pMovidea

branding

The most core function is "to move and organize fragments of thoughts", so I brought "moveable".

Short form movidea

Why reimplement?

Trial and error results in code complexity, which is an obstacle to adding new features.

Focus on Impact

I can't get close to "something that looks like it belongs somewhere."

Don't add features unnecessarily

I won't let it bug you.

Increasing the amount of code gets in the way of prototyping the next feature.

Minimal functionality with minimal code

Not to do

No Canvas

Free handwritten additions are not implemented.

Not implementing lasso selection.

why?

When we created Regroup, the first thing we tested was whether it was possible to smoothly add handwriting, assuming an iPad + Apple Pencil.

We found out to what extent it was technically possible (quite possible), but not so much in terms of user needs.

The realization of this is generating strong technical constraints, so it is better to seal it once and find another way to do it.

To do

testing a microphone

Using [styled-components

Last time it was Canvas, so it was defined in JS as far as style was concerned.

This time we'll use the DOM, but it's not nice to separate it into CSS.

Last time I made it while learning React, I started making it using the built-in useState system.

But since it's "an app that puts stickies on a giant whiteboard", the parent App component will have most of the state

What's Next

Cover the verification of the mistake with a test case before implementing the selection move.

Verification.

Can you make stickies without Canvas?

Important technical points

Dragging, well, you can do that.

I'm concerned about interference with text selection.

Include a version string in the data.

1: Make the exported JSON from the current Regroup readable

The question of whether the data structure should remain the same as before is not considered now.

2: Use it to create test cases

3: Data structure changes are easier to make if covered by testing.

→First, [Expose ReactN and use it from Cypress.

Can you zoom in and out smoothly?

I want to cover the data storage format with thorough testing, since it is hard to notice bugs or lost data when they occur.

You can't protect them with a mold.

Last time I made it without testing and started testing with jest-electron in the middle of the process.

It was difficult to make it testable because it was not designed with testing in mind and it was full of asynchronous processing.

It ended up not being covered in the test.

Last time it was iPad + Apple Pencil assumption, this time it is MacBook assumption.

Last time, the world coordinate system of the sticky was tightly coupled with that of the device, but I think it's better to separate them.

Group Structure

Pointer to child element

Drawing order?

doorplate

This is not a pointer.

Not even a child element.

Can it be a child element?

You shouldn't.

Mere text?

You can renew as many times as you like.

In the menu like edit title

So you should know that this is not your average sticky note.

What to do when it opens

It used to be a nameplate sticky, but not this one.

Wouldn't it be nice if it showed up as a "group title?"

https://gyazo.com/b181a51e078ac34a54860bed29bc2bf7

What should happen when you dismantle 🤔 this

If you don't have a title, just demolish it as is.

When there is a title, it is an intellectual product and should not be lost

Do you want to make it a sticky note?

🤔When you group it back together again?

How about just selecting stickies and copying and pasting them without making an elaborate UI?

[Folded things get one step bigger.

Last time I closed it with a menu from open, and opened it with one click from closed.

Let's try to toggle between "default open" and "default close."

Distinguish by borders?

consideration

done

context menu

Is this appropriate?

The menu wording is written horizontally, so the buttons are too long and narrow to be placed side by side

It's better to have a vertically aligned menu like the old context menu.

---