SSR

System Stack Research

Home About Links RSS

What bothers me about my productivity system

The deeper I get into thinking about it, the more I’m convinced: there’s an impedance mismatch between my brain and my productivity systems. None of the tools and services I use are actually a great match for how I need to use them throughout the day, and if I could unite it all into one system in one place my life would be improved by the consolidation… but I’m not sure that (1) that consolidation can actually be done or that (2) it would actually be any better than what I’ve glued together over the years.

I also struggle with how “spread out” this system is. Do I really need all of these services? Why is the workflow this Byzantine? How can I simplify, reduce external dependencies, and bring things into something less diverse?

Right now

My “system,” such as it is, has evolved over decades of trying to tame my brain. When I first read Getting Things Done and became essentially a religious zealot, I was all-in on OmniFocus, which made sense as I was a Mac user and iPhones didn’t exist. But as time went on, I drifted further and further from “pure” GTD and into a strange mélange of practices borrowed from GTD, from Getting Results the Agile Way, from random blog posts, and things I just made up on my own after working in Scrum and Kanban environments. So how many tools am I actually using to keep track of this stuff?

  • Todoist is where the rubber meets the road; I have all of my “to do list” items in there spread across all my various Areas of Responsibility. I’ve bounced around a bit between Todoist, Remember the Milk, and Amazing Marvin over the years, but I always end up back at Todoist.
  • Trello is where I track every (non-day job) project in my universe. I have Kanban boards organized by “Areas of Responsibility” (which we’ll get to in a bit), so things like “House projects” and “Truck projects” (for the old Chevy I’m working on) and one for this blog.
  • For my day job, Trello isn’t an approved software tool, so I track all the same information in Microsoft OneNote. It’s not as fancy, but even though it’s more legwork and doesn’t look as good, it gets the job done.
  • Org mode (in which I’m also drafting this post) is where I do two things:
    • I track “Open Loops,” or things I’m working on at any given time because when I’m inevitably interrupted, I need to be able to retrace my steps.
    • I also track my “Weekly Outcomes.” These are things I want to get done for the week, separated out into “Must,” “Could,” and “Should” categories. I track the status of these things (TODO,DOING, DONE, and DEFER for things that need to be pushed off).
  • In Word I create a weekly plan. I have a lot of meetings, a lot of kid dropoffs and pickups, and just generally a lot going on. It’s helpful to me to lay out my whole week, map out the meetings, find the blocks of time where I can work on the weekly outcomes–especially the ones on that “Must” list–and generally get a grip. The cool thing about the weekly plan (which I absolutely borrowed from Cal Newport’s Deep Work) is that when the week changes–as is always the case–I just start over and re-plan the rest of the week. It’s just a guidebook, a direction-finder when the week starts to get crazy.
  • Sometimes I make day plans on paper, sometimes I don’t.
  • Big picture mind-maps (things like yearly goals, reviews of all of my areas of responsibility, big-picture brainstorms about various projects) go in Xmind mindmaps. I have an old license for the “old” version (8, I think?) and am going to just use that until it stops working. Mind mapping software in general disappoints me, but I’m not enough of an app developer to make a better one.

The Good

The best thing about this system is that it mostly works for me most of the time. I’ve arrived at it over years of experimentation with a ton of different platforms, and this is the thing that I’ve just managed to glue together.

Todoist is the first star of this show, as much as I may keep trying to move to other apps. It has the best natural-language entry for repeating tasks, which are really important to me because I never remember anything and rely on my task system to do that for me. I really want to like Marvin, but it was a little too fiddly. I see its unlimited potential, but ultimately needed to go back to something with more guardrails.

The real magic that keeps me coming back to Todoist every time I experiment with something else, though, is the filtering. If I didn’t have things organized into projects and then filtered by parent project, I’d be lost. I have a “Today at work” view, a “Today at home” view, a “this week in errands” view… the filtering is where the magic really happens with Todoist, and is the thing I’ve never quite been able to replicate with any other platform.

Another good thing about it is that it’s mostly cross-platform. OneNote and Word are fine enough in the browser, and the fact that all these tools (except Org and Xmind, and Xmind really doesn’t get used much in this capacity) are browser-based or, in the case of Emacs, able to run everywhere, means I can use them on Windows, Mac, and Linux interchangably.

It also avoids data islands, for the most part. Working across many different platforms and also on mobile means I very rarely have to duplicate effort. If I make a change to one of these systems somewhere, it propagates everywhere.

The Bad

That “browser based” part gives me the creeps. I don’t own any of these tools and I can’t control their development roadmaps. The only one that is both open source and not running on someone else’s server is Org mode.

I don’t like relying on so many different systems to accomplish one thing. I could bring it all into Org mode, ideally, but then I’d still have to rely on some kind of cloud storage, and that means either (1) still relying on a proprietary system like Dropbox or OneDrive, (2) not really being able to do much on mobile devices, and/or (3) getting into the business of running my own cloud with Syncthing or something similar. None of those are appealing options, but they might be where I’m headed.

If I were smarter, I’d just stay put and say “this system works for me.” But in the back of my mind, because all of these systems are not owned by me and may not be around forever (excepting Org and Word, which will both persist until the heat death of the universe), I’ve already started to lose trust in the system, and that illusory total trust in a system is, true to what David Allen said in the original Getting Things Done book, the thing that makes it all work. Kick out one of the legs of the stool and the whole thing gets rickety in a hurry. I’m not there, but I feel it coming.

So what do I want?

Besides Fully Automated Luxury Communism, I’d like a system that I can either host myself or otherwise fully control, that would give me great to do list tracking with repeating task options, some kind of Kanban functionality, and the ability to work with Org-like outline structures to track status.

Things I could do now

I could probably replace both Trello and the OneNote project list with Org mode text files. I went through the process to get Emacs approved for use on work machines (!) so I’d have access to the same tools across all my devices. That would eliminate two of the laundry list of systems. The problem with that is that I like Kanban and I like Trello.

I could move the Todoist stuff somewhere, maybe. I’ve tried Microsoft To Do, I’ve tried Remember the Milk, I’ve tried Amazing Marvin, and TickTick and a few other things, and none of it worked the way I wanted it to. I’m probably just “stuck” being a happy Todoist customer, which is fine. But these problems at all the other levels of operation still bother me.

Things I could build

I addressed some of this in my post about why I ordered an M5Paper. Maybe there’s some sort of “e-Ink device that uses Org mode files and the Todoist API” thing in the future. Maybe it looks more like a Raspberry Pi Zero W attached to the back of a big e-ink display, with more horsepower and the ability to run real Emacs/Org and a browser. I’m not sure.

This is a space in which I plan to do a lot more playing as time goes on, because I think “using a bunch of laptops and a phone for this” is a lot less satisfying than “this is my one tool that I use for this.” I’m not sure a web browser or a PC is really the best place for all of this to take place; it’s just the best tool we have at the moment. Cheap computing power in small packages may finally be starting to change that a little.

Basically, the ideal for me here is a Bullet Journal with a brain: a single device or object that is the source of truth and the home of my Big Task Management Tangle. That’s what I’d build yesterday if I already had the skills, budget, and materials.

Next steps

Besides waiting on the M5Paper to get here from China, there are some other things I could do.

  • Wekan is an open source Kanban web app; maybe I could host it somewhere privately. I did this on a Sandstorm instance once… it could be time to do that again.
  • I need to experiment with self-hosted cloud options. I could theoretically use a Raspberry Pi as a Syncthing “home base,” or find some kind of extremely cheap Linux VM I could use for the same purpose.
  • I need to experiment more with the task management and agenda pieces of Org mode, because I’ve never been comfortable with them and I’ve always just given up instead of figuring out how to do it. I need to muddle through and see if it can do what I want, instead of running in fear when I get past the basics of outlining and task statuses.
  • I want to show my work on the “Open Loops” and “Weekly Outcomes” textfiles, because that’s a cornerstone of my Productivity System (it feels like you have to capitalize it and acknowledge the weasel-wordyness of the whole thing) and I haven’t really seen anyone else writing about doing something in the same way.
  • I’d like to experiement with Lotus Agenda, the old DOS app, just because I feel like it could be a source of some ideas to steal adapt. I’ve always been curious about it, and it would give me another excuse to write a blog post.

I’m sure there will be more to come on this topic, as this is more of a list of things that bug me than a proper manifesto or charter for a new direction in this field. Those thoughts aren’t fully formed yet, and may never be.