Cal Newport’s Digital Minimalism has left a lasting impression with me. A lot of his ideas around technology were incredibly useful and helped me come down from some kind of lockdown-inspired extremism into something a little more grounded, and a little less bingey. Whenever I’m in the grips of meta/tool-sickness, once I figure out that’s what’s going on I’ve probably forgotten something useful from that book.

He has a podcast, but I don’t listen to it much. A recent epidsode, however, had some stuff about note-taking and I have been deep in the grips of fussing around with that so I used it for my dishwashing and coffee making soundtrack this morning.

His key take is “get rid of friction,” which … yes. Back in the heyday of 43 Folders, most of my impatience came less from the content itself and more the constant riffing on “methodologies” that sounded more and more abstract, overthought, and overwrought. I just stopped believing any of it. Because there are only a few occurrences of the word “yarn” in my 20-year-old blog archive, I was able to find an entry in the non-public archive:

Date: November 14, 2005 at 10:05:06 PM PST

I’ve never hit a gtd adherent. I need to be up front about that.

I thought about picking fights with a few, I guess, but it’d involve barging into the comments over at 43 Folders like Bruce Lee in “Fist of Fury” and fighting with people who want little more than to be more efficient and get more work done. They don’t deserve to be antagonized for that.

Sometimes I read a comment from someone who insists that his routine involves some insanely arcane and convoluted use of yarn and a special shell script he whipped up that reads crap down from his Backpack account and then squirts it into his Palm, makes a redundant backup on the server he maintains in Malaysia and produces printed 3x5 copies in triplicate, one of which he pins to his infant son’s sleeve before leaving for the morning (“If I died, I couldn’t live with him thinking his father went out the door without an action list and a plan!”).

Sometimes, I was saying, I read something like that and I want to find that person and give him a noogie or burn two of his four backup copies. One, because I imagine that the “system” being described is a giant lie concocted by someone caught up in the thrill of inventing systems instead of actually, you know … using them to get stuff done. Two, because if these people are making these systems work for them then they’re surely VERY POWERFUL BEINGS we should hate and fear because we’re all going to end up working for them.

Newport’s take in 2024 is a little more kind, but comes down to “if you like building systems, build ’em, but, like, acknowledge that you’re indulging a hobby.”

He shared another idea I’ve come to appreciate in slightly different form: While you want to remove friction from the note-taking process, it’s not a great idea to hyper-atomize your notes and truly empty your brain of ideas in the hopes that The System will glue them all back together.

There is such a thing as too little friction. Whenever I’m playing around with a todo thing now I seldom enable “quick capture” or “get this into the system by forwarding an email into it” unless my overall operating state is pretty mindful and deliberate, because I know what it means to capture something without considering it much. At best, congratulations, you’ve just added a puppy to the box without a plan for feeding it or taking it to the vet for shots. At worst, it slips into the bowels of The System and becomes an ongoing source of guilt until you burn the system down and start a new one. The remedies for those possibilities just add more friction at point of capture (so great, you managed to launch capture with a single keystroke, but you still have a metadata chore), or require a disciplined maintenance approach.

That is todos, which are not notes, but the challenges seem similar. I’d also have to fiddle around with org-roam and a few other systems a little more to weigh how much discovery they offer at point of capture. He was wise to keep his criticisms vague, because differing feature sets + extensibility makes generalizing fraught.

I will say that mastering org-capture was a mistake for me, personally, because it became too easy to create a proliferation of atomized, siloed entry points into the system. Friction is a sweet spot thing, and I still struggle to find that sweet spot.

Dune

I rewatched Dune last night to feel prepped for the second part. Initial reviews for the new release have seemed positive, saying that it reaps the rewards of the world-building and groundwork done in the first installment.

I didn’t like part 1 very much. It was fine, but the break-point didn’t work for me and there was just enough deviation from the source material right around that part of the story that I got distracted by it.

That was a bummer, because I’d built the coviplex partially in anticipation of Dune, but between streaming issues that made the picture quality poor and not-unseeable differences of opinion, it was a little bit of a letdown.

Last night it worked much better for me. I was able to shut off the part of my brain that was busy reconciling source and adaptation, and the picture quality was way better thanks to a solid stream, so I caught more. I’d still prefer some slightly different choices here and there, but this is an adaptation of a book I read yearly from age 13 to some time in my 30s. And, tomorrow this time I will be parked in the theater finishing the story. Not watching the (occasionally glitchy, low-res) credits roll and thinking “nobody’s even sure he’s going to get to make part 2.”

Running shoes day

Al wants to start running. I told myself I’d pick it up again when I got my weight down. Well, it’s down and I’ve got a potential running partner. So we’re going to find running shoes today. I do well with Brooks Addictions, but they have changed a few times over the years. I’ve really liked my Hoka Speedgoats for fast walks and hikes on less technical terrain. Curious to see what the shoe people recommend. Anyhow, looking forward to trying to pick that back up again. Endurance running is the physical thing I seem to be built to do competently without a ton of focus, and it’s time to shake off winter.

I fiddled around with Workflowy a while back, and it has been in the back of my head, since. There is something about the whole outliner thing that is compelling, but when I see examples from people who are heavy outliner users in the wild I get this sense there’s a sweet spot between “useful chunking of information” and “stilted and hard to consume/digest,” and … we all have different cognitive styles, I guess is all I’ll say.

As near as I’ve been able to piece together some ideas out there in note-taking land, I’d like my personal notes setup to involve:

• A daily log
• A daily personal task list
• The ability to digress from the log

I’d like my log to be somewhat structured, meaning timestamped entries of a consistent format.

I don’t care if my personal task list for the day has much metadata, because we’re closer to the “shopping list” end of the spectrum than the “project management” list. I guess it can have no metadata besides “have I done it yet.”

Digressing from the log doesn’t mean much besides, “is it easy to think of something, quickly make a new node/page/tiddler and start typing.”

The apparent term of art for all this stuff is “intersitial logging.” That is fewer syllables than “keep track of what you’ve done during the day,” but has the benefit of higher syllabic density.

An outliner is a pretty good tool for those things because it brings some structure, favors the terse, and frees you from worrying about managing the arrangement of the text in favor of managing the arrangement of the content.

Outlining with Streams

Digging around for outliners for Tiddlywiki I came across Streams, which sticks a little outliner widget in each Tiddler. Click the “+” button, and you’re in a node in your outliner. Tab to indent, shift-tab to outdent, grab the nodes by their handles to reorder. Each node, in turn, is its own Tiddler.

I am not entirely sure how I feel about that last part, and if it were not for a filter you can apply to your sidebar to hide all the subtiddlers Streams produces in the open and recent lists, I am pretty sure I would hate it. But you can drop some code into the sidebar shaddow Tiddler to clean all that up and only see the root of a stream.

There’s enough there to play around, anyhow.

I made a simple Espanso trigger to timestamp my log entries:

# Make a timestamp
  - trigger: ":log"
    replace: "''[{{mydate}}] '' "
    vars:
      - name: mydate
        type: date
        params:
          format: "%r"

So, open a new node in the outline, type :log, start typing.

Finding the escape hatch with Streams Fusion

Because I’m a nervous soul who is always wondering how to back out of things like this, there’s also the Streams Fusion plugin, which gives you a little icon at the bottom of a stream to merge all the sub-nodes in a stream into a single, unified chunk of text.

That’s cool, because in a daily log you can be creating a proliferation of digressions and sidebars, not all of which qualify as full-fledged chunks of thought or interesting directions. So, click that button at the end of the day and all the nodes in the outline are turned into normal markup (links intact, if you added any) that looks like an outline, but all the child tiddlers from the root are removed and nodes you linked in a digression become backlinks to the newly merged daily log.

Alternately, you can turn the outline nodes into simple paragraphs. This morning I came across something while I was reading, started outlining, said all I’d had to say about that stream of thought, and went ahead and merged it all into a normal tiddler.

In evolving practice I don’t know whether I’ll fuse many daily pages or not. The part of me that doesn’t like the underlying sprawl of nodes even if I’ve hidden it from myself is still paying too much attention to what’s going on underneath. It’s a fine line between due diligence and unhealthy perfectionism. The benefit of fusing logs will be improved searchability for log entries themselves, since search results will go back to a single day’s log page instead of its child nodes. That seems to be kinder to future me.

Tasks in my log with Espanso and regexps

Once I decided to let the Streams experiment run, I waffled around about how to integrate tasks into my log. The “interstitial logging” people encourage a single, unfified stream of log entries and tasks created as they come up. Tiddlywiki’s core conception of tasks is that they should be nodes (“tiddlers,” yes) with a todo tag. That’s a little cumbersome in the Streams workflow.

So I did a quick experiment with the Sticky Todo plugin as a way to make a Streams node a task.

Sticky Todo uses markup like this to turn any text into a todo that appears in your sidebar:

<<sticky "Take out the recycling">>

That’s not too hard to remember, but it’s sort of type-y, so I made an Espanso shortcut that leverages its ability to do regexps:

- regex: ":todo\\((?P<todo>.*)\\)"
    replace: "<<sticky \"{{todo}}\">>"

So I can start a node with :todo(some task) and Espanso replaces it with <<sticky 'some task'>>

It’s 7 keystrokes minus the task content instead of 13? Plug it into the table and bask in the efficiency!

Espanso is useful for stuff like this where there’s no way I could figure out how to get Tiddlywiki to automate this in any reasonable timeframe.

I know that is easier said than done.

“Bits of our identities” are conversation starters, signifiers, hints, badges, clues to deeper things about us, personal reminders, and anchors.

“Anchors.”

“That was a very anchoring conversation.”

“This is a real anchor around my neck.”

Which is a weird way to start a post about note-taking apps, but here we are.

Like, Emacs is part of my “Unix person” origin story. I can’t name another software tool I’ve used as consistently for 33 years. I suppose the Unix paradigm itself edges Emacs out for personal longevity, but not by much. And when I think about everything I was doing with that first Ultrix account in 1991, “running Emacs” is the only thing that remains from the list. No more Netrek, I don’t use USENET in a way that would be recognizable to Past Me, and if the work I did on the Landsraad assembly hall for the DUNE MUD remains – getting cones of silence to work felt like a real triumph – I haven’t been around to visit it for a few decades.

It is a bit of an anchor in the putatively good sense of the word “anchor.” Technological comfort food. One of the first things that goes on any new machine, and one of the first server-side things I test when I’m trying out a new remote access tool. But also a bit of an anchor in the not-great sense of the word, in that I will pay the “figure out how to express this in elisp” tax for hours, well beyond practicality or reason.

Most recently, I was ignoring some excellent advice from Prot regarding people who want to use his excellent Denote for task tracking:

“If you want my opinion though, be more forceful with the separation of concerns. Decouple your knowledge base from your ephemeral to-do list: Denote (and others) can be used for the former, while you let standard Org work splendidly for the latter—that is what I do, anyway. … “Do not mix your knowledge base with your to-do items.”

… and the complexity was piling up and up, the tradeoffs were getting worse, and there was simply no joy in the experience because I had gotten myself into that bitter “make this problem yield” mindset that eventually leads to less understanding and more hacked-up, suboptimal stuff.

I just don’t have time for it right now.

So it makes sense that my fallback position was “maybe Logseq would be fine,” because it’s got the whole “supports org-mode syntax” thing going on – leave the door open for a return to Emacsland once I have more time – and perhaps because it is just odd enough to tickle another bit of my self-image. Mercifully, the second I tried to solve a problem of moderate complexity I realized how much time I’d have to invest to do anything besides pick code samples up off the sidewalk and pop them in my mouth. So I backed away slowly.

And after that it makes sense that Tiddlywiki got a look because it has been around forever and there is a sort of cheerfully prosaic attitude among its users. But the plugins started creeping in and I was trying to get it to do stuff it doesn’t really want to do without a lot of third party assistance, so the whole “it’s very simple” thing was not allowed to find much expression.

So there I was, and I’m sort of glad that I chased my tail over the past few weeks because it tired me out a little, but left me with an idea of what I wanted to do: Take notes in a connected manner, blend a little of PARA with space for a slipbox approach, and have inline todos, and I wanted it to sync across a few devices.

Obsidian does all that very well. There is always the risk of plugin creep, but in past Obsidian experiments that has been less about extending the core feature set and more about removing repetitive work. The simple mission of “write notes, link between them, keep track of tasks” doesn’t take much, with mobile and sync managed competently. It runs on every platform I’d care to run it on.

It’s a little dull. But after a few days of “just using it” and adding little affordances here and there from past vaults as I’ve remembered them, it has the benefit of just working in a non-dramatic, non-head-desking, simple way. I haven’t had to really think about it much. I haven’t inadvertently broken it or misconfigured it in such a way that I’m scrambling around a minute before a meeting trying to get back into my own notes. It’s of sufficient maturity that you can look up the answers to things and they are often in written form, which minimizes the whole “if I see one more YouTube poster frame of a slack-jawed influencer taking 30 minutes to explain something I could have copied and pasted in ten seconds I’m going to do a murder” thing.

I was inclined to say “and it says nothing about me, at all.”

But it does say a few things: “Doesn’t want to think about this problem he created for himself in any more detail,” “will settle on Markdown even though it is inferior to org,” “can stand being associated with people who think a graph of their notes is interesting and useful if it means not having to think about this any longer,” “will pay for sync,” “considers seven plugins normal and reasonable, would not admit to nine,” and – most likely and eventually – “always seems to creep back to Emacs even though it seemed like he knew better last time he did this.”

Poking around with note-taking options, I finally came back around to Logseq, which I’ll reduce to “org-mode and Workflowy gang up on Obsidian.”

I was primed to not like it, but the org-mode syntax option and some relatively strong opinions about how things should be were compelling enough that I played around with it, then tried to implement a PARA-like structure, then watched a few videos to try to internalize how it wants you to behave.

The page link syntax is interesting. You can do either a traditional wikilinks double-brackets, or you can use #hashtag notation to link to a page, which – thanks to the inbound links section of each page – serves as a tag index.

Logseq solves the blank page problem by … well, it sort of doesn’t. It’s opinionated enough about its “everything is an outline, every line is a node” thing, as well as its “start from a journal page each day” thing, that you can tell it wants you to do something besides “make a page and start typing,” but you can’t be sure what.

I ended up deciding to just go with it. Sitting in the living room, staring at the blank daily journal page, I just made a node in the outline for the first thing that came to mind and dumped out my big projects and OKRs for the coming quarter. Then I dumped a packing list for a trip this week. Then some miscellaneous tasks to get done before I fly out.

Entering tasks, then thinking ahead to how I’d find them once I moved on to a new journal page the next day, led me to the Unfinished Business plugin, which will roll yesterday’s todos into today’s journal page. I think there are other ways to gather todos but my initial read of journal page todos vs. project or area todos is that they should be more ephemeral and bound to the day.

But I guess my summary, having messed around with this for all of 18 hours, and after a morning of using it for work, is that Logseq is very, very into the non-hierarchical, “networked notes” approach. It wants structure to be a matter of emergence.

Its main competitor for my attention is Denote, which is more of a collection of Legos with opinions that are less about your workflow and more about the metadata. I like the way Logseq provides a built-in backlink block automatically, though I could implement that with a capture template for Denote. I like Logseq’s more wiki-ish page creation conventions. I like that there’s a mobile app devoted to Logseq’s point of view, as opposed to the general-purpose mobile org clients.

It’s important, for org-mode people, to not go into this thinking it’s “org-mode with a native toolkit GUI.” You can use org-mode syntax in a list. There are some familiar org-mode conventions (e.g. src blocks). If you like everything being an outline, as in org-mode, you’ll be in familiar territory. But there’s a layer of plumbing that finds its way into the markup that org-mode is not going to make sense of if you ever depart Logseq. It’s instructive to set up a few pages where you’re transcluding nodes, etc. and then open those pages in Emacs to see what it does with them. In some cases, it can’t do anything because Logseq has its own markup overlay.

Dusting off the camera

I’ll be in Vancouver, BC this week for a work event, then staying over for a few days to do some tourism. Given the time of year and likely subject matters it’s an X-T5 kind of week. I’m just taking the 23mm/f1.4 WR. I considered the X100V, but I’d like the extra speed and the image stabilization.

I have not been very excited about photography lately, but the prospect of a new place and interesting scenery got me a little more excited. Looking forward to being out on the streets at night.

Wallabag bookmarking script

I submitted a PR for my Newsboat Wallabag bookmarking script. As with the Newsboat one, you can just run it from the command line, too, if you’re interested in some pre-written plumbing for Wallabag bookmarking. Just takes a URL as ARGV[0] and lets the Wallabag service pull all the metadata.

… and as a followup:

Writing is a tool I use to think, and I don’t share a lot of it. I used to get frustrated because I’d write a few thousand words to make a point or express an idea and then “abandon” it in favor of just talking through things, but that was the attitude of a paid writer who thought of his writing as a product, not a process.

One day, though, when I was on a tech writing team, I became super frustrated with something I was documenting because some of the design decisions around it were sort of bad. So I took the time to write almost 4,000 words about how to un-bad it, didn’t really feel ready to share that with anybody, and set it aside. But I understood the thing I was writing about much better, and my docs around it really improved, and when I got to a place where I felt like sharing the ideas I came up with more widely, I was able to write a much more concise and digestible 1,000 words or so that didn’t include phrases like “this fucked-up Legend of Zelda configuration scheme.”

Since that experience, I got a lot better at just writing and not expecting to share, understanding that what I am doing is processing. Sometimes it helps to use an RFC template, false-start a blog post, or start an unaddressed email, because those contexts serve as cues or mood-setters. But I’m usually content to use whatever to do it because the output doesn’t need to last forever. It’s fine if it gets lost to the sands of time. It’s just a process, and wanting to “save” or “preserve” it makes little more sense than a baseball player trying to save a swing at the ball.

I think having a bunch of time to just fiddle with stuff put me in a mood to over-optimize, so it makes sense that I was getting obsessive and hyper-focused on note taking and writing tools, and getting super architectural about the whole thing. But having had a solid six weeks of being back in the work swing and having to actually use the notes I keep, I’ve been realizing how ephemeral they are. I’m glad I have them, but they don’t need to be architected. I’m careful to tag them up front with people and topics just to make search more useful and focused, but I don’t interlink much, haven’t really used the automated index pages I’ve cooked up, and don’t really care much about the heading structure or tidiness. Most of them will not be useful for much in maybe a quarter or so.

So I just made a folder in Apple Notes and started typing notes in there. It has tagging, it has smart folders, it has rudimentary formatting (title, heading, subheading, lists, todos) and you can make smart folders that pay attention to things like tags, location, unfinished todos, etc. I can do stuff I’ve been doing in Obsidian — inline todos I have a smart folder to surface for consolidation, smart folders for particular tags — but with no plugins that could age out, no need to really “automate” anything, no paying extra for sync, and no real room for mission creep. I know Notes will be around for the next OS release because Apple added some features to the beta, so it’s as future proof as this kind of content needs to be.

I sort of like it. There is not a lot to think about. Even less, in some ways, than flat, plain text, because Notes is just sitting there on every Apple device I own, and in a pinch is available via a web browser.

I also went through Things and asked myself what was up there. I’ve been using it more over the past several months, and there are already dead things, abandoned things, cryptic things. So I got out a legal pad, captured all the living things, and wrote all my active projects down at the bottom of the page. Then I started crossing things off as I did them.

When I looked at all the projects this afternoon — eight of them are “active” right now — I knew what each needed next. If I feel like I’m getting to a state of overflow, I think I’ll just make a list in Reminders for a project that needs more from me. But I live in a pretty well supported environment — there are plenty of trackers, there’s program management, there’s a steady cadence of meetings — I think I can manage.

I did find myself looking through my calendar and sort of Tetrising tasks on my list into my available space, jotting down the target block of time and a time estimate. I am sure apps allow you to do all that, and I know there are apps that also semi-automate the process of doing the Tetrising for you. I don’t know if I need something to do that for me.

I keep coming back to my old HR business partner, Don, who told me — when I was freaking out about an interim leadership role I didn’t think I belonged in — “you know what’s important, and you don’t do what’s not important, and that’s why we think you’ll be good at this.”

I know what needs to be done, know the state of the things I’m pushing forward, and I know when that stuff needs to be done. When things get super choppy I’ll break down and do a big dump so I can negotiate priorities with more precision, but, sort of like throw-away writing, that’s a process. I don’t want to leave it to a tool or suck all the friction out, any more than I want ChatGPT to do my throwaway process writing for me.

Someone did comment to me that they didn’t like the sound of copying things over from day to day. I get the resistance. It’s not necessary effort. The CO I knew who lived out of her legal pad explained to me that the copying was important to her, because it helped her think about what was important, think ahead about tomorrow, and ball up the list from today and toss it in the recycling bin before turning off the lights and going home.

You know, a legal pad is an imperfect system. If you move around a lot, you have to remember to bring it with you. It doesn’t sync to the cloud. It requires you to copy things, scratch things out, etc. On the other hand, it is worse at hoarding obligations than an electronic tool, it makes you think a little harder about what you commit to it, and it is nicer to draw a line through a finished thing than click it out of existence (in my opinion). I like having it sit there on my desk, just to the right of my mouse pad. I can look at it whenever I’m not sure what to do with a minute, and there’s always something. I experimented with pinning my Things “today” list on my screen to do the same thing, but that didn’t take.

Anyhow, I really don’t want anyone else to use legal pads to keep track of what they have to do today, and expect at some point I will be very curious about some other thing and will want to try it out.