Prediction for the upcoming year 2023 on Software Automation & Quality
Autonomous Test Design
Traditional test automation still requires that the tests themselves are manually written. Model based approaches allow tests to be generated from the central model; Autonomous Test Design takes the next step and automatically generates these models. This means that the model is auto generated as a “Digital Twin” of the system to be tested, and from there the actual tests are auto generated and executed. This further simplifies and optimizes testing to improve quality and reduce release times and, as a result, will become the de facto approach to testing in 2023.
Sustainability of Testing
Traditional test automation is based on the need to run a large number of fixed tests at defined periods (overnight, weekends and prior to a release, for example). The execution of each test requires significant computing power and thus has both an energy cost and an environmental impact. With the increase in energy prices and the greater awareness of sustainability, this legacy approach of “non-intelligent” test automation will be replaced by intelligent test optimization – in which the goal is to only run the tests that are known to identify a problem.
Metaverse as a Platform
Vendors need to maintain multiple different channels to engage with customers; the web and mobile are the most common, but there are also dedicated mobile apps, kiosks, IoT devices, ATMs, Set To Boxes, etc. In 2023, expect more conversation around the Metaverse as a significant channel for future customer interaction. This adds a significantly different type of channel for vendors to design, build and test for – meaning that most will need a mobile app, a website and a metaverse implementation. Non-Fungible Tokens (NFTs) can be attributed to Metaverse environments, providing a new way to deliver products and services. Testing the Metaverse adds significant challenges for most test automation technology, which in turn will accelerate innovation in that space.
AI to Provide Assurance of Quality and Behavior
With the increasing complexity of a digital-first world, digital products will come under greater scrutiny. This is already high for safety critical systems but expect it to increase in all areas in 2023. The contents of the product, including all constituent parts and third party components, must be itemized and certified – ensuring that all constituents are authentic and original. As products become more intelligent and AI is more prolific across systems and devices, their behavior becomes more nuanced and complex. The testing of these systems needs a more intelligent technology to understand the responses and validate against acceptable behavior – resulting in the need to use AI to test AI.
The Rise of the Citizen Developers
With an increasingly digital savvy population, traditional non-technical audiences are becoming technically proficient and confident in using more complex systems. Combined with advances in user experience design and usability improvements, these new non-technical users can develop for their specific needs through low-code or no-code technology – in many cases removing the need for a separate requirements document for a technical team to implement. This reduces time to delivery and the risk of misinterpretation and increases overall efficiency.
Traditional test automation still requires that the tests themselves are manually written. Model based approaches allow tests to be generated from the central model; Autonomous Test Design takes the next step and automatically generates these models. This means that the model is auto generated as a “Digital Twin” of the system to be tested, and from there the actual tests are auto generated and executed. This further simplifies and optimizes testing to improve quality and reduce release times and, as a result, will become the de facto approach to testing in 2023.
Sustainability of Testing
Traditional test automation is based on the need to run a large number of fixed tests at defined periods (overnight, weekends and prior to a release, for example). The execution of each test requires significant computing power and thus has both an energy cost and an environmental impact. With the increase in energy prices and the greater awareness of sustainability, this legacy approach of “non-intelligent” test automation will be replaced by intelligent test optimization – in which the goal is to only run the tests that are known to identify a problem.
Metaverse as a Platform
Vendors need to maintain multiple different channels to engage with customers; the web and mobile are the most common, but there are also dedicated mobile apps, kiosks, IoT devices, ATMs, Set To Boxes, etc. In 2023, expect more conversation around the Metaverse as a significant channel for future customer interaction. This adds a significantly different type of channel for vendors to design, build and test for – meaning that most will need a mobile app, a website and a metaverse implementation. Non-Fungible Tokens (NFTs) can be attributed to Metaverse environments, providing a new way to deliver products and services. Testing the Metaverse adds significant challenges for most test automation technology, which in turn will accelerate innovation in that space.
AI to Provide Assurance of Quality and Behavior
With the increasing complexity of a digital-first world, digital products will come under greater scrutiny. This is already high for safety critical systems but expect it to increase in all areas in 2023. The contents of the product, including all constituent parts and third party components, must be itemized and certified – ensuring that all constituents are authentic and original. As products become more intelligent and AI is more prolific across systems and devices, their behavior becomes more nuanced and complex. The testing of these systems needs a more intelligent technology to understand the responses and validate against acceptable behavior – resulting in the need to use AI to test AI.
The Rise of the Citizen Developers
With an increasingly digital savvy population, traditional non-technical audiences are becoming technically proficient and confident in using more complex systems. Combined with advances in user experience design and usability improvements, these new non-technical users can develop for their specific needs through low-code or no-code technology – in many cases removing the need for a separate requirements document for a technical team to implement. This reduces time to delivery and the risk of misinterpretation and increases overall efficiency.
