3 ms·
Edwards hasn't seemed to have solved the core issue with so-called graphical (Edwards is calling it "non-textual") programming languages: they're clumsy to the
by Donald 17y ago
Edwards hasn't seemed to have solved the core issue with so-called graphical (Edwards is calling it "non-textual") programming languages: they're clumsy to the point of being unusable.
The table visualization might seem useful for some, and it would perhaps make a nice feature/plugin for an IDE. But, frankly, I would rather muddle around with boolean algebra than clicking around his GUI.
- alphazero 17y agoI agree with you regarding the clumsy gui -- UML editors suffer from the same general problem. What I took away from his presentation were the 2 core points: 1) linear expression is intrinsic to text but not intrinsic to programs. 2) the machine is more than a tablet. Who knows? May be the ideal interface is actually voice based ...
- GrandMasterBirt 17y agoI disagree... A clumsy GUI is no excuse to dismiss the idea. We can always make a better GUI. The point is this: If we can let the computer do the work, including during programming, then maybe we can focus more on design of the logic rather than keeping the entire programming model in our brain all the time and having to build a new one every time we hit new code. Its definitely a very nice idea. Actually I can think of one program I worked on which is really a logical configuration tool which could use such a language instead of what it does and it will be simplified many times over.
- cubicle67 17y agoFor complex boolean rules, this looks good, however most code that I deal with has pretty simple conditions. Personally, I prefer plain text, then I use pencil and paper when things get complex - best of both worlds. Although, having something like this that could be involved when required, and was smart enough to construct the table from existing code, would be useful.