VS code is also accessed via the keyboard. The main key combination you need to know is Ctrl + Shift + P command Panel , From here you have access to all functions of VS Code, including keyboard shortcuts for the most common operations.
The command Panel Allows access to many commands. You can execute editor commands, open files, search for symbols, and get a quick overview of a file, all in the same interactive window. Here are a few tips:
another example from fman
The quickest way to explore fman abbreviations is via command Panel :
Just press Ctrl + Shift + P (or Command + Shift + P for Mac) open in fman.
After the command palette (see above) for power users and developers in some apps (vscode, sublimetext, some Pythonide, jupyterlab … etc.) has become a normal user interface, I think that in every complex program with many hidden tools / functions a range of commands would be very useful.
In my opinion, a modern range of commands serves three different purposes
- Recognize commands / functions based on the desired actions (i.e. duplicate a file, export a tree view, create a timestamped zip file, … etc.)
- Discovery of links
- Executing commands that have no obvious visual interface or are deeply nested and would require a few clicks. Or for the lazy ones who type faster, then move and click the mouse and don't remember the shortcuts.
The way I see human memory is to map the basic dictionary object (key, value). And our intention is directly related to the action we want to take. Typically, in any software, a user must learn to map their intent to the learned behavior of steps that must be performed in the software to achieve their goal. (Let me know if I can explain all of this better). I find that the command palette shortens this approach and enables a faster learning experience.
I personally find it easier to remember words and actions that I want to achieve than the steps I have to take to achieve them. So I often have to google the exact recipe and run it manually. A command palette helps considerably and reduces the effort. Google has also greatly influenced people to move from structured information to queryable (?) / Searchable information. I've also always thought that the command line is great to run when you know what you want, and the UI is great to recognize.
As for the future, we move from learning behavior to dictation and let the software figure it out. I think that's all about assistants.
So my question is whether this is only useful for certain groups of people (i.e. developers) or whether it is generally applicable to the general population.
MacOS has the help search function, which is similar, or the action item, which is pretty good.
If someone wants to implement such a function in software, what are the guidelines from a UX perspective?