On Mon, 21 Apr 2025 18:21:32 +0200 ichthyo <p...@ichthyostega.de> wrote:
>On 14.04.25 13:15, Will Godfrey wrote: >> I think I've worked this out now. >> >> I've done some reordering of the lists for better separation. >> >> In the CLI we have: >> previous entry >> current input >> line to-send >> >> Each of these have three entries representing: >> root word, >> section, >> offset. >> >> At startup the sets the last see to: >> main >> (empty) >> (empty) >> >.... >> Whenever a successful entry has been sent it is copied to the previous one. > >What I don't yet fully get is how the user interacts with that mechanism. >If I understand correct, the idea is to have a very simple control structure >which builds up a "Sentence" -- once that is complete, a matching window >in the GUI is selected to receive positioning commands, correct? Yes, although once at a specific level you won't need to repeat whole lines e.g: voice show (display the voice window) X 50 (move it horizontally to pixel 60) width 750 (set its width to 750 pixels - scaling the height accordingly) sub hide (switch to the subsynth window and hide it) pad Y 40 (move the padsynth window vertically - if it's already visible) main show (it's always visible so just bring it to the top) >Now, let's assume I am a rather clueless CLI user and don't have a grasp >of the concept behind this mechanism. Let's assume I started playing around, >do I get any clues as to how I can move forward in this "game"? > >I.e. what kind of feedback does the user get while in the middle of that >selection procedure? How to make all this stuff obvious to users is always a problem! A curious point is that as the GUI only maintains one copy of any window type you don't actually need the respective section to be active to show it! I'm not sure what implications this has. -- Will J Godfrey _______________________________________________ Yoshimi-devel mailing list Yoshimi-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/yoshimi-devel
