A list of basic concepts the TA Toolbox is built on.

The concepts layed out below are the result of years of programming data analysis software with Matlab™ in a scientific context. Whereas there is more than one way to do things, many basic building blocks of the toolbox follow a common idea that is not always apparent from the code itself. Therefore, these pages may help new users with a deeper interest and potential new developers to get an idea of what all this might be about. And of course, it is not all my ideas, I learned a lot from colleagues and other people. Nevertheless, the flaws in design and implementation are solely my responsibility.

Please note: Although the toolbox tries to follow the ideas layed out below, sometimes reality may still lag behind.

• Interoperability
  (both between different operation systems and Matlab™ installations)
• Dataset as fundamental unit
• Modular design → easily extensible
• Highly configurable (config files)
• Clear distinction between GUI and functions
  (all functions that work on datasets can be called from the CLI)
• Metadata for datasets
• Robust and error-tolerant
• Export of near-publication ready figures (at least well-suited for presentations)

A list of GUI-specific concepts:

• Intuitive design
• Integrated help
• Loads of keyboard shortcuts (for faster operation)
• History of every operation
• Complex tasks in subwindows