4 ms·

This is a great article. It is great to know that there are other people like this. I always have had these projects that I can't forget off. There have been va

by nzonbi 14y ago

This is a great article. It is great to know that there are other people like this. I always have had these projects that I can't forget off. There have been various, but after I matured, I ended up focusing in two. They are still early, I have been unable to dedicate on them as much as I would have liked. Because I have lived most of my life with limited means. I used to not own a computer, nor internet. While I worked to save my family from poverty, all I could afford to do, was to buy books, and learn programming, and work on my projects in theory. It took a long time, but now I am relatively well off, and have a great full time work. I am advancing my projects, but it is too slow. I am considering to at some point, to take maybe a full year or two, to focus on my projects. My two (still early) projects are: Reproduce the human mind, and emotions, in a computer. Human like AI: http://www.ozkeebo.com/soul.project/ http://www.ozkeebo.com/soul.project/ This is a problem that have fascinated me since child. I have meditated a lot on it, and have this hunch that I can pull it off. I am the kind of crazy and weird person, perfect for it. But I know that it is a hard problem. Maybe I will never be able to complete it, but at least I am sure that I will get great things out of it. Let's see how I will be doing in 10 or 20 years. My other project, is a graphical programming language. This comes from my dissatisfaction in how we express programming languages. I feel like I want to shout to the world, that we are missing a huge opportunity here. I feel like we could be doing much better than what we have. But it is a lot of work, and it will also take me years of work, to prove if I really have a point. I will see. Thanks for a great article, and a great hn thread anyway.

ionforce 14y ago
So what are your hunches that lead you to believe that a graphical programming language is the way to go? Not saying that it couldn't, but it hasn't really been a success so far.
- nzonbi 14y ago
  I am 100% sure, that at least for me, I can imagine and design a graphics language, superior to text representation of programs. And I expect that for most people, once they surpass the learning curve, will also be better. And even more sure about it working for people new to programming. We can represent programs using texts. text is a good way to represent sequences of information. So it fit programs well, to some extent. But text is limited. Programs contains many kinds of information, that don't fit well with a textual representation. Hierarchical data, scope, branching flow, loops, etc. After you realize this, you can also replace text commands, with symbolic representations that are easier to parse visually, and that can convey their function better than just text. You need to maintain an appropriate level of information density. This one of the thing that has made text representation of programs successfully. The other key thing is the UI. We are highly productive in manipulating text with editors. A successful graphical language has to beat that. It has to make possible to manipulate the information more efficiently: faster, and with less muscular effort. Allow me to call this the field of "unconventional UIs". When I was a kid, I only had access to computer for a brief time. But I had plenty of access to console games. I used to love videogames. I spent huge deals of time detailed studying them, drawing concepts, etc. Examining why I found most of them to be bad, some average, and a few gems. A key thing was the UI. The mechanisms to manipulate the game environment. Well, videogames are a rich field of unconventional UIs. And it is a field that I know very well. My experience with all these game UIs, makes me pretty sure. A better, more efficient UI can be designed, to manipulate programs information, on a graphical environment, than existing UI from text editors. that's it in short.

by nzonbi 14y ago

ionforce 14y ago

So what are your hunches that lead you to believe that a graphical programming language is the way to go? Not saying that it couldn't, but it hasn't really been a success so far.

nzonbi 14y ago
I am 100% sure, that at least for me, I can imagine and design a graphics language, superior to text representation of programs. And I expect that for most people, once they surpass the learning curve, will also be better. And even more sure about it working for people new to programming. We can represent programs using texts. text is a good way to represent sequences of information. So it fit programs well, to some extent. But text is limited. Programs contains many kinds of information, that don't fit well with a textual representation. Hierarchical data, scope, branching flow, loops, etc. After you realize this, you can also replace text commands, with symbolic representations that are easier to parse visually, and that can convey their function better than just text. You need to maintain an appropriate level of information density. This one of the thing that has made text representation of programs successfully. The other key thing is the UI. We are highly productive in manipulating text with editors. A successful graphical language has to beat that. It has to make possible to manipulate the information more efficiently: faster, and with less muscular effort. Allow me to call this the field of "unconventional UIs". When I was a kid, I only had access to computer for a brief time. But I had plenty of access to console games. I used to love videogames. I spent huge deals of time detailed studying them, drawing concepts, etc. Examining why I found most of them to be bad, some average, and a few gems. A key thing was the UI. The mechanisms to manipulate the game environment. Well, videogames are a rich field of unconventional UIs. And it is a field that I know very well. My experience with all these game UIs, makes me pretty sure. A better, more efficient UI can be designed, to manipulate programs information, on a graphical environment, than existing UI from text editors. that's it in short.