The current text jumping methods of legend seem to be mainly divided into two types.
-
The first type of jumping is very pleasant. Its main triggering methods are four (alt j); (ctrl f); (click the text to place the cursor and then ctrl alt ); click the pop-up window on the right side of the backlink to jump;
-
Their feature is very fast. My documents are already huge, when I used workflowy, its size had to load 10s every time I opened workflowy. Now it’s even bigger and more than 5 times larger than the content in wfy days. And All the content is in one document (so use can take full advantage of the selection search function). Even so huge. The first type of jump can still achieve instant jump to where I want to see. When I used it for the first time, I felt I am amazed that web notes can have such performance. Its smoothness makes me use these functions very frequently and accustomed to.
-
The first type of jump seems to have some features. For example (alt j) enter after searching the content. By default, after jumping to the corresponding content at the home level, it will place the cursor at the end of the jumped text. This feature is very convenient. It can Let me jump to the content and edit it directly without having to repeatedly click and place the cursor with the mouse.
-
Speaking of placing the cursor, this reminds me of another feature of legend. When you move the cursor, the field of view will immediately focus on the cursor. This is also a very useful feature. For example, when you click on a node in the sidebar, it will zoom in by default. This node. When zooming in on a node, you can only view the content of this node and its child nodes, and you cannot view the content at the same level as it. You can only view nodes at the same level after zooming out (alt <-). But there is a problem ( alt <-) it will focus the field of view to the top layer of the previous node by default. You want to jump to the position where you just zoomed in and read. You just need to move your cursor with the arrow keys and the field of view will immediately focus to the position where the cursor is . Even if the top layer of the home page is 3w nodes away from the cursor I placed, the field of vision is still instantaneous and the jump is extremely smooth, just like the first type of jump.
-
The first type of jump always places the cursor, which reminds me of the cursor focus feature of legend. I am wondering that the code implementation of the first type of jump is to first place the cursor on the target text to be jumped, and then use the lower-level feature of the focused text. Does it realize the jump? No matter how the first type of jump is realized, its extremely excellent code implementation gives me an extraordinary experience
-
There is also a second type of jump in legend. I only found two places to use this jump method. One is the ^v button in the sidebar. Clicking this button legend will scroll from the top of the home page to the corresponding Location. If the text I’m looking at is long and multi-chapter. Or I’m looking at a multi-chapter novel, this feature is logically the most suitable because the sidebar is easy to locate what you’re looking at. Also click ^v it jumps to the home level by default it doesn’t zoom in on the node so you can still view the chapters at the same level as it. But I almost never use this feature. In fact, it’s really not very useful.
-
I clicked on the home page to jump to a far-flung novel content. It may be the reason why my document is too large. It took me 13s to click on it to jump to the corresponding position. The unpleasant memory of opening workflowy and waiting for 10s. The more uncomfortable thing is that it often stops scrolling when it is halfway through, and how much do I need to click it to smoothly scroll to the position I clicked. Sometimes my computer lags when opening a lot of things . This kind of jump needs to scroll while loading the page. I clicked it 5 times to complete the jump. This experience is really frustrating. Now I want to achieve this requirement. I always click directly on the zoom node and then (alt <-) Zoom out the node and then move the cursor to focus the cursor to achieve the function of jumping content in the sidebar. Although this operation has many steps. Repeated use will be more troublesome, but it only takes 1s-2s and I always click the ^v button. It takes about 10s. If I use this function, I’m forced to consider whether I need to waste 10s waiting for this thing before every click.
-
There is another kind of jump that also uses the second type of jump. That is (alt g) Speaking of this, I want to share my legend usage process and the pane layout of legend daily use. I usually open two panes in my daily use. In the past The pane on the left is used to browse the reading content. On the pane on the right, I will zoom in on a node dedicated to taking notes and use it to take notes. But now I have made a slight change. Now the pane on the right no longer needs the function of taking notes. It only You need to complete the search and find function. When using the pane on the left to read long texts, I am always reluctant to use the label next to the text to filter to find relevant content or directly (alt f) to find relevant content. Because every time (alt f) The view always focuses on the top of the home page. I need to scroll the scroll wheel or use the first type of jump to jump to the previous position. After many repetitions, you will feel so troublesome that the left I just want to read and browse the pane, and I don’t want to use the extremely efficient text filtering and searching function of legend. Later, I used the pane on the right to use the searching function. Compared with the past, I now use the filtering function very frequently during the reading process. to find the content. Now I use the (alt n) function to take notes. If it is a note that is very closely related to the reading content, I will directly record it on the child node of the original text. But there are many notes that are only related but not Completely important content I don’t want to record directly in the original text to pollute the original text. I will directly use (alt n) to record efficiently. I only need to record what I think of and then directly (alt n) record and enter. The notes will be placed directly in the home. The top position. Afterwards, I use legend’s various powerful functions to organize notes and then organize notes. I don’t need to think about where my notes should be placed when recording, and whether it will be confusing, just rest assured (alt n) record and then enter. And (alt n) recording will not affect the current state of the left and right panes at all. Very comfortable.
-
After using this method, I really want to use the (alt g) function. I may place many tags in a document. I am not sure whether each tag can search for valuable content. And each Searching all tags, it will be troublesome to search very frequently. The ideal solution is to place the tag (alt g) when the left pane is reading, and then the right pane will jump to the corresponding position instantly and then I just need to simply click the tag I can search for relevant content, which is very convenient. Even if an article has a lot of tags, I can rely on (alt g) and simple clicks to achieve efficient search without interrupting the original reading. But unfortunately the situation Not ideal. The jump used by (alt g) is not the first type of jump. It uses the second type of jump. Every time I use it, I have to repeatedly click ^v and scroll for 10s before scrolling to the corresponding position. A terrible experience. This experience is not as fast as I manually enter the tags in the search box manually.
-
Maybe the second type of jump uses relatively early code? The first type of jump is the code that was perfected later? From the experience, the first type of jump is really much better than the second type. Instant jump and 10s The gap between the left and right jumps. It is a more advanced design. Maybe the code for the first type of jumps can be reused on the only two remaining functions that use the second type of jumps, which should be used frequently. Great functionality, but the implementation of second-class jumps hinders their usefulness.