What best characterizes interpreted languages?

Interpreted languages are specifically designed to be executed line by line or statement by statement by an interpreter, which makes option B the best characterization. This means that as the program runs, the interpreter reads a line of code, interprets it, and executes it immediately. This approach allows for more flexibility during development since changes can be made and tested in real-time without the need for a separate compilation step.

In contrast, languages that require compilation, as mentioned in one of the other options, involve translating the entire code into machine language before any execution can occur. This is a defining feature of compiled languages, which shows how interpreted languages differ fundamentally in their execution method.

Regarding environment dependence, interpreted languages typically have a better cross-platform capability, but they can still depend on the presence of the interpreter on the target system. Instead of being platform-dependent, many interpreted languages can run in various environments provided the appropriate interpreter is installed.

Finally, while some languages might be limited to specific platforms based on how their interpreters are designed, many interpreted languages are crafted to be portable across systems, which contrasts with the notion presented in another option. Therefore, option B stands out as the most accurate representation of interpreted languages, highlighting their unique execution style.