Why I no longer use LOGseq for taskmanagement (but somehow still do)

I don’t really model my work using swim lanes anymore. My main page and its sections are included below.

I get done what I get done today and what doesn’t get done just gets bumped forward to tomorrow. I don’t fret at the bumping. But I do have the concept of hats (or roles, what is ordinarily tied to a bucket or swim lane). That is, if you write a blog post you may wear a Writing, Editing, and Publishing cap. I don’t generally wear two hats at once and I know which hat I’m wearing when I work on a task even if I’m not using buckets. If I need the reminder, I can just add a tag (e.g. #Writing, #Publishing) to the task/subtask. That #Writing tag will link to a page about that hat/activity. The point is my buckets are ephemeral. Maybe a task is appropriately tagged, maybe not, but I know what I’m doing when I’m doing it.

Some of the tasks I mark with a #25m tag. This reminds me that I’m just committed to about 25 minutes of work, what some call a pomodoro. The Logseq feature I lean on most is repeating tasks. In this way every project is itself a repeating task with subtasks. When I put in the time (on any of the subitems) I check it off and it bumps itself into the future. I use the Schedule date repeater to designate the habit and the Deadline only for real world deadlines. Even if a task has a Deadline it almost always has a Schedule. I would call what I do a systems approach (see James Clear’s Atomic Habits). I just keep chipping away a bit at a time until the thing is done.

I read 2 books on Kanban and dozens of articles, but in the end, when used personally, there was no special sauce in the swim lanes, the boards, and the dragging and dropping between columns. My major takeaways were limit the amount of work (WIP limits), limit lead time (ship started things to get them off the list), and have a way to visually see all the work in progress at once. But, like you, I am in a constant state of flux, rejiggering priorities daily. And I rarely completely everything listed for a given day. I don’t beat myself up for just bumping things forward. If something gets bumped forward (avoided) too frequently I mark it as a Lemon and focus on getting it off the list.

Also, and I think we may be on the same page, I don’t actually use parent level outline items as buckets. I use Logseq queries as buckets. By changing metadata I move things between buckets.

I’ve spent years trying different tools and I eventually settled on Logseq. I determined that in almost every case the productivity issue was me or my system and not the tool. I tried Trello for a while. Logseq is as good as it gets. It does well enough in the business of getting things done that I am no longer trying alternatives. It local-first software and Github friendly and that’s good enough for me.

Here are my current query buckets:

 :default-queries
 {:journals
  [
    {:title [:h4 "⌛ Overdue"]
      :query [:find (pull ?block [*])
              :in $ ?start ?next
              :where
              [?block :block/marker ?marker]
              [(contains? #{"TODO","DOING"} ?marker)]
              (or
                [?block :block/scheduled ?d]
                [?block :block/deadline ?d])
              [(> ?d ?start)]
              [(< ?d ?next)]]
              :result-transform (fn [result]
                        (sort-by (fn [h]
                                   (or (get h :block/scheduled) (get h :block/deadline))) result))
      :inputs [:365d-before :0d-before]
      :collapsed? false},
    {:title [:h4 "⏰ Today"]
      :query [:find (pull ?block [*])
              :in $ ?day
              :where
              [?block :block/marker ?marker]
              (or
                [?block :block/scheduled ?d]
                [?block :block/deadline ?d])
              [(contains? #{"TODO","DOING"} ?marker)]
              [(= ?d ?day)]]
      :result-transform
        (fn [result]
          (sort-by (fn [h] [(get h :block/priority "Z") (get h :block/created-at)]) result))
      :inputs [:today]
      :collapsed? false},
    {:title [:h4 "🍿 Watch"]
      :query [:find (pull ?b [*])
            :where
            (or [?b :block/priority "A"] [?b :block/priority "B"] [?b :block/priority "C"])
            [?b :block/marker ?marker]
            [?b :block/ref-pages ?p]
            (or [?p :block/name "movie"] [?p :block/name "show"])
            (not [?b :block/scheduled ?d])
            (not [?b :block/deadline ?d])
            [(contains? #{"TODO","DOING"} ?marker)]]
    :result-transform
        (fn [result]
          (sort-by (fn [h] [(get h :block/priority "Z") (get h :block/created-at)]) result))
    :collapsed? true},
    {:title [:h4 "☀️ Tomorrow"]
      :query [:find (pull ?block [*])
              :in $ ?day
              :where
              [?block :block/marker ?marker]
              (or
                [?block :block/scheduled ?d]
                [?block :block/deadline ?d])
              [(contains? #{"TODO","DOING"} ?marker)]
              [(= ?d ?day)]]
      :result-transform
        (fn [result]
          (sort-by (fn [h] [(get h :block/priority "Z") (get h :block/created-at)]) result))
      :inputs [:1d-after]
      :collapsed? false},
    {:title [:h4 "📆 Sooner"]
      :query [:find (pull ?block [*])
              :in $ ?start ?next
              :where
              [?block :block/marker ?marker]
              (or
                [?block :block/scheduled ?d]
                [?block :block/deadline ?d])
              [(contains? #{"TODO","DOING"} ?marker)]
              [(> ?d ?start)]
              [(< ?d ?next)]]
      :result-transform (fn [result]
                (sort-by (fn [h]
                            (or (get h :block/scheduled) (get h :block/deadline))) result))
      :inputs [:1d-after :7d-after]
      :collapsed? false},
    {:title [:h4 "📅 Later"]
      :query [:find (pull ?block [*])
              :in $ ?start ?next
              :where
              [?block :block/marker ?marker]
              (or
                [?block :block/scheduled ?d]
                [?block :block/deadline ?d])
              [(contains? #{"TODO","DOING"} ?marker)]
              [(> ?d ?start)]
              [(< ?d ?next)]]
      :result-transform (fn [result]
                (sort-by (fn [h]
                            (or (get h :block/scheduled) (get h :block/deadline))) result))
      :inputs [:6d-after :90d-after]
      :collapsed? true}
]}

The “Watch” is a reminder for some family time. We watch shows/movies a few nights a week.

6 Likes

I am the same spot as you. I use queries in the journal (config file) to show my overdue and personal tasks. I use the journal pages to write thoughts, meeting notes, TODOs (I use deadline dates). For projects, I seem a little lost with Logseq. The Kanban plug-in doesn’t allow me to move items back and forth (no offense to the person that wrote it…and perhaps I may be missing something). I have come to the conclusion that Logseq is good for some things and Obsidian is good for the rest.

I am also that person that has used so many programs or tools to achieve the ultimate workflow; however, I will not stay with one tool…I will use a bullet journal, then maybe a book to write notes, I will use Logseq (I use Obsidian to sync Logseq) and Obsidian. I have used ClickUp and Tick Tick. I have projects at work and trying to use the Microsoft software is a nightmare (No, I do not have planner accessible to me). I started using DevOps; however, it wasn’t working out.

Anyway, I don’t think there is a magic tool out there. I do agree with the person that wrote maybe I am the problem, not the tool.

1 Like

Great comment. Can you show me an example of how you structure your TODO task?

Example: TODO /a Do this or do that. I would add a deadline.

Thanks

I use Logseq exclusively, one program, for getting things done although I use an online calendar for scheduling absolute time/place appointments. I have different categories. Wellness is one. It includes:

TODO Exercise 
SCHEDULED: <2022-05-01 Sun .+1d>

Elsewhere in another category:

TODO Get pool excavation quotes
SCHEDULED: <2022-05-02 Mon>

Anything that is scheduled or has a deadline is detected and dropped into my buckets: Overdue, Today, Tomorrow, Sooner, Later (all time based). Both show up somewhere in a bucket on my Journals page. This page allows me to survey the event horizon. I might look only at today and tomorrow most of the time, but I will glance forward a few times a week.

Everything which needs to get done gets scheduled. If it takes more than a single session to complete it gets a repeater. The schedule is more reminder than agenda absolute, even getting the excavation quotes. The benefit of repeaters is I can take some action moving a thing forward without actually completing it. I still get the satisfaction of checking it off for today. There’s a difference between done for the day and done altogether and some things, like exercise, never go away.

In GTD David Allen says something about always defining a Next step, making it discrete and clear to avoid inaction. I do that in some instances, but not always. Much of what I do is ephemeral. If I have some reasonable next steps in my head I don’t don’t always bother to write them down on simple projects. It’s more flexible to not write them down. I can zig or I can zag. The most important thing about a task is it’s clear enough to get traction when its time comes, rather than procrastinate due to ambiguity. For the more complex projects, ones that deserve planning, I will break out details.

As an example of ephemeral, “Exercise” is not the gym, but my commitment to daily activity since I have a desk job. On most days its 45 minutes of walking. On another it might be 90 minutes mowing the lawn. My brain understands what I mean when I say “Exercise” so I don’t worry about its precision.

Dating everything with repeaters serves to complete larger jobs by chipping away at them. I’m not a person who is so regimented that I know what I am doing every hour of the day. I’m sure some can do that, but it doesn’t suit my personality. I apply enough dates to things to keep about 10 things in view a day. Some of those things belong to large projects potentially taking months to complete. I have between 1-3 big ticket items (Goals, Rocks, whatever you want to call them) appearing over the course of a week, the rest is just life and its routine responsibilities. Some of the big ticket items repeat weekly, some every 3 days, some daily. It really just comes down to time and energy and the priority/pace for getting the project to done.

3 Likes

@statix how did you manage to use queries with the kanban plugin?
Or is this other workaround to display blocks next to each other?

Great to see the topic, because I started using Logsec, I’ve been relying less on my dedicated task manager, TickTick.

I have to admit that though I’ve tried, I’ve never used task managers effectively. I typically get stuff done when I need to, and I use task managers to organize larger projects and recurring todos I might overlook without a reminder.

I started using the Interstitial Journal method in Logsec, and I have a Keyboard Maestro macro that reminds me on an hourly basis to log what I’m doing. So since I’m opening Logsec throughout the day, it’s practical to use it as a task manager.

What works for me:

  1. Repeated tasks (though there are some limits like I can’t set for example the first Saturday of each month as a task.)
  2. There are nonotifications for tasks, but at least I’m opening the app on regular basis see the current ones.
  3. Inputting tasks is sometimes not as easy as TickTick, but it’s manageable.
  4. Agenda, Logsec-Calendars, and Todo plugins are helpful.
  5. I can use pages for projects lists and outlines

I’ll continue to use Logsec for task management until I hit some roadblocks that stop me from using it. If it works, I can maybe get rid of TickTick and pay one less subscription, and one less app to context switch to.

1 Like

I tried Tick Tick and couldn’t make it work for me. I also tried Click UP. I have been using Logseq for about everything now. One thing I like to do is drag emails over to my Logseq into a TODO task. I am using the query that was posted in this topic and it is working wonderfully. I tried using the Interstitial Journal, but I guess I didn’t understand how to use it. Can you elaborate a little on how to use this plug in? I will try Agenda. I wish the Kansan plug in would allow you to move the boxes. Thanks

Yeah, I don’t think I’ll be renewing my TickTick this year, especially if I can keep using Logsec.

The macro I use for Interstitial Journal reminders is not a plugin, it’s a macOS automation application called Keyboard Maestro. It’s similar to Automator but way better.

It can perform all types of actions based on the actions you set for a macro. You don’t have to code.

You can download a free trial of the app, and my website https://macautomationtips.com has several articles about how to use it. I also have a YouTube channel, MacAutomationTips, that includes some videos about KM.

Let me know if you have any questions.

Thanks! I may give it a try. I have Typinator, but haven’t used it that much.

I think I’m in the same boat, though I’m only a few weeks into the LOGseq journey. LOGseq is really cool, but I quickly found the out-of-the-box experience to need significant customization and that is not for the faint of heart. I can see the power / potential of Logseq, and thankfully I like technical challenges but the learning curve is steep and I’m not to confident I’ll get Logseq to bend to my will.

Stacks looks interesting, especially since it stores data in local JSON file (which might be useful for integrating with Node Red.) First impressions are that creating tasks have a bit too much friction. One thing I like about Logseq is that I can do a brain dump very quickly, and then easily sort out the results later. I really want to minize the time needed to create and sort out tasks.

The lack of an easy, editable overview of tasks is the issue with Logseq’s UI. I’m trying to use @cannibalox productivity tools kanban feature to create columns, not so much for “kanban” but just for creating that one-page overview that helps me put tasks into context.

The overview is needed for context, whether that context is date-time, location, assigned person, etc. I need to see the big picture, not the daily journal. Then, easily move tasks into the correct context (including DONE).

The UI wastes a lot of screen real estate, which doesn’t help the matter. I’d like to dig into @cannibalox plug-in to see if its possible to tighten-up the formatting. I’m only marginally familiar with CSS so it takes me time to decipher and, again, we’re back to the learning curve.

1 Like

Hey, guys, I’m writing to share a setup that I’ve found that has been working great for me:

  • I’ve adapted the queries that @mlanza shared to get TODO, DOING and DONE tasks

  • I combined the queries with the kanban visualization provided by the logtools plugin (#.v-kanban-wide option) and placed each query in a column

And here is the result:

Some keypoints about how I use this setup:

  • By using queries tasks are pulled from everywhere in my database

  • This wide kanban visualization packs a greater amount of tasks, compared to a vertical kanban (which is what I was using before discovering this plugin)

  • To change the status of a task, I simply click on it and press CTR + ENTER to cycle between TODO, DOING and DONE

  • I also use a week tag for every tasks so I can plan tasks in a weekly basis (I use deadlines for a more granular control in the current week)

This setup is similar to the kanbank view @statix has shared, but the visualization provided by the logtools wide kanban is much better.

This combination of a nice and traditional kanban view and being able to use the power of Logseq queries is a killer feature for me.

I hope it is useful for some of you who posted here: @Deviation @TR-Ranch @ChrisVn

9 Likes

I will keep using Things for general task management, but use Logseq TODOs for pending work inside the knowledge management structure.

Same, now just gotta hope someone writes Things integration plugins for Logseq just as people did for Logseq.

I have been using my same GTD method I used in org-mode to track all of my task in Logseq. Each task will get assigned a context tag, which allows me to stay focused by making a clear distinction what I am expected to do on each task item.

The only real obvious change is the way tasks get tagged, I ended up using namespaces to track my various different context types, and another namespace for reference material.

Example of a multi-level task item that starts with a task and leads down to something I need to read and eventually digest.

So at any point of my day, if I want to spend some time reading, I can pull my reading task items and pick from a list, or if I want to work around the house, pull my house items… etc etc…

3 Likes