. Following two prior workshops of the same name in 2004 (Hong Kong) and 2005 (Edinburgh), this rendition will also focus on the special concerns that are raised because of the "web" basis of the application.

• The 6th International Workshop on Automated Verification of Critical Systems, Nancy, France, EU, 18-19 September 2006. While not specifically focused on web applications, the pervasiveness of the web clearly is a factor in this workshop, which focuses on current methods (model checking, probabilistic systems analysis).

• Workshop on Constraints in Software Testing, Verification, and Analysis, Nantes, France, EU, 25-29 September 2006. The constraint-based approach is still new to the area of program analysis, but the results when applied to topics like test case generation and heuristics suggest that some significant new results will be possible.

The key idea in the eValid solution was to construct a set of test scripts performing typical actions, run them on 24 x 7 on a 10/hour basis, and wait to generate enough instances of the "intermittent" failures were documented in playback event logs to permit changes to be made to the server software.

After several weeks of regular monitoring, detailed analysis of the captured failures led to repairs of each problem, and the observed failure rate dropped significantly. See the Intermittent Server Failure Monitoring description for full details.

Ajax lets Web developers create "rich content, thin client-side" applications with somewhat less effort than with alternatives like Flash or Java Applets. As such, Ajax fits in with methodologies like Extreme Programming (XP) and allied types of rapid prototyping. The resulting product can have superior look and feel and contain powerful functionality as well.

Google comes up the clear winner in several different application categories, including calendars, Email, information managers, and spreadsheets. But Google is not alone in exploiting the benefits of AJAX, and the article points out a number of non-Google applications that are winners as well.

• The Excluded URLs Report, so you can see the effect of the Blocked URLs list during a scan.
• The OOPs Page Report, so you can identify pages that a server reports to the user as being unavailable in a "OOPs Page Message", but does NOT deliver with a 400+ error code.
• The Page Links Report, which brings out a complete list of all of the links found in the scanned site in a table that can be sorted up/down by parent or child URL. Note: this same information is already used in the dynamic displays in the eValid 3D-SiteMap, but the linear form has been requested by users for some time.

The Apdex Index is based on whole-page loading times, and on a pre-defined threshold time value T secs. The recommended T value is 4.0 secs. If pages generally load faster than the threshold, a user is presumed "satisfied" with the performance. If pages load slower than that, but faster than four times T (16 seconds, for example) a user is viewed as "tolerating" that performance. Anything slower (greater than 16 seconds, for example) and the user is termed "frustrated".

The Apdex Index formula is as follows:

      ( #satisfied + ( #tolerating / 2) ) / #total

After applying this formula to a website (a set of pages) or to a particular page (downloading it 100's of times) Apdex's comparative scoring rules are as follows:

Score		Interpretation
0.94 Up		Excellent
0.85-0.94		Good
0.70-0.85		Fair
0.50-0.70		Poor
Below 0.50		Unacceptable

Getting the values from an eValid site analysis run is a simple matter, as shown in this step by step Apdex Analysis Setup instruction page.

As an auxiliary application beyond measuring general website speed, the Apdex folks also point out the Apdex Indes is a fairly good way to rank order Business IT Alignment when different parts of a website are aiming at difference performance levels.

But it looks like this idea's back, as witness these two events:

• The Second Sorkshop on Mutation Analysis (Mutation 2006), schedule for Raleigh, North Carolina in November 2006.

• The First International Workshop on Random Testing (RT/06) Portland, Maine, July 2006, part of the ISSTA 2006 Conference.

It used to be said that mutation testing was never going to succeed because the machine requirements were too great, but that doesn't seem to be an issue any longer. It seems an old idea (a good idea) has been resurrected!

The papers will appear eventually in the International Journal of Software Engineering and Knowledge Engineering. They are looking in particular for general methods for testing, but also for papers that deal with testability, reliabilty, and performance.

For information you can contact the Guest Editors: Alexander Keller, IBM Watson Research Center alex@us.ibm.com and Jean=Philippe Martin-Flatin, Ecole Centrale Paris, France jp.martin-flatin@ieee.org

• The International Workshop on Software Verification and Validation is being held in Seattle, Washington (USA) on 21 August. This workshop, a part of the Federated Logic Conferences (FLoC), aims to focus attention on new and innovative methodologies and technologies for analyzing and verifying software.

• The workshop on Formal Apsects of Testing and Runtime Verification (FATES/RV 2006), also a part of the Federated Logic Conferences (FLoC), aims for the special problems of analysis of software at execution time.

• The International Workshop on Software Certification, held in conjunction with the FM06 conference at McMaster University, 26-27 August 2006. It aims to deal with reliability issues for a wide range of mission critical software applications.

His current research interests include software process modeling, software requirements engineering, software architectures, software metrics and cost models, software engineering environments, and knowledge-based software engineering. His contributions to the field include the Constructive Cost Model (COCOMO), the Spiral Model of the software process, the Theory W (win-win) approach to software management and requirements determination and two advanced software engineering environments: the TRW Software Productivity System and Quantum Leap Environment.

The Complete Interview between Barry Boehm and Michael Milutis, Executive Director of the IT Metrics and Productivity Institute, took place in April of 2006.

Two critical automatic synchronization commands are already build into eValid. They handle most cases when eValid does not synchronize automatically. The commands are the Validate and Synchronize on Screen Image command (SyncOnText) and Validate and Synchronize on Text String command (SyncOnScreenRectangle)

To dig deeper check out the Playback Synchronization page. The main point is that this support is built into eValid -- you don't have to write a program to get it.