Hi John/all. I still have some Git exercises and maintenance to do tonight, but I tripped over this http://planet.factorcode.org/ and want to see where it might lead.
The syntax highlighting is interesting to me. I'm wondering whether we can change the Listener GUI into a color vocab browser that focuses more on code and less on the navigation links and doc and examples, as in the current Browser. There is a place for those too, obviously, but I'm trying to change the emphasis a little to make development faster and to reduce the amount of typing of words into the Listener. One possibility: Split the Listener vertically, placing the usual Listener command line progression on, say, the right side, and another pane on the left. The idea is that, when you need to know what a word means, because you are trying to use it, or are wondering whether you even have the right word or concept in mind, you can type it into the Listener, and instead of doing \ some-word see you could instead dynamically search for the word's definition by either clicking on the word or just hovering over it. I'm thinking that if you access and display the word's definition on hover-over, instead of on click-on, you could rapidly scan colored formatted word definitions in the left pane. You could have, say, a line of words entered into the Listener, scan over them left to right (or whichever way) with your mouse, looking at their color definitions flash through the pane on the left. When you hit a word that is especially interesting/confusing, you click on it to lock that definition into the left pain. You can now wander up to the definition in the left pane with your mouse and continue the process of scanning over its words, drilling down toward primitives, until you are convinced you understand enough to complete the expression at hand. Every time you want to drill, stop scanning with your mouse and click on the interesting word to popup yet another pane below the one you were in when you clicked on the interesting word. Or, you could stack your definition windows on top of each other, as you drill, and provide a context menu "back" command, or just use some forward/backward arrows in the GUI to go back and then go forward again, just as you would in the HTML help Browser. So why do this? If you are in a coding app and not a browsing app, you could not only scan and study as you drill, you can also scan and edit the command line you are building. Maybe Ctrl-click on a word to append it to the end of the Listener's command line. (Maybe the other way: Ctrl-click for doc and click for appending.) I want to get around faster, and to assemble correct Factor expressions, the first time. I don't want to see an example for a word before I see the code for the word. The reason for this is subtle but important: I want to understand the words increasingly quickly when I read them and their definitions, and not by remembering necessarily several favorite examples, which memories I will acquire anyway if I can use the word successfully in my own word definitions a few times. Having a bad time reading words creates a drive to improve the words, especially because you are now reading definitions more often than examples (which I think I would want to put in a nearby pane or in an examples tab in the same pane). I don't want to use English explanations and examples as crutches for reading Factor fluently. I want to write better words and patterns of them. Being required to read definitions first may be a healthy source of pressure in that direction. I'm also thinking that the stack-effect data could be more binding than it currently is. I noticed while compiling some new words that the compiler will infer a stack-effect, and tell you when you've written it incorrectly. This looks almost binding. So my suggestion/question is: Can we use the stack effects to check at least the superficial (type- and count-based) correctness of a forth expression, as it is assembled on a Listener command line? Also, concerning literals and stack objects, when you mouse-scan those, you could get an inspector, also on the left in its own pane, under your last viewed definition pane. Click on the literal or stack object, and your inspector locks in place for further study and drilling. Similarly provide arrows to go back and forth through your stacked inspectors, so that you don't clutter the screen with windows, thereby created larger saccades, which often reduce thinking- and coordination- efficiency (at least for me). I'm open to suggestions on any aspect of this, but my overall objective is to integrate the contents and value of the help Browser into an expanded Listener GUI, so that coding and learning are faster, because all the code is alive, connected, colored, formatted, and append-able at the Listener's prompt, with minimal GUI gesturing and less typing. Thoughts anyone? Shaping ------------------------------------------------------------------------------ Centralized Desktop Delivery: Dell and VMware Reference Architecture Simplifying enterprise desktop deployment and management using Dell EqualLogic storage and VMware View: A highly scalable, end-to-end client virtualization framework. Read more! http://p.sf.net/sfu/dell-eql-dev2dev _______________________________________________ Factor-talk mailing list Factor-talk@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/factor-talk
