On the micahel blog (http://blogs.msdn.com/micahel/), we can see two concepts related to the model: Linear-linear design and pom, which are checked on both Baidu and Google. There are very few links, I don't know if micahel is a self-developed product. Because these two concepts have something to do with testing, I will sort them out based on my own understanding for your reference. Pai_^
First, let's look at the linear model. It is the abbreviation of the logical functional model. After translation, it is the logical functional model. Specifically, a model is obtained based on the user's behavior, and then the test case is designed based on the model. In this way, the design case is very consistent with the user's actual operations, this ensures the test efficiency. I used this method during the review, and it still seems quite effective.
Let's look at pom again. It is short for physical object model. It is translated as a physical object model, a concept related to UI-based testing. For the UI, it can be seen as a combination of multiple interface elements. The specific functions can remain unchanged, but the interface elements corresponding to the functions will change. So how can we test this situation? In this way, Pom is introduced to abstract functions from specific interface elements. In this way, the designed use cases do not need to be changed even if the interface elements change.
The above is one of my understanding of these two concepts, which is not necessarily completely correct. We also need to try and summarize them in practice. We also welcome your comments. Aside from the two concepts, the ideas contained in them are quite interesting. Pai_^