Theme
Description
Participants positively highlight that the data pipeline code is written in human-like language and can be read like english text, which is especially important for novice programmers (PL5.1 - Syntax is important for novices).
Both Python and Jayvee are mentioned in a positive light, to no surprise with Python generally being considered one of the easier to read general-programming languages.
An effect that contributes to the feeling of reading human language when reading Python code is that it requires comparatively few special characters (see also PL5.4 - Expressive and verbose syntax]). Syntax that is not related to the data pipeline, such as import statements (see PL6.4 - Meta infos (e.g., imports, library names) obscure the logic of the pipeline) seems to have no effect on how human-like the language is perceived, potentially because these syntax elements are only found at the start of a data pipeline model.
In contrast, Jayvee makes more heavy use of special characters but seems to be equally easily readable due to how close the mental model of an ETL pipeline aligns with Jayvee.
Representative Quotes
The data pipelines written in jayvee is easy to understand and easy to read as English text. It is helpful while reading the code. It really understandable and feasible.
- S14
Jayvee has a very human like language, almost like pseudo code which can be immediately understood even by non-programmers in my opinion as long as they have a basic theoretic knowledge about pipelines.
- S31
(Python) Easily readable, close to natural language. No unnecessary syntax.
- S38
Jayvee seems to be quite less computer language syntax and more abstracted English texts for each of the block of the pipeline. Thus, making it a lot easier to understand as it is pretty much just reading exactly what is happening in the pipeline.
- S50