A picture from Fifth Annual Austin Workshop on Test Automation (AWTA).
(top row ) Drew Pritsker, Ken Sommerville, Cem Kaner, David Vydra, Danny Faught, Barton Layne, Ross Collard, Carl Erickson, Bill Nasuti, Yuesong Wang, Carl Nagle, Pete TerMatt, (second row) Michael Kelly, Bret Pettichord, Rob Sartin, Todd Bradfute, Pat McGee, Bob D'Antoni, Jamie Mitchell, (third row) James Bach, ( not pictured ) Charles Rankin, Chris Morris.
“A team with a star player is a good team, but a team without one is a great team.” - Author unknown.
The above quote from Author leads us to discussion on great teams and its characteristics. The article stems from experience gained while working for different teams, observation of team members behavior under time pressure coupled with complex nature of project.This holds good for Software Testing team which finds prominence place in a project activities and requires right mix of people for performing these activities.
Why does some software testing team fails and others succeed? Is there any solution for this problem. The answer is “Yes”/”No” – depends on how the team member aligns himself towards common goal of the team not at the cost of suppressing his team members interest but working together with common understanding of problem at hand. The success also depends on leadership attributes possessed by Test Leads –“Captain of ship”.
The objective of this article is to help software test engineers or any person who believes in team work,to understand characteristics of high performance team and how to cultivate them in their own teams.
Success of team in long run doesn’t depend on individual who is considered “STAR” but does depends on all who form clusters of stars that makes great team.
Characteristics of Great Software Testing Team
Initial stage - Ask yourself following question:
Does your new team member knows the reason he has been selected for the team?
New members of the team are often puzzled about their presence in team.Although you may argue that he/she need not know purpose and just work on task assigned to him/her.This is assumption made by many higher management people.By clearing defining the roles and responsibilities helps individuals to understand the project in bigger context.That includes the relevance of his/her job,skills of individuals that could be contributed towards the projects,share common team goal which was defined earlier.This does bring great commitment towards the work and hence contributes towards its quality.
Ownership:
When project complexity increases in terms of tasks and team size, it would not be possible to keep track of individuals tasks by single leader.Hence the solution to this would be assigning Ownership to individuals. However this virtual leadership often act as a impediment rather than solution if not considered appropriately. Mere appointment of individual as Owner without considering a serious thought of whether he/she could manage their team would not bring desired result.
Individuals acting as owners should have mindset which matches leaders mindset and the pride on their part to act as future leaders. These are people who could make difference by carrying along with them their team members and the same people by showing Indifferent attitudes towards their team will disintegrate the team. The task of owners is not merely restricted to assigning task to team members but to understand task at hand, situation at much broader perspective and bringing common level of understanding among their team members. Support their team member at the time of difficulty of handling task,word of encouragement,correcting their mistakes by not acting as lead but as a peer,acting up on ideas or taking advice for appropriate situation from experienced members would certainly benefit towards shared goal. Collaboration and a solid sense of interdependency in a team will defuse blaming behavior and stimulate opportunities for learning and improvement.
Knowledge of seasoned players in the team
The term-seasoned players indicates the person who has spent considerable amount of time in same project or similar kind of work. They are resources who have vast knowledge about project. By channeling their knowledge in proper way,the entire team could be benefited.These individual should show an act of diligence towards others work rather than arrogance.It is commonly said “Past success breeds arrogance”. They are higher performers who’s absence could be felt in a team but it should be not sole criteria as there are equal chance for others who has similar caliber to act at this position.
Motivation – Key Factor
Motivation is not all about giving speech when members of team are assembled but rather every effort should be made to tailor these speech to address each individual. This means each of team member has unique qualities and unique working style. This task is rather complex than said for Test Lead since it will bring effort on leaders part to sense the team member’s feeling not only to task assigned to members but also on project as whole. Positive attitude of lead will energies team – This is quoted from experience working for one of great test team.If the leader complains about long working hours or insisting the team members to work at schedule which is impossible to meet, your team will reflect your attitude. He/She is true leader who inspite of unreasonable schedule instills the confidence among team members to believe in their abilities and at the same time working at the background on his part to justify his team members effort working on unreasonable schedule but bring an extension to these schedule to make his team members job simple.
Recognition
Everyone likes to be recognized for his/her work.When an individual is awarded for his/her work,the responsibility of team lead should bring reason for individual recognition in front of others. The team lead decision for these kind of task should be impartial.This does bring great respect for the awarded individual by members in the team. They would be acting on similar grounds and ultimately team benefits from their collective response. Very often that members working for virtual leader often are not recognized since due to zero visibility to the leader of team. It is virtual leader who has to bring on table the accomplishment,contribution done by team member towards their task.This indicates that virtual leader is future leader who does take care of members of his team and well received by members of his team to whom they always wanted to be associated in future.
One-One basis Meeting
It is often seen that roles and responsibilities for the members are defined and assessment is done at the end of project.Agreed that it is formal process.But informal talk like One – One basis adds to this formal process as well. These informal meeting should address issues at present whom members wont feel like conveying during group meeting, future opportunities for members, identifying future leaders/owners of the team and equally acting on issues at hand after feedback from team members.Timely and appropriately delivered feedback can make the difference between a team that hides mistakes and a team that sees mistakes as opportunities. The responsibility for poor performance is usually a function of the team structure rather than individual incompetence; yet, it is individuals who are sent to training programs for fixing. If team members feel like they are pitted against one another to compete for rewards and recognition, they will withhold information that might be useful to the greater team. When a team has problems, the effective team leader will focus on the team’s structure before focusing on individuals.
“Don’t tell people how to do things, tell them what to do and let them surprise you with their results.” - George Patton
Conclusion
There are plenty of things to be considered while building successful team.The key words – Unity,Trust,Respect for others opinion and acting without fear are ingredients for great test team,in general for any successful team. After reading this article look at your team and question yourself “Are you working in great test team” or “ Will you make every effort to build great test team”.Then don’t wait,try next second to build “Great Software Testing Team”.
“Coming together is a beginning, Keeping together is progress, Working together is success”. - Henry Ford
Today I found one useful site on Connection strings. This is one of main thing we always do or search for help before starting the core work in coding...:-)
I am a Tester/End User, I'll expect what I want/my requirement not what the product/software is giving to me. In my usage I don't bother about "software limitations" or "Not an issue" kind answers...:-)
Today I tried the below scenario and I felt like It should not work like this. I am not sure how many will agree with me.
Like a “black box” flight recorder for your software. BMC’s patented Black Box software is analogous to a black box flight recorder on an aircraft. Deployed on servers (BEA WebLogic, IBM WebSphere, JBoss, Apache Tomcat, Microsoft Windows and .NET) and the Windows-based clients that access them, BMC Black Boxes capture application execution at multiple, synchronized levels, based on a dynamic, user-defined recording profile.
Eliminate the need to recreate problems and end user environments Just as you wouldn’t re-crash a plane to determine the root cause of a problem during a flight, with BMC Black Box, IT teams don’t have to recreate end user environments or attempt to reproduce a reported error or defect in order to fix it, which provides a significant efficiency gain. It also eliminates the wasteful ping-pong communication that often occurs among end users, support staff, and developers or between developers and testers when trying to gather information about the defect or reported issue.
In effect, Black Box logs provide a “digital documentary” of application execution that provides all of the information needed for diagnosing the root cause of any unexpected application behavior. A Black Box log can be stored in a centralized repository or saved as a highly-compressed file that can be e-mailed or attached to defect tracking, systems management, or service desk for seamless integration with problem resolution processes in phase in the application lifecycle.
Gather the information you need without compromising performance or privacy BMC Black Box does not require any source code instrumentation or modification of executables, and can be controlled and analyzed while applications are running in production. You can fine-tune Black Box recording profiles (RPRs) on-the-fly during run time, in order to ensure the right balance between the depth of information captured and overhead. For added privacy and security in regulated environments, BMC Black Box recording profiles can be encrypted and set to eliminate sensitive data before being sent for analysis.
Flexible deployment options for proactive monitoring or reactive capture of problem information Black Boxes can be run continuously or just when problems occur and can be embedded into software that is delivered to end-users. There are four common deployment schemes:
Distributed: In a distributed mode, Black Boxes run on clients and/or servers connected to one or more BMC Application Problem Resolution repositories. The distributed mode is typically used to monitor and capture problems in a LAN/WAN environment, either in the test lab or production environment.
Standalone: When servers or clients are not accessible via a network (e.g., in a remote, isolated environment or in highly secured environments), BMC Black Box can be used in a standalone mode. In this mode, it runs as an independent application and is operated by an end-user. When application recording is completed, the log is saved as a file and can be sent by e-mail or FTP for root cause analysis.
On-demand, Web-based: BMC offers a customizable Web portal module that automates the process of installing Black Boxes and recording a problem scenario at remote end-user sites. When using the portal, the end user visits a Web page that automatically installs, configures, and activates the BMC Black Box seamlessly.
Embedded: Organizations that wish to automate support for their applications can use a COM version of the BMC Black Box. This object exposes all necessary APIs to set a recording profile, record problems, and save logs. This option can provide ISVs with a significant competitive differentiator for their support organizations and allow in-house application developers to deliver applications with enhanced supportability.
What Information can the BMC Black Box Capture?
Microsoft Windows/.NET
Java EE
User experience
Client side: Video-like capture of the end-user’s screen, keystrokes, and mouse clicks. Server side: User sessions, ASP.NET Web pages and services accessed by each user.
BMC Application Problem Resolution can synchronize the Windows-based client user experience view with the server side application execution details.
System configuration
CPU type, display resolution, installed RAM, Windows version, Swap file size, .NET Framework version, connected drives, system environment variables, regional settings, services, drivers, and more.
OS type and version, JDK type and version, Application Server type and version, server resources (JDBC, Threads, Memory), server parameters (class path, JVM), and more.
Application configuration
All system resources touched by the recorded application, including files, registry keys, modules and COM/COM+ objects, running processes, and .NET configuration. When present, IIS and IE configuration are also captured.
Configuration of application components, deployment structure and times, JNDI info, URL mappings, implementation structure down to class and method level detail.
System performance
Values of selected performance counters, such as memory usage, CPU utilization, GC, connection pools, and ASP error rate are recorded over time.
Alerts of application and application server problems, metrics of application server and application components performance. Application server and host, virtual and physical memory, JDBC, JTA, JMS, thread pools, execute queues, EJB, servlets and JSP.
Application/Component/Transaction performance
Execution duration, Web page hits and errors (ASP and ASP.NET), COM/COM+ objects, database calls (direct and stored procedures), Web services calls, function calls, and more.
Execution duration, Web page hits and errors, database calls (direct and stored procedures), Web services calls, and more. Transaction performance data, enabling you to quickly detect performance degradation, isolate slow-performing tiers, and drill-down to analyze individual instances of problematic transactions for root cause determination.
System-level execution
All application calls made during recording, including database calls, Web services calls, IIS operations, .NET Remoting calls, .NET garbage collection, network events, component invocations, Active Directory calls, file accesses, registry accesses, process start and exit, and more.
All application calls made during recording, including database calls, Web services calls, network events, component invocations, and more.
Code-level execution
.NET Tracing calls; full trace of Visual Basic and C++ function calls, arguments, return values, variables, exceptions, stack, and more.
Complete code execution including: method calls, object state at each method call, arguments, return values, and stack. It also captures exceptions, which may indicate a code problem. BMC Black Box does not require any change in the application and runs against optimized production applications.
Internal logging events
When applications already have some type of internal logging (either by using the .NET Trace object or by simply writing to a text file), the Black Box can record logging events and embed them into the log.
The Black Box can trace calls to logging frameworks and other logging calls and make the data available as part of the ongoing recording (embedded in the log).
Other custom events
Black Box can attach other files (e.g., configuration files) to its log, to include any data point needed for root cause determination. Users can also add custom events to the log through a documented API.
The Black Box can trigger custom actions that can perform arbitrary functions (e.g.. send e-mail with file attachments). An API is available for users to implement their own actions.
Component execution
BMC Application Problem Resolution can synchronize the Windows-based client view with the server side component execution details.
Any Java class (e.g., POJO, custom exceptions, custom MVC frameworks), Servlets, JSPs, EJBs, JDBC, JNDI, JMS, JTA, JCA and networking. BMC Application Problem Resolution presents component invocations on a timeline as well as performance data for each invocation.
Testersdesk.com - Useful Site for Testing Engineers. Just Login to site, Select your Test requirement, enter the input it will give/generate the data which you required
It has some nice features/functionalities all are self explanatory with nice help.
I still remember the days where we need exact 1000 Chars of Text, 2000 Chars of Text, files required with different sizes like exact 10KB, 1MB and 10MB to validate the some Text Fields in my applications. I used to the take help of Winword application to enter the text and count manually and some times using Word count option. Using this tool it is very simple..enter the length and type of data..it will generate the data for u.
Please find few screen shots from the site. Just login to site. I am sure you will also like this site.
Today I read a post by Mr. Mohan Bharadwaj on Testing Challenges. I really liked the post.
Every now and then I hear people saying that we don’t have enough time for testing or our estimates have gone wrong due to some resource issues, however we can resolve these things by doing risk analysis, we need to identify the areas where testing should be focused.
Since it’s rarely possible to test every possible aspect of an application, every possible combination of events, every dependency, or everything that could go wrong, risk analysis is appropriate to most software development projects.
This requires judgment skills, common sense, and experience. (If warranted, formal methods are also available.) Considerations can include:
- Which functionality is most important to the project’s intended purpose? - Which functionality is most visible to the user? - Which functionality has the largest safety impact? - Which functionality has the largest financial impact on users? - Which aspects of the application are most important to the customer? - Which aspects of the application can be tested early in the development cycle? - Which parts of the code are most complex, and thus most subject to errors? - Which parts of the application were developed in rush or panic mode? - Which aspects of similar/related previous projects caused problems? - Which aspects of similar/related previous projects had large maintenance expenses? - Which parts of the requirements and design are unclear or poorly thought out? - What do the developers think are the highest-risk aspects of the application? - What kinds of problems would cause the worst publicity? - What kinds of problems would cause the most customer service complaints? - What kinds of tests could easily cover multiple functionalities? - Which tests will have the best high-risk-coverage to time-required ratio?
Not enough time for thorough testing?
This is a cry frequently heard as deadlines approach. You can hear most of the managers screaming about this issue…. However there could be a number of answers:
1. The testers were not able to complete testing due to a new release being loaded. 2. The bug was not in an earlier release (reload that earlier release and see). 3. The bug could not be tested for earlier because some part of the release did not work and inhibited the test’s ability to “see” the bug. 4. The bug was in some part of the system not originally planned for the release for which a test has only just been written. 5. The bug was found while running some other test. 6. The bug was in a part of a system which was not the focus of testing. 7. The bug would have been found eventually, but the tester hadn’t run the test (which would have found it) yet. 8. And yes, maybe if we’d been more thorough we’d have found that bug earlier.
It’s always good to keep in place a corrective action in place so that the impact of the issue can be minimized and the stake holders and the client/s do not lose faith on you and your team.
How many of us are not asking the team members to stretch or extend to complete the Task?
I am sure everyone will ask..(:-)
As you all know there is no tool or approach available to give 100% correct estimation for Testing Effort. In My knowledge (my opinion) this will depend on Situation, Project, Resources, Technology. If I am having everything (clear specifications, good resources and required technology) in my hand i will choose WBS for Effort Estimation. If I don't have clear requirements I will say 40% of development Team. I feel I should have thorough knowledge on all approaches and should able to choose the right approach based on the situation.
There are some good links available in net on Software Testing Effort Estimation. Please verify the below links.
Test Effort Estimation Using Use Case Points by Suresh Nageswaran - CTS (Click Here)
Test Effort Estimation: Learning from an experience Article by Deepak Goel (Click Here)
Test Effort Estimation by Murali Chemuturi (Click Here)
This is my favourite subject, I am looking for more stuff on this subject..Will post soon...
Dani has generously donated his book to the QTP community, it is presented under the Creative Commons Deed, which loosly means you can use it freely as long as you provide proper credit to the author. Please remember that in this case, AdvancedQTP was just a publishing platform - the book was wholly written by Dani, and he should get full credit for his tremendous work.
