Hi folks,
A couple weeks back I got a kind email from @zowiewho over at Shiny Frog that encouraged me to try out Lettera for a specific use case: the “company brain”. This is a growing pattern I wrote about recently where instead of using Google Docs or Notion, teams are using a folder of synced Markdown files alongside agents and triggers to maintain their company docs and knowledge.
Currently teams doing this either use Obsidian or code editors to work in a Markdown project folder – but neither app is ideal for this. I use Bear for my blog post drafts and sometimes wish Obsidian was more Bear-like, so was curious to try Lettera for this!
What follows is my ranked thoughts on the biggest obstacles to recommending Lettera for this use case. (That is, for working in a shared folder of Markdown notes and docs.) Take all this with a grain of salt because I’m not sure how this meshes with the underlying vision for Lettera – this is just specific to people and teams trying to use it as a tool for building and maintaining knowledge bases (either “company brain” or “personal knowledge management”.)
- New file workflow is cumbersome.
- A very common workflow is to quickly create a new file and jot something down, starting a small note. ⌘N almost does this by creating a file and saving automatically, which seems great.
- But when you create a new note, Lettera creates a file called Untitled, then dumps you in the note contents instead of the filename field. (see 2 below for file renaming).
- For this kind of workflow, you generally want the same UX as Bear and Obsidian: you ⌘N, then immediately type the name of the document. (Or don’t, but by default you can do this immediately.)
- In practice, this means people will end up with Untitled.md gremlins across their project, some of which have meaningful content and some of which don’t.
- This is annoying for a personal folder, but a huge problem for a shared one.
- I saw on the forum that you can ⌘↵ then ↵ to edit filenames. This fine in theory – I could learn it and do it every time – but almost nobody will discover this.
- This is the top issue for sure, and I think will block people who try to use Lettera for a brain or shared docs folder.
- Fixing filenames is cumbersome.
- It looks like you should be able to edit the name of a file by dropping the cursor into the filename in the top left, like in Obsidian. That would be great.
- Instead, you need to click, wait ~100ms so it’s not counted as a double click (which would just reveal the file icon), then click again, then it pops another field with the whole filename selected instead of where you clicked to insert.
- If you type at this point, it also removes the .md extension.
- This indirection between the thing you often want to edit and actually being able to edit it is friction.
- It would be nice if Lettera displayed the filename larger and more H1-looking like Obsidian, but that is bonus.
- It’s not clear how to jump to a note.
- An extremely common workflow is to jump to a note you know the name of.
- In Obsidian this is what ⌘O does to avoid conflict with ⌘K for Edit Link, but it would be more consistent with other apps to make ⌘K open a “jump bar” when no text is selected.
- That is, ⌘K, type “lette”, then just return since I see it’s filtered down to just my doc titled “lettera-feedback”.
- It’s not clear how to jump to another folder.
- Not everybody will do this, but I often have 2-4 different folders of text/markdown open: company brain, personal blog repo, personal brain, software repo that has a lot of .md docs in it, etc.
- It would be extremely useful if the ⌘O/⌘K jump bar described also also jump-switched to other windows in Lettera by the name of the folder.
- That is, not replacing the foreground window with another folder window, but foregrounding the window the destination folder is open in (or opening a new one if it’s not open currently).
- Together with the above, this would let you for example open Lettera, then type ⌘K “comp” ↵ “goals” ↵ and then you’re at the goals doc in your company brain in two seconds.
- It’s not clear how to close a document.
- Lettera seems to have no concept of a folder/project being open, but no file being open. In most apps for working in folders like this, you can ⌘W the open files until you have a clean slate. In Bear you have to hit esc to close the file you’re done with, which isn’t so bad since you don’t tend to have multiple Bear windows open and use them for file navigation.
- Not being able to close your docs in Lettera means you need to pick between keeping the doc you’re finished with open (which feels like there’s more to do on it but there isn’t) or closing the whole project and needing to re-open the project next time.
- Generally for these kinds of use cases, you would keep one or more projects open all day every day.
- It feels like ⌘W should first close the open note, and only close the folder if no other notes are open when you do ⌘W again.
- Acting like Bear where Esc does this might be learnable for muscle memory, but it’s a weirder workflow for working on files in a project than ⌘W, and you get people accidentally closing their whole project with ⌘W and not realizing Esc is an option.
- Right click → Copy Image seems to just copy the text “image.jpg”?
- It’s too hard to tell what folder is open.
- It shows at the top of the left sidebar, but grey and small and easy to scroll off. This is the anchor context for which window you’re in.
- Also once you open a file, it doesn’t show in the Window menu which window is which.
- When you close a folder, it loses the state of what doc was open, what folders were expanded and not, etc
- This is much worse given the current fact ⌘W seems like it will close the doc but closes the whole folder, so I get a lot of projects closing I didn’t intend to close, then losing my state.
- It’s not clear how tabs and multiple-files-open-per-project are meant to work.
- I found the feature but it didn’t behave how I’d expect. This isn’t so important to my own workflow but some people do this.
- It’s not clear how to show hidden folders.
- Folks who work with agents for example will often often need to edit .md skill files in
.agentsat the root of their folder. - Showing dot files/folders is also useful for codebases, which increasingly have markdown worth maintaining.
- Folks who work with agents for example will often often need to edit .md skill files in
- It feels like you should be able to drag-select a table row to select it and delete.
All this said, I think it’s great you all are working to bring Bear-like sensibilities to this category.
I used Lettera instead of Bear for writing my blog post this month, and it was awesome to be able to have an agent fix typos and otherwise operate and then eventually publish the post “in situ” rather than my old workflow of eventually copying from Bear to a real markdown file once I’m 90% done. I haven’t tried Lettera on iOS yet, but Bear-style smooth iCloud sync between phone and Mac would be nicer than Obsidian’s current partial implementation.
So hope this is more helpful than annoying, and best of luck with Lettera! 