Well, it kinda depends, doesn't it?
If you're talking about code, then yes I agree 100% that it doesn't make very much sense.
But if we're talking about drag-n-drop to build a GUI, it makes very much sense to use a visual tool to define a visual thing (e.g. a dialog box, or an application window) instead of fiddling around with snippets of code to build it.
Perhaps the most important distinction between the two would be to say that the GUI builder is declarative, and the manual code way of building a GUI is imperative. I think that's important.
That kind of thinking could also be brought back to the original question of drag-n-drop coding. While it may be inherently bad to do this for imperative coding, maybe it works better for a declarative style?
Thoughts?
/J
| [reply] |
When you use those top-of-the-line VC++ gui builders, where a right-click on a widget brings up a submenu listing EVERY attribute of the widget, (from bindings, callbacks, colors, etc), the GUI-builder can be very effective. BUT.... all the free ones(and most opensource) I see
have very limited sub-menus. I suppose eventually we will get one that is cheap, widely available, and works on all platforms. We all dream of such a programming tool, but it is still elusive. I won't pay to find out either, they all promise alot of features, but when you get to the nitty gritty, they disappoint. ;-) One or two month trial periods is not enough. Who is going to devote a month full time to testing a gui builder,
knowing you have to pay if you don't decide fast enough.
| [reply] |