Why reinvent the wheel?

Just like everything we do, we start off with a problem that our kids are facing. Writing is a major part of learning Chinese, especially when it comes to word shape and the stroke order.

However, all Chinese characters are being written in a box that looks like the word 田, hence the name 田字格. A more advanced version of it also has diagonal lines in it.

But using this word 早 as example, the structure of the grid does not lend itself to frame the word. Without any guidance, it is difficult for a new learner to find the starting and ending points of each stroke.

What do we want to achieve?

We want to develop a grid system that closely observes stroke points for common Chinese characters that can help to guide learners.

The result is as shown here. The grid system is built up using non-concentric circles and guidance lines. It may feel and look lopsided, but it better matches the writing style of Chinese characters, which is very perfectly illustrated by the same character 早.

How did we get the grid?

Through cold hard data science, of course.

First we analyze all the stroke points (starting, turning, and ending points), and plot them into a heat map.

With this heat map, we identified the clustered regions and captured them in circles.

Finally, we find the shapes and positions that can capture these main regions. It just happened to be nicely captured by 3 circles, a straight vertical line, and a slanted horizontal line. Additionally, we added a few extra lines to provide further support for the characters.