prev Kozaneba Development Diary 2021-10-01
Used to explore other people’s writing, useful
-
Arrows are very useful
- No longer limited to “relationships expressed through proximity.”
-
I want regular or double lines, not just arrows.
- It’s neither transitional nor confrontational.
- A case in which there was originally a kozane called A, and then a group was formed to explain the contents of A in detail.
- The most intuitive way to connect them is with equals.
- The functionality provided now would give the group the title A.
- The statement “this and that are the same” is forcing a proximity representation at stage 3.
- Close it because it’s too bulky when left open.
- But if you close it, you have to open it to see it.
- If they’re not in close proximity, but connected by a line, they can be placed apart.
-
When an arrow is drawn against a group, the ends are wrong.
- In some cases, they’re far apart, in other cases, they’re wedged in.
- Bounding box for groups?
- If you write an arrow on a child element and then close it?
-
The additional arrow UI is still a little difficult to work with at times.
- Do we need an arrow mode only for drawing one arrow?
-
Ability to inherit the maximum size when the group is closed
- As I mentioned before, still necessary.
-
It could be made smaller than the normal size.
- When I was using it for personal use, I deleted anything that wasn’t important when I created Kozane in the first place.
- In correcting other people’s writing, I left out particles, conjunctions, etc.
- Then I had to enlarge everything else because those are 1x size.
-
sentence in both ways.
- Twitter
- I noticed something interesting when I used Kozaneba to correct other people’s writing.
- When I personally make a kozane, I make it as shown on the left, even if the input is the same sentence. However, when I was correcting other people’s sentences, I unconsciously made them the way I did on the right.
- Twitter
- From the right state in the previous figure, we can expand important function words without changing particles, etc., and pick up only the expanded kozane as a result, resulting in the left state in the previous figure.
- When I work alone, the left shape is created from the beginning, and then the spatial arrangement is started from there. This may seem like skipping a line to the rest of the world.
- I had been holding off on a prototype algorithm that would produce a kozane like the one on the left when a sentence is entered, because it was not completed to my satisfaction, but in fact, it may be more satisfactory for users to simply chop it into words and produce a kozane like the one on the right.
- Unlike when I created that, now I can group even multiple words into one lump.
Funny when closing a group containing arrows
- Binomial would be OK, NaN when there are more than that.
- Fixed.
@firebase/firestore: Firestore (8.8.0): You are overriding the original host. If you did not intend to override your settings, use {merge: true}.
- Fixed.
- Fixed.
-
Ability to inherit the maximum size when the group is closed
-
Ordinary line
-
double line
-
- Hmmm… it’s going to happen.
-
- Shorten it a bit, but not at all.
-
- If you work hard at math, you can get good at intersections, but don’t work hard and try to fool them.
-
- hmm
- hmm
I worked hard on my math.
- Conclusion:
This page is auto-translated from /nishio/Kozaneba開発日記2021-10-04 using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I’m very happy to spread my thought to non-Japanese readers.