4 ms·

(Appending this to my own comment as a way of acknowledging the comments made in response to my question) I stand convinced that, though the majority of people

by grellas 13y ago

(Appending this to my own comment as a way of acknowledging the comments made in response to my question) I stand convinced that, though the majority of people will not be necessarily learn to code in the sense of developing full-fledged programs, there is much to be gained by exposing young people generally to the concepts behind programming and to encourage them to learn these concepts: this will inspire some to make programming their work and it will benefit all by giving them at least a basic working literacy about how the digital world around them actually works. Thanks to all for your thoughtful and instructive responses. HN is truly a great forum.

dools 13y ago
basic working literacy about how the digital world around them actually works I think it's useful to draw a distinction between "coding" and automation, and I wrote about it here: http://www.workingsoftware.com.au/page/When_does_automation_become_coding http://www.workingsoftware.com.au/page/When_does_automation_...
- erichocean 13y ago
  +1 I have put a great deal of effort into improving the level of automation that can be performed without writing code, or working with graphical mathematical structures (such as wiring together nodes in a graph, or building a flow chart). The work done on Logo, a language designed to teach children how to program, has been particularly helpful. In the system I'm developing, everyone is using a "debugger" (without knowing that's what it's called) even in the process of getting onto the system. A lot of the progress we've made has involved aggressively simplifying the development process into something that's at the level of using a spreadsheet (and just as concrete for the user). For example, if you have to develop an API, you've already excluded most of the people from being able to use your tool. API design is hard and takes a great deal of experience. That's the opposite of what it takes to get productive value from a spreadsheet. I agree that there's a huge amount of benefit a user can get from being able to automate their own work, and to be able to easily and effectively share that automation with others. Today's scripting languages aren't even close to the "spreadsheet" ideal I think is possible.

by grellas 13y ago

dools 13y ago

basic working literacy about how the digital world around them actually works I think it's useful to draw a distinction between "coding" and automation, and I wrote about it here: http://www.workingsoftware.com.au/page/When_does_automation_become_coding http://www.workingsoftware.com.au/page/When_does_automation_...

erichocean 13y ago
+1 I have put a great deal of effort into improving the level of automation that can be performed without writing code, or working with graphical mathematical structures (such as wiring together nodes in a graph, or building a flow chart). The work done on Logo, a language designed to teach children how to program, has been particularly helpful. In the system I'm developing, everyone is using a "debugger" (without knowing that's what it's called) even in the process of getting onto the system. A lot of the progress we've made has involved aggressively simplifying the development process into something that's at the level of using a spreadsheet (and just as concrete for the user). For example, if you have to develop an API, you've already excluded most of the people from being able to use your tool. API design is hard and takes a great deal of experience. That's the opposite of what it takes to get productive value from a spreadsheet. I agree that there's a huge amount of benefit a user can get from being able to automate their own work, and to be able to easily and effectively share that automation with others. Today's scripting languages aren't even close to the "spreadsheet" ideal I think is possible.