Avoiding pitfalls while adopting new IP development technologies

Date:   Monday , September 03, 2007

Technology changes are more frequent now as compared to a decade back. With this development, it has become necessary for organizations to keep the TDNA (Tech DNA) quotient high in order to survive as well as to gain a competitive edge in areas that include defining solutions and operations management. There are many examples where enterprises, including government organizations, are strategically looking at identifying technology trends to gain optimization, reduce TCO (total cost of ownership), and prevent single vendor dominance. Some of the examples are as below:

* Some European government agencies have made Linux and Open Office as the standard core technology stack.
* A study presented in the recently concluded User Interface Experts Group annual seminar has shown that the web sites that implement RIA like GoogleMaps have a higher user base as compared to conservative web sites.
* One of the largest casinos in Las Vegas has switched to using RFID to prevent thefts.

The following experience of ours demonstrates the need for TDNA in operations management.

In one of the long running projects, a customer chose to change the platform from ASP 1.0 to ASP 2.0 with Rich Internet Architecture with AJAX as User Interface layer.

Our team was performing automated testing using an open source tool successfully for more than a year. While the new platform was still at drawing board stage, our testing team assembled and performed an impact analysis on the change. We found that the open source solution does not address AJAX testing portion.

Since a majority of the UI consists of AJAX components, the teamperformed a tool evaluation and found one of the latest versions of a commercial tool suitable for the requirement. The team went one step ahead and created a Proof of Concept to the customer with leading AJAX driven web sites as candidates. The team proved this in a demo to the customer.
The customer was happy with the results and the pro-activeness of the team to bring the limitations of the existing open source tool to their notice at an early stage of system lifecycle, and agreed to consider purchase of the new version of the test tool.

However, not all technologies identified can be implemented with success. One of the best practices to follow could be to pore through the industry analyst reports.

For example, the Gartner Hype curve given in Gartner Research summarizes the visibility and maturity of the presently hot technologies and forecasts the productivity they will have. The Hype curve is made up of the following phases -- technology trigger, peak of inflated expectations, trough of disillusionment, slope of enlightenment, and plateau of productivity.

Of the five phases listed above, it would be safe to go ahead with phase four and five, as technology gets stabilized at these phases.

The experience we gained while adopting an open source performance-testing tool for a long-term project with complex infrastructure set up presents a good example. At the initial stages the tool worked well in the user modeling as well as in the playbacks. However, when we went into run-time analysis the tool left us handicapped. Since it was an open source tool, dependency was on open source community to cater to the need, and it did not happen. We had to take recourse to a commercial tool to cater to the requirement as well as rework on user modeling activities and effort overruns.

To sum up, though multiple choices are available in identifying and leveraging technology one needs to be careful in choosing the right technology to avoid unexpected and undesirable end results.

He can be reached at kidambi.seshadri@marlabs.com