Lou: This is episode 56 of Write While True. Write While True is an infinite loop, and that’s because we think of writing as an infinite game, a game we’re playing for fun and to get better at, like a game of catch.
So in each episode, we’ll tell you something we learn about writing, and then we’ll throw you the ball with a writing challenge or a prompt.
I’m Lou Franco.
Brian: And I’m Brian Hall. And today, I can’t wait, we’re talking about writing tools, talking about your writing stack.
And we’ll actually spread this out over two episodes.
Today we’re talking about the idea phase, outlining, collecting notes, and drafting. So just the writing part. We’ll save publishing, editing, collaboration tools for a future episode.
So let me just toss it right back. Lou, tell us about your stack. Tell us about how you got to your current stack.
Lou: For about five, six more years, I’ve been using Obsidian as a second brain solution. So Obsidian is more, I think people would think of as more of a note-taking app than a writing app. But you certainly can use it to write, because notes are just writing.
But I’m practicing what has been called like Zettelkasten style note-taking. So I have lots of small notes in a system that collects markdown files, simple ASCII text files into folders, and they can be cross-linked.
And I also do longer form writing right into Obsidian, mostly first drafts.
So that’s where almost all of my writing starts out as a tiny note, a collection of tiny notes, and then a draft made up of a synthesizing of tiny notes. So a lot of my writing starts that way, whatever it’s for.
And so that’s the first part of my stack, the first tool I touch when I think to write something down that’s new.
Brian: So you actually do the thing where you collect the notes, and then you synthesize the notes, and then they turn into larger publications or articles or blog posts.
Lou: Yes. As described in like, there are some several books that go over this technique. I think the famous one is “How to Take Smart Notes.” I’ll put a link in the show notes.
I also talked about it in an old episode years ago about how I do this. But the general idea is it kind of makes writing easier.
Writing when you’re trying to draft an actual thing you would like to publish is more like collecting and refining than trying to write something new. The new writing is tiny.
So like a thought about a book or a thought about a chapter in a book or a thought about a YouTube video or a summary of a conversation or anything like that. And then you take that and you try to connect it to your overall big ideas that you’re kind of developing.
So when I was thinking about technical debt, if I had a conversation with somebody that was even remotely connected to technical debt, I’d summarize the conversation and then work it into a paragraph relating it to my ideas about technical debt.
And I have other things like this, like I have an idea about AI safety. And then when I hear something that’s related, I try to make a paragraph about that. And then those paragraphs might end up in a piece together, connected by new writing, but that writing is usually easy to write. So that’s what, that’s how I use obsidian. And that takes me from nothing to almost a first draft, let’s say.
Brian: Let me jump in here because I, this all makes such total sense and I’ve read and I’m familiar with the methodology. I’ve used most of the tools and I just utterly failed at it multiple times. So I used to use Roam Research, which is a similar, similar Zettelkasten tool. I’ve used Obsidian extensively. I used to do Notion also works for this.
And what I found every single time was that I just collected and collected and collected. I would paste lots of things and write lots of things and dash off lots of notes. And I never had the discipline, the interest, I’m not sure, to collect them and synthesize them and turned them into something.
And I found that during this period, when I use these tools, I did still write and publish, but I didn’t really use my notes. It’s like capturing the notes was just part of thinking and organizing my thoughts, but I didn’t actually open them back up and make use of them. And so these days, what I use for this kind of purpose is I actually type cat and then the greater than symbol and then /dev/null and hit enter.
And I just type into the void. I just type stuff out and I hit control C and it disappears forever.
Lou: Oh, wow.
Brian: And I, and there’s no difference to my productivity.
Lou: I used to call that write only memory in a, in one of the jobs I was in.
Brian: Yeah. So I’ve tried them all and I, I just failed. So I’m not trying to discourage anyone else from using this kind of a system, just to copying to my failure.
Lou: I’m going to, if you’re thinking about what we both said, I’m going to offer this as something to try. If you think you might fail: a really, really important part of this is linking.
So if the notes are not linked together, then it, they will be really hard to use. So notes are constantly referring to other notes. And the other part that I would say is important is that each note is very single focused. So you’re not trying to develop an entire argument in a note.
You’re trying to develop one or two paragraphs of an argument, one point. And you’re linking to further things or to references.
And then what, what I do is I pick a random starting point and just kind of look through the notes, following links kind of aimlessly until it kind of gives me a little inspiration. Like, oh, I’m seeing a connection here. I’m, I want to think, oh, I thought of it this way. Now a new take on it. And then I can develop like a, a 250 to 500 word blog post from that.
So I’m kind of mining it for ideas for content.
Brian: That sounds really fun. I never did that. I should have given it a try.
Lou: Or whatever’s working for you is, is also valid.
Brian: Yeah, I guess so. I guess. I mean, it’s fine. I don’t miss it.
I just feel like I wasted a lot of time back in the day, collecting and taking notes that I didn’t use. I don’t, it’s forced me to assume acknowledging who I am, I guess, and I’m not somebody who is.
Lou: Can I ask for, for your book, “Your Website Sucks.” Well, how did you develop the drafts of chapters for that?
Brian: I had a single outline of topics or ideas, or the book is about problems with website user experience, basically. And so I just had a bullet list of here are some of the things that could be in a chapter. And then I would just sit down and crank out a chapter in a single session.
And then from there, that’s a manuscript. And then that’s what got revised and turned into the book.
So I think I tend to produce a bit more at a sitting and in between sittings, it all just bounces around in my head. And I’m sure I lose lots of things in this way.
Lou: Yeah. I mean, the important stuff probably stays though.
Brian: Yeah, hopefully. But so I tell myself, yeah, I did produce a lot of trash notes, honestly. So yes, I do think that works as a filter.
Lou: Yeah. No, I definitely have. One of the things that it also helped me do is like explore something that is ultimately not that interesting, but it’s out of my head now.
Brian: Yeah.
Lou: So I was like, I don’t have to, I’m not thinking about it anymore because everything I think about that subject is written down somewhere and not, you know, maybe I guess one day it might be interesting.
Most likely it’s not. Most likely it should have been in dev null as well, but it’s not.
Brian: I mean, it’s fine though. Text is cheap. It doesn’t hurt to have a markdown file that you don’t revisit, I guess, unless you judge yourself for that kind of thing, which I guess I did.
So you have this really nice, organized, deeply interconnected workflow around Obsidian and capturing ideas and notes and synthesizing them, following the connections.
And then you tend to produce a draft right there in Obsidian.
Lou: It depends where it’s going to go eventually.
So if I have to give it to somebody, like, let’s say I had to give someone something and that is going to be not controllable by me, what that’s going to be, I’ll draft it in Obsidian.
But for my blog, which I use WordPress, I draft in WordPress because it’s just easier to get the formatting correct.
Like if I, if I copy paste markdown, it’s okay, but not, it’s not high fidelity enough for me to trust it completely. And so I’ll have to, I’ll have to look at it. And so I tend to just draft it right in WordPress. And that’s where, and the links would be totally different. Like the link links in the WordPress or to other articles in WordPress. I can’t link that way and Obsidian. So it’s just a way easier to do it right in WordPress for me.
And, and so since WordPress has gotten so good at like saving partial, like auto saving, it doesn’t, it doesn’t feel dangerous to do that right in a web app. Normally that would be big. I think it would be a little scary, but the modern web is better at, um, preserving, like not losing that data. Like it’s not like literally a content area that you lose. Um, it’s, it’s, it’s got modern word processing features, so it’s fine. And so I do that blog is right in WordPress, but almost anything else is going to be in Obsidian.
And then I would also say that for the book, I felt because it was so long and it was really beyond what Obsidian could manage. Or I could manage in Obsidian. And I decided to use Scrivener because it’s still has the idea of disparate content pieces being collected together.
So a Scrivener for people who don’t know it is a tool. You could think of it like a word processor. You can definitely use it that way if you want to. But where you should more think of it as is a hierarchical structured document of snippets that get compiled together into a document. And it’s very easy in Scrivener using the tree structure in the sidebar to reorder snippets.
So if you write things at a multi-paragraph idea level into those snippets and then construct your chapters from, and you might think of it that way as headings and subheadings in a book, then it’s very easy to rearrange the book. And that’s not true for most word processing systems and even Obsidian.
I mean, it’s a different way of thinking about it. Scrivener is always thinking about it as its final manuscript. So it’s going to definitely be a something, a PDF, a doc or something, a book. So it’s more thinking about it as eventually a serialized document. So it’s just really good at keeping everything straight, keeping all the links straight.
You can say chapter and then a special code. And then if that chapter changes its position, that’ll change the number in the reference. It’s very good at understanding like what it would be like to write a book and rearrange that book.
So during my book’s drafting phase, when I was realizing that I had not really settled on a structure, I got it all into Scrivener and where I could like try different ideas, try different ways of thinking about the order of the chapters. And it kept everything straight for me and I really liked it.
It is inferior as a writing tool. And by that, I mean the, the, what you, I think you would just think like the normal editing features of a writing tool seem to not be as good as most. Like there’s like almost table stakes for that, like how the, how the cursor works and how copy paste works and how, um, simple, the really simple stuff works was not great in Scrivener. It seemed to have bugs in that area, which would, which was really surprising, but in the end I still decided it was worth it. Uh, cause they weren’t, it wasn’t an everyday occurrence, but every once in a while it would act in a surprising way to me just on normal editing. So I, I think what I, I wouldn’t use it again, but I was so deep in it that I couldn’t really give it up.
Um, but I think that made me, makes me a little less likely to use it in the future because the day-to-day editing was sometimes a little annoying.
Brian: Speaking of day-to-day editing being annoying. So let me unveil my stack and go. It is all, it is completely just a function of having gotten frustrated with every piece of software I’ve ever used.
So I mentioned Roam Research, Notion, Obsidian at some point, either I’ve tried to do something that I couldn’t do, or the app has crashed or shipped an update that broke something I depended on. I just have been burned by every piece of software I’ve used heavily in this way.
And so I now do everything in Vim. It’s Vim or nothing. It’s plain text markdown in Vim or nothing. And so I use one tool for capturing notes and outlines and things.
If I know I’m going to, if I need to save the information, things like links and facts, I’ll save with a tool called JRNL J R N L. It’s a command line script that basically just makes for quick note capture, writes them all to a single text file on your local machine.
And then it’s really good at searching them and their timestamps and tags and stuff. But it’s basically just, I can press one key and type a note and it’s saved in a text file.
And then for blog posts and even books now, it’s at least up to the point where I have to share it with somebody. It’s just a single markdown file. And that’s all I do.
And I really love Vim. Speaking of being able to move the cursor around and copy paste it, that matters so much. It matters so much how that feels, right? You don’t want to focus on getting that done when you’re trying to get things out of your brain. And so for my own joy and comfort, I’ve decided that if I can’t do it in Vim, I just won’t fool with it. And so this is not necessarily a wise decision. It leads to a lot of kind of bike shedding, over-engineering later steps of the process.
But I love it. And so I open it up every day and type. That’s my stack.
Lou: I mean, are you using plugins that like, for example, a spell checker? No, actually.
Brian: Probably should.
Lou: Okay.
Brian: But I don’t. You know, it’s either low enough stakes that I don’t care. I’ll find it when I find it or never.
If I’m actually going to publish a book, it’ll go through at some point. Proof-reading. And we’ll catch it then.
So yeah. Plugins-wise, gosh, nothing much that really helps with the writing. I really prefer a tool that gets out of my way, I guess.
Lou: Yeah. I mean, I use Obsidian with zero plugins. I know people plug in that thing up.
Like crazy. And it’s like, I want zero. I want my editing. And Obsidian does have a great editing experience. It’s just locally installed software. It’s a good editor.
I think it’s just, I’m assuming it’s an electron app using whatever, the Atom or the VS Code editor or something like that, which is a very good editor. And, you know, mostly what I’m looking for is, it’s as fast as my typing is as fast as how it goes.
It should be fine. I’m running like a modern computer. It shouldn’t be. This is a problem for some software. It shouldn’t be, but sometimes it is. But it doesn’t have that problem.
It has very reliable editing features. So I’m never worried about that. And the storage format is just markdown text files. So it’s very, there’s nothing tying me. It’s just folders and markdown text files. Nothing about Obsidian’s storage that I can’t take to any other tool.
Like, in fact, I sometimes do edit the files outside of Obsidian if I want to do something with a tool, a command line tool or something.
Brian: I’ll say Obsidian is a beautiful piece of software that performs really well. And I have no actual beef with Obsidian. There’s something wrong with my brain.
And if I open it up, I will immediately start over engineering the links and tagging. And I am that person who will look for a plugin that I don’t need. So sticking to Vim and a single flat file that is plain text in markdown format is to rein myself in. It’s not a judgment on the tools.
Lou: I get it. I definitely have that feeling sometimes with some things. It’s just the Obsidian has been, it’s just an, it’s okay.
It’s like, like, I don’t have to like pull back even more.
I have the same feeling of simplification, but I stopped at Obsidian because it was, it was, I was nervous about using anything else where I can’t, couldn’t, where the files were in some kind of system that I could not, you know, was proprietary or especially for something as important as, as what these notes are to me.
And, um, so I decided that I needed that Obsidian did it so that I didn’t have to go further, but I would have, if, if there wasn’t something like this.
I will tell you the biggest drawback is I have decided not to adopt any kind of syncing system, which allows my notes to be edited on anything other than my laptop.
Because those are not good with Obsidian, the choices are not good. And I haven’t been happy with them. And when I say not good, they’re not good for me. And so, uh, so I can only edit the files in, on my laptop.
They are all put into GitHub, into a private repository, so I can read them from anywhere, but I can only really, I mean, I guess, I’m sorry, I guess I could edit them with a regular editor on something else.
But I can’t run Obsidian on something somewhere else and, uh, and, and see those files. But I obviously they’re, they’re in GitHub. So I could, but I tend not to want to be doing it’s, I think of it as push only from my laptop and read only everywhere else.
And it’s been fine. I thought I was going to hate that. But, uh, in, in practice, this is my, I haven’t wanted to edit files on a phone or, or, or an iPad. I just like, I usually have my laptop with me if I’m going to be somewhere like on work travel or anything. So it’s not gonna be a problem to, to do it there.
And if I do have some crazy idea out in the world, I’ll just make a note in note in the notes app on my phone and that’s fine. I’ll just incorporate it later. It’s not, it’s not a big deal. I don’t need it to be.
Brian: I’ve made the same compromise. I don’t have them running on my phone. Obviously I don’t care.
I don’t want to write on my phone, but I do use a pencil and a pencil sharpener and a notebook to mainly when I want to do something that is unstructured in my head, when it feels really messy and I want to draw pictures and boxes and arrows, I’ll use that to get things straight.
Lou: To draw a mind map or something.
Brian: Exactly. Pretty soon it will become bullet lists that can go into a linear outline that can become prose, but yeah, I definitely use paper and pencil for stuff as well.
And that’s, that’s the stack.
Lou: I think of the, the, the capture part, people often call this capture. Uh, I don’t care how I capture as long as it’s into, I don’t need that to be right into my ultimate system. As long as I trust that whatever has been captured will end up in there.
And as long as it’s like in my main paper journal, which I review, I know I will review. And if, or if it’s in the notes app, which I know I will review, it’ll be fine. And then with those two, one of those is going to be on me.
I guess I can’t even imagine like where, if I’m somewhere where I don’t have my phone or a journal, like I should not be writing or capturing it.
Like, I hope I’m not thinking about that stuff.
Brian: It’s a really important point. I guess I’ll be explicit.
I don’t actually believe in the idea, in the notion that some genius idea will come to you and you have to capture it or you’ll lose it forever. I trust that the genius idea will return to me if it’s truly a genius idea. And in that way, I let myself off the hook for having to capture everything, remember everything, process everything.
It’s. It’s enough. If it’s important, it’ll come back to me. And if it goes away, someone else can have it.
Lou: Exactly. Or there’ll be another one. Maybe not that one, but another one like that.
Brian: Yeah.
Lou: Ideas are not like you don’t run out of them. They’re like a renewable resource. And so like, it’s fine.
So given all of that, our web of tools, I guess I have a web of tools.
And, but what it’s if someone’s thinking about like, you know, as developers, like, you know, you called it our stack and like, you know, or maybe always like fiddling with our stack or thinking about it more than, or out of proportion.
Brian: Yeah. Because it’s fun.
Lou: What do you recommend to people if they’re just, well, they’re maybe putting something together or considering things?
Brian: Yeah, I would suggest, and you might not need to hear this, I guess, but I would suggest auditing your stack.
What do you use to capture ideas, flesh out ideas, take notes and write drafts.
And I would look at it with an eye to, is there a way to simplify it? Hmm.
Or is there a comfort tool that you might as well use? I guess I I’ve moved over the years from more web based apps to local based to literally a command line text editor.
It, if you can make a move in that direction, if you really are comfortable in, in VS code, maybe you try writing markdown in VS code and see how it feels like seek a bit of joy and comfort in your tools.
And if you already have it, that’s great.
But if not, sit with it and try something out.
Lou: Okay. So that was episode 56 of Write While True. It’s a podcast where we love infinite loops as long as they’re fun.