5 ms·
I taught programming a long ass time ago, a Java and C++ class. While I get that these languages are fairly complicated, the things people got stuck on was unde
by rhacker 8y ago
I taught programming a long ass time ago, a Java and C++ class. While I get that these languages are fairly complicated, the things people got stuck on was understanding control flow, expressions and data storage.
In other words, it wasn't the syntax, but the concept of a program running from top down (including loops or conditionals), how expressions work, and what a variable is and means. These fundamental concepts are sooo far beyond the comprehension of people it took a LOT to get down. After that stuff was understood, learning curly braces, keywords or indenting (for style not meaning) was so damn easy for people.
So I would expect an evidence based programming language to look completely different than this.
Edit:
It looks like a primary use-case for this language is for blind people - for which use-case I know nothing about. It wasn't clear from the Quorum language website, but it is clear from the video linked by another commenter.
That definitely would affect how I would have written my comment if at all. Take my original comment with a grain of salt that may not even be relevant here.
- nickpsecurity 8y agoOn top of that, it should actually be successful in education. An example of one designed for education with some good results is Pyret: https://www.pyret.org/ https://www.pyret.org/
- mwcampbell 8y ago> Edit: It looks like a primary use-case for this language is for blind people - for which use-case I know nothing about. That does appear to be the original use case. Even now, Quorum seems to be most successful in that niche. In case anyone reading this isn't aware, blind programmers are working successfully in all kinds of programming languages. On that topic, check out this comment by one of my blind programmer friends [1]. I used to think that Python would be unusually challenging for blind people, due to significant whitespace, but many blind programmers are perfectly productive in it. And the NVDA [2] screen reader for Windows, which is developed primarily by blind people, is written in Python (and some C++ where necessary). Nobody made them use Python; they chose to. And they've attracted a thriving community of contributors and add-on developers. So syntax isn't really a problem even for blind people. It seems that Quorum has had some success in helping blind students start learning to program. But I'm skeptical that a new language was really necessary or even helpful. I think a mainstream language would have been fine. The tailored IDE, libraries, and tutorials are probably bigger factors. [1]: https://news.ycombinator.com/item?id=9284744 https://news.ycombinator.com/item?id=9284744 [2]: https://www.nvaccess.org/ https://www.nvaccess.org/
- jareds 8y agoThanks for linking to that comment. I'm glad to know I'm not the only blind person who doesn't understand how I don't get into a fatal car accident at least once a week.
- PinkMilkshake 8y agoMicrosoft has put a bit of effort into accessibility in Visual Studio. This video demonstration blows my mind every time I watch it: https://www.youtube.com/watch?v=iWXebEeGwn0 https://www.youtube.com/watch?v=iWXebEeGwn0
- jareds 8y agoI'm a totally blind programmer who saw this start to develop based on email list postings. As far as I remember originally the plan was to make Netbeans more accessible with screen reading software. This was done by creating Sodbeans, https://sourceforge.net/projects/sodbeans/ https://sourceforge.net/projects/sodbeans/ I tried it out several years ago, and didn't find it to be worth using. By that point I was already quite comfortable with Eclipse, and Netbeans/Sodbeans didn't offer anything to make it worth switching. After working on Sodbeans I guess the researchers decided that they could only do so much to make Netbeans accessible so created a new language. By that point I was already a fairly experienced programmer, so lost any interest since I didn't see the point in learning a new language created at least partially based on the needs of blind programmers that I would most likely never use professionally. I don't have any opinion on whether this is useful since I taught my self to program using GW Basic 20 years ago so have no idea how someone who is blind would go about learning programming in 2018 with current technology.