Messages are currently checked for special character strings like ==, >>, ::, --, and ** after each keystroke, so that effects like boldness, quoting, and card types can be applied in real time.
This is a good default for most usecases but can be very frustrating when you’re trying to write notes that include those characters without those special meanings – which is extremely common if you happen to be studying programming. I find myself constantly fighting RemNote to format my notes the way they’re written and not as cards, bolded strings, or other things. (This can be especially troublesome with certain card formattings that can’t be undone with a CTRL+Z, you have to erase and start writing it anew.)
Logseq avoids this problem by deferring the parsing until the note has been fully written. Compare RemNote on the left to Logseq on the right, writing the same input.
Not only does it make it easier to format your writing the way you want it, I suspect this is the reason Logseq has noticably lower CPU/energy usage during writing sessions.
I would be greatly appreciative of an option to either defer parsing the markup until the text cursor leaves the rem, or to show the Markdown representation of the rem whenever the text cursor is inside it and the rendered representation when it’s not. Either would solve the problem and make RemNote that much more comfortable to use IMO.
