This is a translation of my article for the “Testkrant Jaargang 2012 Editie 1“. The article is based on my experiences at Polteq.
The world changes and this sets new requirements for testers, especially the technical aspect of testing. Cloud computing enables us to use functionality from the cloud. For cloud computing we need to test if the functionality of the cloud service correctly supports the business processes, but we also need technical tests to check if the service correctly integrates with the existing systems. We also need to monitor the service in production by using test automation, which needs technical skills.
The other big “hype” that requires technical testers is Agile. By using multi disciplinary teams, the tester gets closer to the developer. Direct communication between these two functions will increase since less documentation is used. To enable good communication, the testers will need to know something about development and vice versa. The iterative and incremental character of Agile requires to execute regression tests more often. This demands for more test automation. This article will describe some situations for cloud, Agile and test automation that will clarify the need for technical testers.
Cloud
A growing number of companies uses services from the cloud (e.g. e-mail, CRM or environments). All services are provided via the internet and should be integrated in the current business processes. Testing the communication with the service is a technical task. The focus will be on the messages that are used to communicate with the service which requires looking below the GUI.
Using cloud services has an impact on internal development projects. Whenever we need to test chains of events after internal changes a connection to the (live) service is often unwanted. So we need to use a stub or mock to simulate the cloud service in order to test our processes without touching the production version of the cloud service.
A service comes with agreements and contract with the supplier about performance and other aspects of the service in production. We need to check in production if the supplier will keep up to the agreements. There is the need for frequent (or even continuous) tests that will monitor the service. This implies test automation which will be covered further on in this article.
Agile
Agile uses four simple basic priciples:
- Individuals and interactions over processes and tools
- Working software over comprehensive documentation
- Customer collaboration over contract negotiation
- Responding to change over following a plan
The importance of interaction can be found within the multi disciplinairy teams. Less documentation can only work of the team members communicate well on what they are doing. To understand what the other team members are doing, you will need to know about the other disciplines. We expect from our testers that they have a basic knowledge of programming and have some basic programming skills. This will help to discuss about software architecture and possible solutions, which will be defined by the team.
The short development cycles force us to test incomplete products. The tester will need to apply the use of drivers, stubs and mocks to be able to test. Creating or setting up these facilities to support testing needs to be signaled (and preferably also executed) by the testers.
The working software that gets delivered every iteration will need to work together with the previously delivered items. Each iteration the functionality will increase which also reflects on the size of the regression test set. This set grows, but the time to execute it is only limited. The frequent execution of the regression test set, together with the time pressure, make a good business case for test automation. You should not think to lightly of test automation within Agile.
Another requirement as a result of the frequent deliveries is good version and configuration management. The testers will need to help to make this a succes. When and how to integrate? Which demands do we have for the different environments? Technical knowledge will help to get a better and clearer view of what should happen.
Test automation
No need to explain that test automation has technical aspects. Since test automation is an important part of testing with cloud computing and within Agile, test automation deserves some special attention.
Test automation can occur on different levels. The foundation for test automation is set with the unit test. These unit tests are in general created by the developers, but this does not mean that testers are not involved. We can help the developers apply test design techniques, since we have more knowledge of testing. Pair programming of a tester with a developer can be of value here. The tester can assist the developer in thinking about which test cases need to be created. In addition to this assisting, we can also review the unit tests when we have enough knowledge and skills with respect to the programming language.
Functional test automation is closer to the testers. With Agile development the graphical user interface keeps changing to meet the (changing) requirements, so succesfull automation needs to be at a layer below the GUI. So once again we need to know how to manipulate the code to be able to test the functionality. When automation is used that does use the graphical user interface, this GUI must be manipulated in some manner. When we use Fitnesse or Cucumber to achieve this, it seems like little or no technical work, but we do need some programming. Both Fitnesse and Cucumber need an interpretation layer to talk to the test object. Allthough both come with a lot of standard functions and options, they almost never fit completely for your specific situation. Therefore we need to add some functionality, which needs to be programmed. More and more it is expected from the tester that they can do this.
Multiple contexts require continuous testing. Within Agile for the integration and regression, with cloud for monitoring the supplier, etc. The only way to achieve continuous (or at least very frequent) testing is by automation. Thinking about how to realise this and actualy implementing it requires knowledge of architecture, programming and testing.
Conclusion
The software we create is getting more and more complex. This does not only require more technical knowledge from the developers, but also from the testers. In test jobs we see requests for programming skills (preferably in multiple programming languages) and people expect the testers to be able to discuss about architectural problems found during development. The integration of different systems, as seen with cloud computing, requires the testers to be able to read technical logging, use drivers, stubs and mocks, etc. The closeness that we find in the Agile teams implies more technical discussions and communication.
The tester needs to keep up in the quickly changing world, where Agile and cloud are becoming the standard. The only way to keep up with the speed of development is to use test automation. The tools that will be needed for this should not only be used, but also understood and implemented. So yes, the need for technical testers grows.
Keep learning and broaden your knowledge, so that we – as a test community – can fill this need!
asked at the end of the webinar. I want to compliment Ruud with how he handled the questions! As the result of this webinar, I hope that people realise that testing really needs to adapt to the way that software is used nowadays. Much uncertainty is introduced with the use of cloud computing, but by selecting the right service and supplier you can allready mitigate a lot of risks. Don’t forget to keep testing in production, to check if “everything still works”. One of the questions mentioned that this is monitoring and not testing. I think we need to broaden our definition of testing and make monitoring a part of testing. Realise yourself that the maintenance department might not monitor on the service, since it is maintained by the supplier… I think what needs to be done in production is testing! A continuous end-to-end test to check if the processes still work, is that monitoring or testing? Doesn’t the word test in E2E test imply a testing activity?
Testing Experiences 