Personal systems

created: 2020-06-03 · modified: 2020-10-31

Basics

Personal systems distilled down to a few core practices:

  • Need a place to jot down fleeting information. This is spur of the moment thoughts you’d like to save for later, observations you’ve made worthy of keeping, etc.
  • A way to manage tasks, i.e. a TODO list. Ideally this would come with a calendar of sorts.
  • Need a place to store more polished notes, a place where scratchy fleeting notes mature and become useful over the long-term.
    • Includes usefullness of backlinks
  • A way of creating notes related to external sources e.g. books, articles, research papers, etc. This is a primary source of new information for your system, and can often lead to more complete polished notes. Includes details about
    • sources where external sources may be found
    • how external sources are managed before/after being read (bookmarks, read-it-later app, archiving useful references, etc)
    • how notes are taken on a particular resource (articles, PDFs, physical books). Are there ways annotations or highlights are being managed? How are physical notes converted into digital notes?

That’s pretty much the basics. By no means do these practices make a perfect system; other ideas and principles can vastly improve upon just these few. But these are the axioms, in a sense.

Implementation

Digital

  • Plaintext only: no exceptions here. Everything should be stored in plaintext. Apps go down, services die, formats fade…but plaintext will always be accessible. It’s the best long-term option. Markdown is good.
    • Note that this does not mean that you can’t use an external app to operate on your collection of plaintext files (e.g. Obsidian
  • Self-host your notes. Again, services die and so will your notes if you don’t own them yourself.

Why?

Nobility, responsibility, sanity (anxiety reduction)

Philosophy

  • Completeness
  • Efficiency
    • Economic driven: design embraces a cost-benefit specific-generalized tradeoff awareness when building system structure. To cover desired landscape, one can get the bulk of it with a large components (notebook, resource archive, knowledge base), but there will always be smaller scales necessary to fill in the cracks (research lab, trackers). This is analagous to ice and water when filling a cup (large structure like ice cubes can only fill so much, then you must use a finer scale to fill in the remainder).
  • World model representation
    • Prosthetic: at the end of the day, the system is prosthetic, an extension of one’s mind. But it intends to remain just that: an extension. It is not to replace the mind (as seems to be the case for the individual that wrote “Mind Mapping”)
  • Built around learning processes, memory models
  • Sufficient dependence on and easily extendable hierarchical structure (resource archive → collection subsets → internal notes → nested headings)
  • Encourage highly connected pieces (linking within knowledge, referencing canonical topics in writings)

Balancing Autonomy

  • Balance between entirely decoded mental structure, complete dependence on system for managament (e.g. Mind Mapping) and light prosthetic knowledge systems and scheduling
  • A system should be able to satisfy the user’s desire, being skewed in the preferred direction. That is, as long as they are aware of the tradeoffs involved. Ideally a well designed system should strike a good balance between extending the mind and reducing working memory stress while also enabling sufficient autonomy away from the system and building long-term memories (I feel as though I’ve been on both sides of this: using system lightly, to recently more heavily. The sheer amount of thought that goes into changes, their implications, loss of generality of specificity? Will I remember where these things are? etc. It’s hard getting this balance right for the individual.)

Cluster v Stratified Management

  • Managing from above all at once? Or from within use case? May or may not be a useful anology

Indexing

  • Most of the system’s job is to serve as an expectable infrastructure for whatever may need to be entered. Thus, a large part of the system design is ensuring that this indexing is clean, without redundancy (canonicalize everything), and simple. The worst that can happen is indexes getting needlessly complicated and too much friction when making additions.

Components

  • Information management
    • Knowledge base
    • Resource archive
    • Notebook
      • Daily notes
      • Writings
      • Progress log
    • Research lab
  • Task management
    • Task system
    • Objectives
    • Calendar events
  • System management
    • Dynamic reevaluation
    • Multi-scale consolidation
    • Personal tracking
      • Daily activity/habit logging; need history to build from to understand where you’re going
      • As-you-go logging (e.g. daily log summary)
      • Automatic time tracking software
    • Rich changelog

Inspiration

  • Mental models
    • Spaced repetition for memorization
      • Weekly and monthly consolidation is a form of spaced repetition on daily thoughts (light inspiration for feature)
    • Feynman technique for understanding
    • Recall practice
      • Recalling (daily, weekly, monthly) events and thoughts, dream recall, DNB practice (light)
  • Journal structure
    • Bullet journal
    • PARA method
  • Note taking methodology
    • Simple nested heading
    • Markdown-like mark up
    • Color emphasis
  • Goal setting
    • Situational analysis
      • Backcasting and pre mortem
    • Time traveling
  • Practice
    • Failing early and often
    • Make explicit fail logs, document why and how you failed (extremely important)
    • Pomodoro technique for focusing, task work
  • Sources
    • Wikipedia as a system
    • Gwern personal structure
    • Less Wrong sources
    • Reddit sources
    • Arbitrary
      • Weak top ranking article (bad structure, bad system): link

External Systems

Observations from gwern.net

Site Philosophy

  • Focus on “long content”, dynamic content to be changed and improved over time. Collection of writings is considered a continual WIP,

Information Sources and Ideas

  • “I grab things I see on Google Alerts & Scholar, Pubmed, Reddit, Hacker News, my RSS feeds, books I read, and note them somewhere until they amount to something.”
  • Link to Finding my ideas

Managing Information

  • Important facts or quotes → spaced repetition system
  • Processing (research) papers: read paper, making exercepts along the way. Push exercepts and paper to (Evernote)
  • Custom archival of web pages (to prevent loss from rot)
  • RSS subscriptions for managing feeds
  • Event reminders and follow up sessions marking in Google calendar
  • Time for personal data (e.g. in the morning), as well as automated recording

Reflection

  • Most of the managment process is handled by simple systems. Most are things I already do: manage a calendar for events, track personal information like the progress log (and previously habit tracking), make brief excerpts whilst reading.
  • Things I do not currently use from this list
    • Integration with spaced repetition. I’ve yet to find (or really look) for a nice workflow to transform information from the system into something like Anki. I’ve long thought about it, however, and it is essential for long-term memorizaiton of small chunks of information.
    • Managing a feed: I do not currently have a nice system for stream management (working on it now though). RSS or Feedly seem like a decent approach, but ideally I’d like something connected to Notion. I currently manually curate things I use on Reddit, Twitter, and elsewhere on the internet.

LW :: How I Am Productive

Obviously inspired by the bullet journal

  • Important rule: write things down, from simple ideas to things I see
    • Gets idea out of head, no stress to remember and dwell on it
    • Things you think aren’t worth writing down often come back to bite you; write it down anyway

Organization

  • Record events (time and place) in a calendar
  • Task tracking: keep a to-do list
  • Zones: organizing the current stage of material in the system
    • Action: things that need to be done now
    • Waiting: things that eventually need to be done, but require something else first you’re currently waiting on
    • Reference: material you might want to reference but is not associated with a task (e.g. passwords)

Prioritization

  • Do what’s important and urgent first, and what’s neither last
  • Planning a day in advance: blocking out periods of time in the calendar for working on particular tasks

Do

  • Pomodoro Technique: 25-5 minutes focus-break ratio
    • During 25 minutes of focus: do nothing but the task at hand for the entire time, no multi-tasking, no distractions
    • During 5 minute break: do anything but the task at hand. Cut completely from task. Helps restore focus, can provide new perspective

Review

Review materials on multiple time scales

  • Daily review: keep track of daily habit success (e.g. diet, exercise). Record time spent on various tasks, think about successes and mistakes from the day, create/tweak the next day’s plan
  • Weekly review: sift through zones, ensuring material belongs where it is. Do high level weekly planning
  • Monthly review: reflect on habits practiced, what to add and what to remove. Writes up reflection and publishes on blog.
  • Six month review: consider life trajectory and goals, what’s being done to accomplish them, mistakes to avoid, new goals to pursue, write a personal mission document for reference over the next six months

Reflection

  • Some decent pointers in this essay. Most mentioned things are actively part of the system, likely due to the overlap with a bullet journal approach. I mostly reject the “zone” implementation, as I have a similarly effective system.
  • Things I don’t do that I’d like to consider
    • More detailed planning of future days. Integration of tasks and events seems a little scary, but perhaps it’s a good idea. Or at least assigning goal times to the tasks that are scheduled for a particular day.
    • More scrictly implementing focus blocks: the Pomodoro Technique seems like an interesting approach
    • Multi-scale review: be more detailed when tracking personal information. Perhaps rethink current tracking systems to be more friendly. Weekly and/or monthly reviews seem like a good idea for periodic note consolidation and system cleansing. Also for ensuring higher level goals are being met in the long(er) term.

LW :: My Workflow

Devon’s Memex

  • Uses Highly (a chrome extension) to highlight web documents as they’re read. It then saves this highlighting and displays it anytime the web page is visited. I like the idea of using this as it’s a means of direct interaction with the source material without having to bring an entire article into the system. Should consider using this.
  • Looking at her note structure, it almost seems as though there’s a lack of formality and structure. Planning or compilation notes (things like the class or coffee note) are sort of on equal group with weekly outlooks and things. A topic gets its own page and from there it usually makes enough sense maintaining the internal info. If no page exists for what you want to store, create a new one. This is sort of what I’ve done with the raw notes in the notebook, perhaps embrace that a little more when it comes to things that may not fit anywhere in particular.

How to Think

Reflection

  • I think I naturally am inquestive and generative when taking notes, although more likely in a casual setting. When taking lecture notes, I think I’m more afraid to be sloppy and jot down thoughts or questions in preference of an error-free reference. However this is a bit backwards, and you should seek first to understand and then produce a correct reference afterwards. This might be something to keep in mind.
  • Have heard loads about working backwards from goals. Start doing this more often, create a template to write up before starting any objectives.
  • Long-term plans have been focused on via objectives. Something I lack here is a more formal “personal mission statement”. This should be completed.
  • Quick mistakes and documenting errors reminds me of ACT studying habits. This was an incredibly effective strategy. Build this better into studying, specifically when doing practice problems (e.g. homework).
  • On protocol documentation, this is exactly my recent thoughts with detailing processes explicitly and with diagrams. Nice to see some reinforcement here, and should take this seriously.
  • Documenting everything is the focus of this system, so we’re on our way here.
  • Logarithmic time planning feels very natural, don’t think this is something that has to be specifically enforced.

HN :: How to organize personal knowledge?

HN :: Do you keep a personal knowledge repository?

PARA Method

  • The PARA acronym stands for Projects, Areas, Resources, Archives:
    • Projects :: a series of tasks linked to a goal, with a deadline
    • Area :: a sphere of activity with a standard to be maintained over time
    • Resource :: a topic or theme of ongoing interest
    • Archive :: inactive items from the other three categories
  • I don’t mind the hierarchical structure of this system. It’s pretty natural and nothing new; tasks are assigned to projects which belong to “areas of responsibility” like Hobbies or Product Development. This all makes sense, and isn’t anything special.
  • There is virtually no information about Resources, or how exactly they are supposed to look within the system. I also almost entirely disagree with their definition of “resource”; they aren’t topics or themes. To me, a resource is some discernible body of work related to a topic/field, such as a book or article or website.
  • Overall I find the site for this method pretty weak. It gets away with not being horrible simply because of how general and obvious it is. I hate the feeling of the site and how the author is touted as a “productivity coach” and sells a $1000 course for “building a second brain”. It all just feels a little scammy.
  • As I’ve stated in the past, I don’t subscribe to the idea one pre-existing system will work for everyone or even anyone particularly well. You have to mold the system to your needs rather than conform to a strictly structured set of practices. However that’s a pro for this method as it appears to allow that room for flexibility and building out your own “areas”. That said, there just not that much going on here and no interesting points or ideas for me to extract. Everything is just completely general and obvious, and my system makes use of virtually all of the concepts discussed here in some way or another. If there’s anything for me to process further it would how exactly areas are used; I’ve sometimes struggled with where new page structures (e.g. Leetcode log) should go other than just the general notes page. But this describes areas as having projects, so maybe the Leetcode log would belong to the broader area of programming?

PARA Inspired Notion

  • Implements the idea of the “area” from PARA by having a simple page template for tasks and projects. All area projects and project tasks are located in the same database (similar to my objectives and tasks), but remain separate when visiting each area/project individually.
  • This system implements a number of areas including Academics, Learning, Health. This is a fairly compact and useful collection of pages. Each of these inherits from the same “area” base template, having projects and tasks links for that area. This is nice that the structure is generalized, but there are literally only three areas. I find that the term “Objectives” is actually more fitting than projects, especially for Health. Each area then has its own custom set of notes/databases (e.g. exercises database). I think the looseness here works pretty well.
  • That’s pretty much it; there isn’t a whole lot of difference between this approach and the way you would intuitively approach building a collection of pages for each of the areas. There is virtually nothing on Archives and Resources.
  • Seeing this system and the PARA technique makes me aware of the fact I’m not missing out on anything with my system. Early on I was slightly worried that I wasn’t doing everything the best I could be, and sure there are some areas that could use review and possibly changed up to improve efficiency. But since really leaning in I stopped worrying about this because I felt it hit pretty much everything I was wanting out of it. I have processes for describing wants and observations from working with the system, and those suggestions can be implemented easily.
  • Another quick thing that I noticed here was how little focus seemed to be placed on managing scheduling. There appear to be dates worked into assignments and other deadlines, but there’s no global addressing of events. To me this makes the system feel somewhat incomplete.
  • All in all I’m slightly uninspired here. I’m trying to keep an open mind and not be biased by my own practices, but there isn’t a whole lot of value I can take away from this system. It appears to work well and I don’t have many issues with it, but there aren’t many things I can bring into my own system that don’t exist in some way that’s as good or better than what’s seen here. If there’s one thing I’ve taken away from seeing this and reading up on PARA, it’s slightly larger scale organization of projects and/or pages from something like the Notes section in the notebook. It feels as though there could be a way to build out this infrastructure just a little better, or add some more global organization on this front. I still like the idea of Fields being a collection of global categories that almost everything can be placed under, but it doesn’t fit well with slightly edge pages like Bills or Quote and Fact Dump.

Johnny.Decimal

  • This is literally just indexed nested folders. There is almost nothing else do this method. There is two-digit number for a broad category of things, and then a two-digit number for the ID of more specific documents with those categories. That’s about it.
  • This system doesn’t claim to be a large, involved PMS or anything; it claims to be a system for organizing projects. I like the simplicity, but again there’s absolutely nothing new here. It’s just number nested folders, and telling you not to make too many of them so things don’t get complicated. This is pretty much not helpful to me at all. My Documents folders are already split up into separate categories, and can hold their own folders/documents pretty straightforwardly without numbering them.
  • With this page and the the PARA method, a lot of these “systems” feel like people just trying to formalize some extremely basic, general practice (like literally creating nested folders and indexing them), slapping a catchy name on them and promoting them as a life-changing infrastructure. They’re useless and boring. I like my system.

Planet

  • An interesting personal site management system somewhat similar to the XXIIVV wiki. Includes a way of processing pages, projects, articles, and tracking activity/progress toward tasks. This all has a nice, interactive web front for visualizing this.
  • Main takeaway here are the project pages, tag structure, and statistics/monitoring.
    • The tag structure is fairly arbitrary; pages and articles are tagged with whatever topics they are associated with. There does not appear to be any particular hierarchical structure, just loose topic association.
    • The project pages have tracked versions, relevant links like Github repos and documentation, as well as percentage association with creation categories like “programming”, “writing”, “visual”, etc.
    • Log entries include dates, a tag for the type of work, and the duration of that work. These are compiled into a Github-like activity map, as well as accumulated into other useful activity statistics. With something like a Notion API, I could export my own tasks, their descriptions, time worked, etc to a local database and run incremental processing on build. Or just whenever the site is compiled, pull from the Notion API for my tasks and process the statistics then.

Additional Sources