About Arborosa

Arborosa is a software tester, walking fanatic, father of two and husband living in Utrecht in the Netherlands. This blog is intended to display my thoughts and opinions on software testing, books, blogs, experiences and anything else that I find interesting enough to write and publish about.

Test Types – N, O, P

Test Type
A particular type of testing that has an approach, goal and/or use of oracle(s) that provides information that is typical to that test type.

This is the sixth post in a (sub) series on Test Types. Please add any additions or remarks in the comment section.

Negative testing
Negative testing ensures that your application can gracefully handle invalid input or unexpected user behavior. It is the process of validating the application against invalid data.
Although most testers will agree on the importance of negative testing it sometimes is difficult to put the necessary time into it. Especially when many developer or business stakeholder argue against too many “hypothetical” tests in favor of confirmation tests

Neighbour testing
Testing the connectivity and data exchange between the application and its direct ‘neighbours’.
A colleague of mine used this type of testing in a Chain Test Plan that he wrote. I found it a useful expression to describe limiting the (initial) scope of a chain test to its immediate surroundings.

Network conditions simulation testing
Testing traffic, exchange of data and behaviour of systems and interfaces while simulating different network conditions.
Think of introducing latency, interruptions, data package loss etc.

Non-functional testing
Testing how a software application or system operates.
A very short description that covers a world of possibilities. While behaviour is the subject of many requirements documents, user stories and the like, non-functionals are much less intensively described. Often they are only identified by a number of constraints and boundaries describing what the system should be able to handle or what is specifically not allowed to do. Don’t be fooled by their absence. Non-functionals have the possibility of influencing the behaviour of applications so that it looks as if the behaviour is incorrect or correct while it is not. Moreover non-functionals can have a major impact on user experience. There are many quality attributes and testing types connected to non-functional testing that can be used for investigation.

Operational testing
Operational testing is testing focussed on usage of the application or system in its intended environment, its intended usage and by its intended users.
Environment here extends beyond the hardware, tooling or user interface and such. It includes applicable standards, procedures, processes and culture. Similarly for users this not only relates to consumers or end users but also to operations, customer service, engineers, etc. And usage extends beyond day-to-day use but also relates to maintenance, backup and restore, etc.

Packaged application testing
Testing application packages or application suites on the inter functionality of the separate parts in their interaction and their behaviour as a whole.
There are a  number of suppliers offering package application testing services for products like SAP and Oracle. Although I see the benefit of testing process flow and behaviour throughout such a group of applications that more or less work together I am not quite sure how this differs from any other testing.

Penetration testing
Investigating a system and its environment on the possibilities of gaining unwanted access to a environment or system or the possibility to retrieve not be disclosed information.
Penetration testing together with performance testing (see next item) is in my opinion one of the specialist fields in software testing. While every testing should have some basic understanding and skill in this area it involves both continuous renewed skill, knowledge and practice to excel in penetration testing. 

Performance testing
Determining how a system performs in terms of responsiveness and stability under various work loads.
While the focus is often particularly on high loads or peak loads special attention should also be given to low loads, intermittent loads, loads addressing only specific parts of the system. Also load should point towards both load given to the system as load produced by the system.
There are several sub-areas that fall under performance testing that each have a specific area of performance that they address: Load testing, Stress testing, Soak testing, Configuration testing, Spike testing, Isolation testing, etc..

Test Types – J, K, L, M

Test Type
A particular type of testing that has an approach, goal and/or use of oracle(s) that provides information that is typical to that test type.

This is the fifth post in a (sub) series on Test Types. Please add any additions or remarks in the comment section.

Keyword-driven testing

Keyword-driven testing is a methodology used for both manual and automated testing but is best known in combination with automated testing. In this method of testing the documentation and design of testing is separated from the execution. Keywords or action words are defined for each of the actions to be executed. These words are then used in a pre-defined format (often a table) to enable a combination of actions to be executed (and possibly evaluated) by either a person or a previously set up test automation framework.

Load testing

Load testing is the process of putting demand on a software system or computing device and measuring its response.
Although closely related to stress testing load testing has the aim to measure behaviour for different loads and not necessarily for peak loads.

Localization testing

Localization is the process of adjusting (internationalized) software for a specific region or language by adding locale-specific components and translating text.
Localization testing then is performing checks and validation that the translation and locale-specific components are functionally correct and understandable.

Manual scripted testing

Manual scripted testing is the process of manually executing previously designed test scripts in search for software behaviour that does not match the behaviour as described in the script.
On paper this type of software testing is still relatively popular as it is based on the early descriptions of structured testing and as such appeals to non-testers. A fair number of testing practitioners however judge executing test scripts as inefficient, ineffective and boring.

Manual support testing

Oddly enough there are two definitions in circulation for this.
“Testing technique that involves testing of all the functions performed by the people while preparing the data and using these data for automated system.
This one I believe should be discarded as being part of test automation preparation. Although I have to admit that at the start of test automation this activity is often under estimated. The second one I think is much more useful.
Testing manual support systems involves all the functions performed by people in preparing data for, and using data from, automated applications. The objectives of testing the manual support systems are to verify that the manual support procedures are documented and complete, support responsibility has been assigned, determine that the manual support people are adequately trained and to determine that the manual support and the automated segment are properly interfaced.

Memory profiling

Memory profiling is the process of investigating and analyzing a program’s behavior to determine how to optimize the program’s memory usage.

Migration testing

Migration testing is the activity of testing data, functionality or behaviour of software after migration to a new platform, database or program for intended and unintended differences.

Model based testing

Model based testing is the automatic generation of software test procedures, using models of system requirements and behavior. Once created these test procedures can be run repeatedly.
Model based testing has been around for several years now but has never really made a big impact on testing as such. The reasons for this I believe are that the “automatic generation” and repeatability of the tests require a considerable technical- and time investment and are not nearly as easily achieved as portrait by model based testing tool vendors. Next to this the success and quality of the tests is highly dependent on the ability to formulate the software behaviour into a model and on the quality of the then created model itself. In practice model based testing will find bugs but not as many as often promised and certainly not all.

Mutation testing

Mutation testing (or Mutation analysis or Program mutation) is used to design new software tests and evaluate the quality of existing software tests. Mutation testing involves modifying a program in small ways. Each mutated version is called a mutant and tests detect and reject mutants by causing the behavior of the original version to differ from the mutant. This is called killing the mutant. Test suites are measured by the percentage of mutants that they kill. New tests can be designed to kill additional mutants. Mutants are based on well-defined mutation operators that either mimic typical programming errors (such as using the wrong operator or variable name) or force the creation of valuable tests (such as dividing each expression by zero). The purpose is to help the tester develop effective tests or locate weaknesses in the test data used for the program or in sections of the code that are seldom or never accessed during execution. (Wikipedia)

Test Types – G, H, I

Test Type
A particular type of testing that has an approach, goal and/or use of oracle(s) that provides information that is typical to that test type.

This is the fourth post in a (sub) series on Test Types. Please add any additions or remarks in the comment section.

Glass box testing
Testing or test design using knowledge of the details of the internals of the program (code and data) (BBST definition)
Too many this might sound familiar to White box testing. I think however that the perspective is quite different. Here the perspective is from a tester rather than a developer.

Gorilla testing
Gorilla testing is testing on particular module, component or functionality heavily and with large variety.

Grey box testing
Grey box testing is testing with, limited, knowledge and access to the code and internal structure and details of the software.
It is my opinion that most testers, safe pure programmers and pure black box tester, are doing grey box testing. As such they stand somewhere on a scale between the two extremes. Their place is sometimes determined by (enforced) choices of role or scope but more often by their programming and design knowledge and capabilities.

GUI testing
Graphical User Interface testing is testing the application’s user interface to detect if the functionality of the interface itself and the functionality that is directly influenced or dependent by the user interface functions correctly.
With the current growing attention to automated testing, and especially to testing on API level, the GUI is quick to be overlooked in its importance. It remains however know more than ever the first point of contact of any user with the system.

Incremental integration testing
Incremental integration testing is an approach in which you first test each module of component individually and then add them one by one together and test the integration. You can do this top down, bottom up or functionally incremental.

Installation testing
One side of installation testing is aimed at ensuring that all the necessary components are installed properly and are working as required once installed. The other side of installation testing focusses on what users need to do to install and set up new software successfully.

Integration testing
Integration testing is testing where, previously tested, individual modules and components are combined and tested as a group. It tests not only interactions between individual components but also between different sets of components and parts of the system within its direct environment. Integration testing focusses on different aspects such as functionality, performance, design and reliability.

Inter systems testing
Inter systems testing focusses at testing on interconnection and integration points of separate systems but working together.

Interface testing
Interface Testing is performed to evaluate whether systems or components pass data and control correctly to one another. It is to verify if all the interactions between these modules are working properly and errors are handled properly.

Internationalization testing
Internationalization is designing software systems in such a way that they can be adapted to different languages and regions without engineering changes, loss of functionality, loss of data or integrity issues. Internationalization testing is aimed at uncovering these potential problems.

Test Types – D, E, F

Test Type
A particular type of testing that has an approach, goal and/or use of oracle(s) that provides information that is typical to that test type.

This is the third post in a (sub) series on Test Types. This post covers test types beginning with D, E and F. Please add any additions or remarks in the comment section.

Data integrity testing
Data integrity testing focusses on the verification and validation that the data within an application and its databases remains accurate, consistent and retained during its lifecycle while it is processed (CRUD), retrieved and stored.
I fear this is an area of testing that is often overlooked. While it gets some attention when functionality is tested initially, the attention on the behavior of data drops over time.

Dependency testing
Examines an application’s requirements for pre-existing software, initial states and configuration in order to maintain proper functionality.

Destructive testing
Test to determine the softwares or its individual components or features point of failure when put under stress.
This seems very similar to Load Testing but I like the emphasis on individual stress points.

Development testing
It is an existing term with its own Wikipedia page but it doesn’t bring anything useful to software testing as such.

Documentation testing
Testing of documented information, definitions, requirements, procedures, results, logging, test cases, etc.

Dynamic testing
Testing the dynamic behavior of the software.
Almost all testing falls under this definition. So in practice a more specific identification of a test type should be chosen.

End-to-End testing
Testing the workflow of a single system or a chain of systems with regard to its inputs, outputs and processing  of these with regard to its availability, capacity, compatibility, connectivity,  continuity, interoperability, modularity, performance, reliability, robustness, scalability, security, supportability and traceability.
While in theory end-to-end testing seems simple. Enter some data and check that it is processed and handed over throughout all systems until the end of the chain. In practice end-to-end testing is very difficult. The long list of quality characteristics mentioned above serves as an indication of what could go wrong along the way.

Endurance testing
Endurance is testing the ability to handle continuous load under normal conditions and under difficult/unpleasant conditions over some longer period of duration/time.

Error handling testing
Use specific input or behavior to generate known, and possibly unknown, errors.
Documented error and of exception handling is a great source to use for test investigations. It shows often undocumented requirements and business logic. It also interesting to see if the exception and errors occur based upon the described situation.

Error guessing
Error guessing is based on the idea that experienced, intuitive or skillful tester are able to find bugs based on their abilities and that it can be used next the use of more formal techniques.
As such one could argue that it is not as much a test type as it is an approach to testing. 

Exploratory testing
Exploratory testing is a way of learning about and investigating a system through concurrent design, execution, evaluating, re-design and reporting of tests with the aim to find answers to currently known and currently as yet unknown question who’s answers enable individual stakeholders to take decisions about the system.
Exploratory testing is an inherently structural approach to testing that seeks to go beyond the obvious requirements and uses heuristics and oracles to determine test coverage areas and test ideas and to determine the (relative) value of test results. Exploratory testing is often executed on the basis of Session Based Test Management using charter based and time limited sessions.
It is noteworthy that, in theory at least, all of the test types mentioned in this series could be part of exploratory testing if deemed appropriate to use. 

Failover testing
Failover testing investigates the systems ability to successfully failover, recover or re-allocate resources from hardware, software or network malfunction such that no data is lost, data integrity is intact and no ongoing transactions fail.

Fault injection testing
Fault injection testing is a method in which hardware faults, compile time faults or runtime faults are ‘injected’ into the system to validate its robustness.

Functional testing
Functional testing is testing aimed to verify that the system functions according to its requirements.
There are many definitions of functional testing and the one above seems to capture most. Interestingly some definitions hint that testing should also aim at covering boundaries and failure paths even if not specifically mentioned in the requirements. Other mention design specifications, or written specifications. For me functional testing initially is conform the published requirements and than to investigate in which way this conformity could be broken. 

Fuzz testing
Fuzz testing or fuzzing is a software testing technique, often automated or semi-automated, that involves providing invalid, unexpected, or random data to the inputs of a computer program. The program is then monitored for exceptions such as crashes, or failing built-in code assertions or for finding potential memory leaks.
Yes this is more or less the Wikipedia definition. 

Test Types – B, C

Test Type
A particular type of testing that has an approach, goal and/or use of oracle(s) that provides information that is typical to that test type.

This is the second post in a (sub) series on Test Types. This post cover test types beginning with B and C.

Backward compatibility testing
This testing can be done on several levels. On platform level it means to investigate if an Application/Product developed using a previous version of a platform should still work in a newer version of a platform.
On document level it means to investigate if a document created with a previous version of the product should still works on the new version of the product.
On feature level it means to investigate if input, usage and result of a feature in the previous version compares to input, usage and result in the newer version.

Benchmark testing
Benchmark testing is the act of running (parts of the) software, in some circumstance, to assess its relative performance in comparison to an existing ‘benchmark’.

Beta testing
Beta testing testing is the follow up of alpha testing. During beta testing the software is released outside of the development team and offered to (limited) groups of people within the same organization and possibly a limited set of (potential) end-users. Beta testing offers the possibility for the software to engage in real world scenarios and receive early feedback with regard to bugs, use-ability, competiveness etc.

Big bang integration testing
In Big Bang integration testing all components or modules are integrated simultaneously, after which everything is tested as a whole.
I am not a big fan of this type of integration as it is very difficult to isolate causes of failures and bugs. Also test coverage is only high level. It interesting however to see that, while not on purpose, this type of testing is often implicitly done when continuous integration is implemented and the integration interval, and thus the test interval, is daily or parts of a day.

Black box testing
In science, computing, and engineering, a black box is a device, system or object which can be viewed in terms of its inputs and outputs (or transfer characteristics), without any knowledge of its internal workings. Its implementation is “opaque” (black). Hence black box testing can be describes as approaching testing without taking any knowledge of the internal workings of the software into account.
Personally I do believe that software testing has progressed beyond the notion of believing that one could do with mere black box testing. I do think that it can be valuable approach alongside grey box- or white box testing. Although in practice I never seem to make that distinction anymore when I define a test approach.

Bottom up integration testing
Bottom up integration testing is an approach to integrated testing where the lowest level components are tested first, then used to facilitate the testing of higher level components. The process is repeated until the component at the top of the hierarchy is tested.

Change related testing
Execution of tests that are related to the code and/or functional changes.
Some view this as being regression testing. While this is probably part of change related testing it potentially ignores the tests that are targeted on confirming and validating the changes as such.

Compatibility testing
Compatibility testing is testing executed to determine whether or not the software is compatible to run on:

    • Different types of hardware
    • Different browser
    • Different devices
    • Different networks
    • Different versions
    • Different operating systems
    • Different configurations
    • Etc.

It is basically the testing of the application or the product built with its (intended) environments.

Competitive analysis testing
Testing targeted on how sales or customer experience are impacted:

  • If a feature or functionality is not available
  • If a feature or functionality is not working
  • If a feature or functionality is not available but is available at a competitor
  • If a feature or functionality is not working but is working at a competitor
  • If a feature or functionality is available and working but a competitor has a different version (cooler / less cool; cheaper / more expensive; faster / slower; etc.)

While this may seem more marketing than testing (context driven) testers are often at the front runners in identifying these differences.

Compliance testing
Testing focussed to establish if the software is compliant to known criteria such as:

  • Open standards (to ensure inter operability with other software or interfaces using the same open standard)
  • Internal processes and standards
  • International standards (e.g. IEEE, ISO)
  • Regulatory standards

Component testing
Testing of separate software components without integrating them to other components.
A component can be any portion of the system under test (e.g. module, program, class, method, etc.). It is recommendable to select components of a similar abstraction level.

Concurrent testing
This is two very different meanings:

  • Testing throughout an iteration, sprint or development cycle concurrent with all other development activities. Whereas concurrent means both simultaneously, cooperative and collaborative.
  • Testing activity that determines the stability of a system or application under test during normal activity.

Conformance (or conformity) testing
I see three meanings of conformance testing:

  • Testing meant to provide the users of conforming products some assurance or confidence that the product behaves as expected, performs functions in a known manner, or has an interface or format that is known.
  • Testing to determine whether a product or system or just a medium complies with the requirements of a specification, contract or regulation.
  • A synonym of compliance testing

Conversion testing
Testing of programs or procedures used to convert data from existing systems for use in replacement systems.

Test Types – A

This sixth post in the series of software testing overviews introduces the first of some 80+ different test types. That number itself is completely arbitrary. While searching for and investigating testing definitions I found over a hundred definitions of test types and I have chosen to leave out a number of ‘test types’. My choice to do so is based on the interpretation that some test types rather described a test level or a test technique and I could not see how to make a useful test type out of them.

So what then do I call a test type?

To me a test type is a particular subject of testing that has an approach, goal and/or use of oracle(s) that provides information that is typical to that test type.

While going through my overview you might find that some of the test types I mention do not entirely fit the narrow description of a test type as provided above. The reason that they are mentioned in spite of this is that I felt that they were so often mentioned as a test type that they should have a mention in this post just for that reason.

This post differs somewhat from the earlier posts as the definitions used are often rewritten by me to form a single aggregate definiton as many different ones for the same term exist. Where useful I have added comments as additional information. Also since a post with over 80 descriptions would be too long I have split up the overview into alphabetical sections. To begin with the letter A.

Just as a reminder these are not necessarily my definitions but a collection of definitions I encountered. Finally there are no sources or attributions for the individual test types as this would make this a totally different exercise.

A/B testing

A/B testing originates from marketing research used to investigate the more efficient of two possible solutions by presenting them to two different groups and measuring the ‘profits’. In software testing it is mostly used to compare two versions of a program or website, often of which only one contains changes on a single or a few controllable criteria.

Acceptance testing
During acceptance testing requirements, variables, parts of a program or specific behaviour of a program is compared against measurable aspects of predetermined acceptance criteria. This requires at least four things.
First identification of the requirements, variables and parts or behaviour (coverage). Second expressing these in measurable aspects.
Third the aspects need to represent the defining elements of the acceptance criteria. Finally the acceptance criteria themselves should represent the needs and wants of the stakeholder. The goal of this interpretation of acceptance testing is to provide stakeholders the possibility to accept the software. And provide the possibility of sign-of.

A more pragmatic way to look at acceptance testing is to allow the stakeholders, often end users, to evaluate the software and see if it meets there expectations and (operational) needs. In practice this is often done by proxy by stakeholder representatives. Sometimes the testers are selected as the representatives. I do not think that is a good idea as testers then step outside of there boundary of information provider, are often commited to the solution and their role in creating it and more essentially testers are not the end users. 

Active testing
Testing the program by triggering actions and events in the program and studying the results. To be honest I do not consider this a test type as in my opinion this describes nearly all types of testing.

Ad-hoc testing

Ad-hoc testing is software testing performed without explicit prior planning or documentation on direction of the test, on how to test or on which oracles to use.
Some definitions see this as informal, unstructured and not reproducible. Informality and being unstructured (if seen as unprepared) is certainly true as this would be the point of doing it ad-hoc. The not reproducible part depends on whether you care to record the test progress and test results. Something that is in my opinion is not inherently attached to doing something ad-hoc but highly advisable. Why elso would you be testing if you are able to tell the testing story.

Age testing
It is a testing technique that evaluates a system’s ability to perform in the future. As the system gets older, how significantly the performance might drop is what is being measured in Age Testing.
To be honest I found only one reference to this test type but I find the idea interesting. 

Agile testing
Agile testing is mentioned often as a test type or test approach but I have added no definition or description here. In my opinion agile testing is not a software test type. Agile testing rather is a particular context in which testing is performed that may have its particular challenges on test execution, on how tests are approached and on choices of test tooling but not a specific test type. 

Alpha testing
Alpha testing is an in-house (full) integration test of the near complete product that is executed by others than the development team but still is executed in a development environment. Alpha testing simulates the products intended use and helps catch design flaws and operational bugs.
One could argue that this is more of a test level than a test type. I care to view it as test type because it is more about the type of use and its potential to discover new information than that it is part of the software development itself. I specifically disagree with the idea that this is an intermediary step towards, or is part of, handing over software to a Test/QA group as some definitions propose. In my opinion testing is integrated right from the start of development up until it stops because the product ends its life-cycle or due to some other stopping heuristic. 

API testing
API testing involves testing individual or combined inputs, outputs and business rules of the API under investigation.
Essentially an API is a device-independent, or component-independent access provider that receives, interprets/transforms and sends messages so that different parts of a computer or programs can use each other’s operations and/or information. Testing an API is similar to testing in general albeit that an API has a smaller scope has, or should have, specific contracts and definitions that describe the API’s specific variables, value ranges and (business) rules. Testing an API should however not be limited to the API alone. Sources, destinations (end-points), web services (e.g. REST, SOAP), message types (e.g. JSON, XML), message formats (e.g. SWIFT, FIX, EDI, CSV), transport- (e.g. HTTP(S), JMS, MQ)  and communication protocols (e.g. TCP/IP, SMTP, MQTT, TIBCO Rendezvous) all influence the overall possibilities and functionality of the API in relation to the system(s) that use(s) the API. Typically API testing is semi- or fully automated and requires sufficient tool, message type, and transport- and communication protocol knowledge to be executed well.

An alternative meaning of API testing is testing by using an API. In this case the API is a means to an end in gaining access to the subject under test and feeding it with data or instructions and gathering responses.

Regression Testing

As a follow up in the testing definition series it was my intention to continue with covering Test Types. Initial investigation showed what I had already feared. Such a post would become a Herculean task and probable my longest post ever. So I will continue with that particular endeavor sometime later tackling it one step at a time. This post for starters covers one of the most common but also one of the most peculiar types of testing

“Regression Testing”

Regression Testing is so common as a testing type that the majority of books about software testing, and agile for that matter, that I know, mention regression testing. Almost as common however is that most of them either or both do not tell what regression testing is or do not tell how one should actually go about and do regression testing. To be fair an exception to the latter is that quite a few, particularly the ones with an agile demeanor, tell that regression testing is done by having automated tests but that is hardly anymore informative is it.

Before I go further into regression testing as being peculiar first inline with the previous posts a list of regression testing definitions:

  • Checking that what has been corrected still works. (Bertrand Meyer; Seven Principles of Software Testing 2008)
  • Regression testing involves reuse of the same tests, so you can retest (with these) after change. (Cem Kaner, James Bach, Bret Pettichord; Lessons learned in Software Testing 2002)
  • Regression testing is done to make sure that a fix does what it’s supposed to do (Cem Kaner, Jack Falk, Hung Quoc Nguyen; Testing Computer Software 2006)
  • Regression testing is the probably selective retesting of an application or system that has been modified to insure that no previously working components, functions, or features fail as a result of the repairs. (John E. Bentley; Software Testing Fundamentals Concepts, Roles, and Terminology 2005)
  • Retesting to detect faults introduced by modification (ISO/IEC/IEEE 24765:2010)
  • Saving test cases and running them again after changes to other components of the program (Glenford J. Myers; The art of software testing 2nd Edition 2004)
  • Selective retesting of a system or component to verify that modifications have not caused unintended effects and that the system or component still complies with its specified requirements (ISO/IEC/IEEE 24765:2010)
  • Testing following modifications to a test item or to its operational environment, to identify whether regression failures occur (ISO/IEC/IEEE 29119-1:2013)
  • Testing if what was tested before still works (Egbert Bouman; SmarTEST 2008)
  • Testing of a previously tested program following modification to ensure that defects have not been introduced or uncovered in unchanged areas of the software, as a result of the changes made. It is performed when the software or its environment is changed. (Standard glossary of terms used in Software Testing Version 2.2, 2012)
  • Testing required to determine that a change to a system component has not adversely affected functionality, reliability or performance and has not introduced additional defects (ISO/IEC 90003:2014)
  • Tests to make sure that the change didn’t disturb anything else. Test the overall integrity of the program. (Cem Kaner, Jack Falk, Hung Quoc Nguyen; Testing Computer Software 2006)

Looking at the above definitions the general idea about regression testing seems to be:

“To ensure that except for the parts of the areas* that were intentionally changed no other parts of these areas or other areas of the software are impacted by those changes and that these still function and behave as before”.
(*Area is used here as a general expression for function, feature, component, or any other dimensional divisions of the subject under test that is used)

The peculiar thing now is that however useful and logical such a definition is it only provides the intention of this type, or should I say activity, of testing. Regression testing could still encompass any other testing type in practice.

To know what to do you first need to establish which areas are knowingly affected by the changes and then which areas have the most likelihood of being unknowingly affected by the change. Next to that there probably are areas in your software where you do not want to take the risk of them being affected by the changes. In his presentation at EuroSTAR in 2005 Peter Zimmerer addresses the consequences of this in his test design poster by pointing out that the wider you throw out your net for regression effects the larger the effort will be:

  • Parts which have been changed – 1
  • Parts which are influenced by the change – 2
  • Risky, high priority, critical parts – 3
  • Parts which are often used – 4
  • All – 5

Once you have identified the areas you want to regression test you still need to figure out how to test those areas for the potential impact of the change. The general idea to solve this, in theory at least, seems to be to rerun previous tests that cover these areas. As this might mean running numerous tests for lengthy periods of time many books and articles propose to run automated tests. This will however only work if there are automated tests to use for testing these areas to begin with. And even if there are you still need to evaluate the results of any failed test and there is no clear indication of how long that may take.

How do you know that these existing tests do test for the impact of the change? After all they were not designed to do so. For all you know they might or might not fail due to changes to the area that is tested by them. Either result could therefore be right or wrong in light of the changes. The test itself could be influenced by an impact of the change on the test (positive or negative) that was not considered or identified yet.

All in all regression testing is easily considered to be necessary, not so easy to determine, difficult to evaluate on success and considerably more work then many people think. Even so next to writing new tests it probably is the best to solution to check if changes bring about unwanted functionality or behavior in your software. My suggestion to you is to at least change the test data so that these existing tests have a better chance of finding new bugs.