When placing the insertion point after the three dots of a folded header, as in the image, and hitting return, the header now unfolds, and places the insertion point on a new line below its (‘child’) contents. I would expect, instead, a new line to open below the folded header, leaving the header folded. Why surprise the user?
When a folded header is copied, its ‘child’ content is copied with it; so far so good. But when it is then pasted somewhere, it appears unfolded. There seems to be no good reason for this; again, the user, expecting things she doesn’t intentionally change will stay the same, may be surprised by the present behaviour.
I would really like having a “Fold All Sections Except This One” command, “this one” being the one that the insertion point is in. (I used to have this in Vim. Bear ate my Vim!)
When folding a section, everything down to the next section title is folded. If you add a blank line at the end, it technically belongs to the folded part, and it would be odd not to expand the folded part if you are, in fact, editing it.
It depends on where you paste it. If the end result of pasting changes the range of the folded part, then you are, in a way, editing it, and we expand it for you. If we didn’t, you would have the odd effect that the markdown below what you pasted would just disappear (becoming part of the hidden fold), and that would be confusing.
This sounds like a good idea. We will think about it.
I think I understand [2]. Thanks for [3]. But your reasoning in [1] seems incorrect. I placed the insertion point after the three dots, which clearly represent the folded part. So it seems to me that hitting return there should add a new line after the folded part. And then there is no justification for unfolding it.
I did notice that it is also possible to place the insertion point between the header text and the three dots symbol. In that case, hitting return should expand the folded part, and add a new line to it. But placing the cursor after the dots should result, logically, in a new line being added after a still-folded section.
The folded range is not a static property. If a section is folded, then everything down to the next section title is hidden regardless of when the edit was done. Adding a blank line after the dots is before the next section title and will belong to the folded range after the edit. If we didn’t expand, the blank line would disappear into the fold along with your cursor.
Folding is activated by a Header. Is that correct?
If so, then if you didn’t want to fold something, couldn’t you create a new header, or simply create a new area or paragraph outside of the folded text under the current header?
i.e.:
Header Folded text (readable)
(Close header)
Header 1
[text hidden]
New text
Header 2
[text hidden]
Or are you asking to have text inside the folded or hidden header ‘not’ fold and remain visible after the header has been folded?
If that’s the case, then wouldn’t creating a separate area outside of the folded text work?
Please excuse me if I am completely misunderstanding.
That’s precisely what I want. But when I position the cursor outside the folded part (represented by the ellipsis symbol), and hit return, instead of opening up a new paragraph outside the folded text, it unfolds the folded text and then creates a new paragraph under it. Unlike @rexikan, I don’t see any reason for the folded part to unfold. A new paragraph could simply be created below the folded text.
Frankly, I think they didn’t think this through. If I can choose to place my cursor before or after the ellipsis symbol, hitting return when I do either of those things should produce different results. For a UX/GUI champion like Bear, this is an ugly flaw.
Your assumption seems to be that the dots represent a static folded part of the document, and in that case, it makes sense to be able to add things after it. But that is not how Bear works. Folding a section inserts a hidden marker after the heading in the markdown, and everything after it until the next heading is folded away, even if added to it.
When you press return after the dots, you will be adding to the folded section, and rather than have it hidden, we unfold. This is also how Obsidian does it.
We are always looking to improve Bear, and I take notes from discussions like this. Perhaps we can find a way to make it more intuitive.
The ellipsis symbol appears as soon as a part of the document is folded. It clearly refers to the folded part. In its folded state, obviously the part is static, because it would have to unfold first to undergo any changes. So my assumption seems… warranted.
Say a rabbit is inside a suitcase. The suitcase carries a label saying “Rabbit inside!” The rabbit is elided, and alluded to. We now place a second rabbit next to the suitcase. This second rabbit will not magically merge with the first. The suitcase will remain unchanged, just like (pace Schrödinger) the rabbit inside it.
This was just for sports, of course. Thanks for taking my suggestion into consideration, and I hope you will see the light.
In my opinion, reasoning from the perspective of what technically happens in the code makes no sense, the average user doesn’t usually know about that. The reasoning should be based on reasonable interpretations of users.
Based on that, I agree with the expectation @HvdG had.
If the technical side makes the expected behaviour impossible, that’s another matter, of course.
I understand your interpretation, and while it might be reasonable, it is also the wrong one. Folding is not a hypothetical feature, and I am trying to explain how it works so you can align your expectations as the implementation is unlikely to change at this stage. Also, different users will have different expectations, and the current implementation at least aligns with what other apps do.
That said, I do value the feedback and we are consistently considering how to improve the user interface both for usability and to set the right expectations. We put a lot of thought into understanding how different users interact with the app in order to improve it.
Header
Header 1
