18 comments

[ 3.3 ms ] story [ 49.3 ms ] thread
This is a blog post telling the story of Allcancode that started with a coding game featured in code.org and played by millions of kids around the world and then transformed to a tool provider that streamlines software development.
No, it's the story of a guy with passion behind an unmarketable idea.

Now they are desperate to keep their 'AllCanCode' trademark alive!

It's sad really.

it apears they have not entered the market nor created a prototype yet though. maybe the next pivot will be a platform where designers can find developers. youre very lucky as a dev to find a good designer and vice versa.
We will start with a private beta in about a month. The tool is already in use in-house to deliver real projects.
I wonder whether there is a Go4-like approach to control Marco, you know, that kind of PatternPattern that makes your code hard to read.
There are many things that could be done with Marco as experiments. Companies cannot survive on experiments, though :-) so we had to move on.
Did you build your tool with your tool? That's really the only question that matters as far as I see it. If not, why not?
> That's really the only question that matters as far as I see it.

Why do you feel this question matters at all, let alone be the only question that matters?

The Java compiler is written in Java, Rust compiler in Rust, Windows OS is developed on Windows PCs. Eating your own dog food, this is called.
If it is being proposed as a new way of doing general software development, yet it could not create the system itself, then it has clear and profound limitations. Most markedly in the case of visual programming paradigms, it means they haven't solved the problems that have tripped up visual programming all along - what happens when things get complex?

The approach of encapsulating things to present bundles of code as a black box is a good one, but it needs to be paired with a powerful navigation interface that allows flying down into those black boxes pretty instantly, and figuring out how you can be viewing the internals of those alongside the 'outer' code at the same time to understand how things are interacting when things go wrong.

> If it is being proposed as a new way of doing general software development

Which it isn’t. It’s being proposed as a tool for rapid application development.

Yes, we are building the tool using our tool! I am glad that at least one person got the importance of that!
You can't build a BPM using a BPM, however there is a time and place for a BPM.
There once was a tool that simplifies software production built into nearly all 8-bit machines. It was called BASIC.

Millions of kids learned how (and how not) to code using it. Many of them gained a career as a result.

Unfortunately kids in the future had to do without it. Because there were a -thousand- 'better ideas'.

You are absolutely right! I used to be one of those kids that learned machine code and BASIC while still in primary school. But technology evolves and expectations grow. That is why we need better tools all the time.
Hypercard was a better tool. And then they took -that- away.

The fundamentals of coding don't change. Built-in BASIC was -so- easy to use. Which 'better tools' can say that?