Thank you! That all makes sense to me… except “useless”.
It’s not actually a type of data and shouldn’t be put on the same dimension with other types mentioned. Useless data can simultaneously be ordinal, or interval, or any other data type. It actually depends on the context: for some tasks one piece of data might be useless, still being important for the others.
One other question I have is how does this classification match with higher-level data types like visual data (images), audio, texts etc. Probably there should be some way to reduce them to one of the types you describe, but is there only one way of such “mapping” or it really depends on situation? I know it might be unrelated and silly question, but when talking about data types in machine learning, images and large texts are what comes to my mind first. So we might take that into consideration for more “multilevel” classification. Just a thing to ponder about :)