Aside from everything else, visual programming as it's usually implemented is inherently slower than typing text when the latter is not ‘hunt-and-peck’. Because

Aside from everything else, visual programming as it's usually implemented is inherently slower than typing text when the latter is not ‘hunt-and-peck’. Because, after some experience with the keyboard, your brain knows exactly how to jerk a finger to get a particular character and can do that very fast (touch typing particularly excels at this since the normal positions of the fingers are fixed—and it's aptly called ‘blind typing’ in some languages). Meanwhile, to lay stuff out on the screen you need to 1) find with your eyes where to grab a block and 1b) where to put it; 2) move the mouse to grab the block and 2b) to drop it, and sometimes also do that with connectors too. Both of these kinds of operations—visual and manual—are way slower than mechanical jerking of the limbs, especially due to them being prone to error. Even worse if you have to drag-and-drop instead of just clicking. All this fiddling requires you to pay close attention to things on the screen, while typing mostly allows the visual system to coast along or tune out altogether. Multiply this by hundreds of words a day, and you'll see that visual programming is in large part an exercise in mouse-brandishing amid visual load, in the vein of FPSes. And usually you have to regularly move the hands to the keyboard anyway, to name your blocks, variables and whatnot. On phones, the difference shrinks since typing relies on just two thumbs with no fixed position and no physical keys—while onscreen manipulation gets more immediate compared to a mouse. However, phones suffer from short supply of screen area where you pick and place the blocks. In my experience, it would still make sense to choose blocks with typing (by filtering in real-time from the entire list), and to minimize the work of placement—like in Scratch or Tasker. Visual programming might have distinct value if it allowed to manipulate higher-level concepts compared to text-based coding. However, it turns out that the current building blocks of code work pretty well in providing both conciseness and flexibility, so visual blocks tend to just clone them. Again, the situation is better is you can ride on some APIs specific to your problem domain—like movement commands in Scratch or phone automation in Tasker and ‘Automate’. Similarly, laying out higher-level concepts like in UML or database diagrams has its own benefit by making the structure and connections more prominent.