Finding the right place: seeking advice on where to store my Johnny.Decimal system (in Windows) and single vs. multiple systems

I’ve completed reading both the website and the Workbook on the Johnny.Decimal System. While I plan to go through the entire process properly, I have a few preliminary questions I’d like to get help with to better plan ahead. I’m hoping to draw on the experiences of those who have been using the system for a while, as your insights might help me avoid some common pitfalls that I might not even know to look out for yet.

Background

I’ve been on my organization journey for nearly three decades, beginning with a very simple structure that naturally evolved as I grew older and took on more responsibilities. I’ve always been organized, but like many, I made my share of mistakes: As the volume of information I needed to manage grew, I created a system that became too granular, making it unsustainable; I filed things based on where they came from rather than where they would be most useful; and at times expanded the system without sufficient forethought.

A good while back, I was exposed to classification systems and learned how powerful they can be. This experience gradually reshaped my entire concept of organization, and my own organization system evolved accordingly, and I began organizing my information into distinct subjects and categories. This structure served me well for quite some time. However, a couple of years ago, my needs shifted, and I had to revise the system. Due to various reasons, I didn’t give this revision the careful consideration it needed, and over time, a degree of disorganization started to creep back in.

To prevent the disorganization from spreading further, I decided to revise the system again. However, I soon realized I was running out of ideas and wasn’t quite sure how to tackle some of the classification and structural challenges that likely contributed to my earlier missteps. In search of new perspectives and solutions, I embarked on research to see what inspiration I could find. After several months of exploring various systems, I found that most didn’t resonate with how my brain works. Then I discovered the JD System, and it quickly stood out. Not only does it share similarities with the approach I’ve been using for nearly two decades, but as I read through the website, I realized it addresses some of the decision-making processes I had been struggling with.

Furthermore, for the longest time, I had a strong aversion to using numbers at the beginning of file or folder names, which made me avoid incorporating codes into my systems. This limitation always held back their full functionality. Over the years, however, that aversion faded as I recognized the benefits of using classification codes. I gradually became accustomed—by choice!—to including dates and numbers at the beginning of file names for chronological sorting, version tracking, and easier identification, such as with project codes, so this is good a time as any to make the switch to a complete and proper classification system—and I find the JD numbering sensible and straightforward.

Question 1: System location

My first question is where to store the JD system. I’m using Windows, and my current folder structure is outside the “known folders” (the default folders created by the operating system in the User folder, such as Documents, Pictures, Videos, etc.) because I prefer not to use them. While reading through the form looking for some ideas and inspiration, my impression was that the majority of users (or at least the active ones) use Mac and place their JD system in the Documents folder. While I don’t use Windows’ default Documents folder, I can create another dedicated folde for the system, but I wonder if it might not be better to create the system directly in the user folder instead of accessing a folder just to get to my other folders, which I use each every day. However.

So, where would you suggest storing the JD system? Do you incorporate the default user folders into your JD system, or do you add the JD system alongside or within these folders?

Question 2: Single or Multiple Systems?

Currently, my main areas (in JD’s terminology) are:

Personal

Business I run my own single-person business, which includes everything from admin to client work and marketing.

Knowledge Management Notes, books, and research.

Development A collection of scripts, programs, websites, and other code I’ve written or use. Not that I’m a great coder. It’s more of a hobby.

Media Family photos, audio, and video.

I know that once I go through the discovery process, some changes will be necessary. For instance, my website might move under Business or become its own area. I also have a folder with portable applications currently in Development that doesn’t quite belong there—these are programs I use occasionally but didn’t develop myself. I’m unsure where they should end up, so if you have any suggestions, I’d welcome them. Other changes, additions, and omissions are also likely.

While reading the website and workbook, I initially thought that each of these would become an area within a single system. After all, the information is already roughly stored in an Area > Category structure. However, now that I’ve completed reading through the website, workbook, and some forum posts, I’m doubting this decision and wondering if it might be better to make each of these areas its own system. The challenge, and what sparked this doubt, is how to handle client work. I’ve read the advice on regarding the freelancer problem, and need to think about this more. Currently, I have a system in place, but it might fit better as an area within a system rather than just a category within an area. I’ll expand on this in a separate post.

Here, I’d like to draw on your experience and get your advice because, honestly, I’m unsure which option would be more appropriate in the long run. Reorganization is inevitable at some point, but I’d like to avoid having to redo my entire system simply because I started with a structure that was too restrictive.

I also realize that the recommendation is not to manage media using the JD system. However, I do want to include the main categories (photos, video, audio) as part of the index. From there, I plan to sort the data by date, name, album, or whatever makes sense, without necessarily assigning them an ID.

Many thanks for taking the time to read through my introduction and questions. Any input or feedback you can provide would be greatly appreciated.

Hi @Shai ,

Welcome!

  • I am on Windows (mainly), Linux (some) and Android (phones/tablets).

  • I am Apple/Mac/iOS illiterate, as I strongly dislike Apple’s end-to-end control.

  • I recognize parts of the journey you describe it, as I’ve been doing some of the same.

My previous system was like this (on all computers I used);

Folders on C:\"my initials"
strict set of subfolders, less than 10

Folders in "Downloads"
set of subfolders to sort different types of files

When I moved machines, I only needed to bring these two directories + whatever was under “User\appdata\Local & Roaming”

  • I discovered JD early this summer, and I’ve recently completed (98%) my transition to JD. (471 671 files, 2,7 TB from the past 25 years now lives in my JD structure.)

  • I ended up like this:
    Snag_a3504ada

  • I would like to avoid keeping several systems (because I don’t want to worry about “which system did I put it in?” (one of the problems with my previous system), I do believe the 20-29 and 30-39 might have to live at work (because IT Policy and information security).

  • (I am open to using separate JD-systems for projects or other ventures.)

Mistakes I’ve made (so far):

  • Not allowing categories to be broad enough, to allow all the files have “their natural place”. This made me end up having to re-do some of the JD structure. I now have some “overflow” that hasn’t gotten a place yet - temporarily living in the “#0 System” structure of each area. This is the 2% job that remains, but I keep cleaning and structuring to integrate the remaining stuff.

  • I allow myself to use sub-folders (breaking the 2 level rule) when it is necessary or natural, but I keep it structured. I see that some places this means “clickyness” to get down to the subfolder/content that I am looking for, but I can use Everything if I want to search more quickly without the mouse clicking.

What works and what I like so far:

  • I do not regret putting the JD storage system / folder structure in a NAS. (see backup link below for description). I has many advantages over keeping the folders on local machine.

  • I use this syntax in the index to show if there is a folder created.

  • I use Logseq for the Index. I considered Excel or Treesheets, but I am very happy with how it turned out in Logseq. It’s efficient and clean.

  • I have automated my backups.

  • This forum / community is pretty awesome!

Hi @fender
My answer for question 1 :

my JD system is located on the root of my Google drive. With the drive app.

I use my local storage only as an inbox (desktop and download folder).
Both are cleaning each week.

The drive app mirror the drive folders locally so I can make a full backup on a backup drive.

My files are accessible on any device.
Yes you have many pro/cons for using a cloud service.

Archives, very big projects or files are located on my SSD external drive.
I report the path on my index file.

The index file is my first search source.
I try to avoid file system navigation to find a file or a directory.

Here my searching process :

  1. Did I remember the AC.ID ? If so I use it to glove inside it with spotlight.
  2. If no I search into my Index file or navigate into it. Then fire the number inside spotlight.
  3. If the file is located on my external drive, connect it and navigate.
1 Like

Hello @shai, and welcome!

My system is in a folder called 001 in my home directory (linux).
I toyed with using the default directories as part of the system, but they just don’t make sense (putting all documents together, etc).

If I understand you right, I went the other way: I started off with several systems and have consolidated them all into one.

I’m fairly new to this, but I don’t foresee any problem with my approach of each client project being an ID within a category. Give yourself longer IDs (000 or 0000) to make sure you don’t run out, and if each ID folder has standardized subfolders, you can fit a lot in there.
Let us all know what further thoughts and questions you have!

I haven’t quite got to this yet, but I’m also still trying to decide on this issue. Either way, ‘albums’ will be listed in my JD system somewhere. I’d like to make albums (of photos, videos). I’m pretty sure I’ll store them elsewhere because otherwise my main Syncthing folder will get too bulky. I think I’ll end up uploading all the actual files to a server or NAS and generate web-based albums from them (with various resized versions, etc). My JD IDs for these albums will be date-based when that makes most sense, and will have a non-date-based unique ID when it’s more of a long-running album (e.g. “my daughter’s birthday party” versus “photos of my daughter at home”). I’m also pretty sure these IDs will not be JD IDs, but listed in a text file within an ID ‘AC.XX Photo Albums’.

I’ve also strongly considered the approach I think you’re describing: actually putting the files and albums in the JD system. For that, I would need to make this it’s own area since that is the level of granularity I use for syncing, but it doesn’t justify taking up a whole area.

Hi everyone,

First off, apologies for the delayed response. Sod’s law, the day after I posted, I came down with something and have been under the weather since. I’m back now and have had some time to think things over.

I really appreciate the warm welcome and all the insights you’ve shared. It’s been very helpful in shaping my approach.

@fender:

Your detailed journey through JD is truly inspiring. I can definitely relate to your initial struggles with category granularity. I’ve faced this issue before, and in some ways, I’m dealing with it now. The one thing I’m keen to avoid with my new system is overlap caused by categories that are too granular. I know some tweaking will be inevitable, but I plan to start broad. I’m quite happy with how broad my categories are at the moment, but I’ve noticed that when I expanded my system last time (rather hastily and without enough consideration), I fell back into the habit of being too granular.

Like you, I think I’ll need to break the 2-level rule occasionally—probably with my client projects system. To me, three levels deep (my current setup looks like this: Project folder / Client name / Year (and sometimes there also sub-folders for months for regular clients) / Project ID / Project content) feels manageable. By the time I’m at Client name / Year, I’m already organized in my mind.
I still need to figure out how to make my client system work within JD. Maybe some changes are required or there are better approaches. I’ll create a new discussion for this in due time.

You make a great point about avoiding multiple systems to prevent overlap. I’m torn on this. I think my systems will be distinct enough to avoid confusion, with a clear understanding of what goes where. But then I start thinking about Development, or how some code could belong to a business (like my business website) or personal projects (such as various AutoHotkey scripts I write to improve general efficiency and productivity on my computer), and that’s where doubt started creeping in.

I’m considering starting with a digital index file for each category. I’m using Everything Search, and there must be a way to create a profile that will limit the search to the JD index folder and only Markdown files (I’ll be using Obsidian on the backend). This should make finding the next available number both easy and reliable. I’ll need to experiment a bit, but I think I’ve got the basics of the index figured out. Thanks for the tip about marking which items are folders—I might adopt that approach.

@Mentat:

Thanks for sharing your setup with Google Drive. I’m using Syncthing to sync all my data across devices instead of relying on the cloud. There’s nothing wrong with the cloud, of course, but I’ve never been a fan of how it forces everything into a special folder. A matter of preference. I’ve found that Syncthing achieves the same result, minus the redundancy of having a cloud copy. Instead, I back up my data to two different drives, with an additional copy stored in the cloud—though I use the cloud as a remote drive rather than an intermediary.

Like you, I plan on making the index a central part of my system. I’m using a file manager called Directory Opus, which allows me to quickly navigate folders and perform all sorts of other neat tricks. I’m also using Flow Launcher, which is similar on Windows to Spotlight on macOS. It can be used to search the index (which will be in .md files) and quickly jump into a folder.

There will likely be some trial and error involved before I settle on the right workflow, but I’m confident that working with the index (managed in Obsidian under the hood) will be both straightforward and incredibly useful. I’m already using a primitive version of the index based on the USN Journal of NTFS (Windows’ file system), so I have a good sense of how powerful an index can be for finding what I need without having to dig through the file system or even open a File Explorer/Finder window.

@hans:

Your insight on consolidating systems into one is exactly what I was looking for. Thank you for sharing. Initially, my idea was to use a single system, but as I worked through the workbook, doubt started to creep in, and I found myself leaning toward creating multiple systems. However, the concern was, and still is, that having multiple systems could and would lead to more fragmentation and friction over time. May I ask what prompted you to consolidate your systems into one?

After giving it some thought, I’ve decided to place my JD system inside my home directory (User folder on Windows), just as you’ve done. My current system is already there, and it makes managing and backing up easier—especially since most programs store their config files in a hidden folder called AppData within the User folder. Using the Home directory proved to work well for me, it makes the most sense, so I’ll stick with it. I do avoid Windows’ default folders (I don’t delete them since some programs rely on them, but I do hide them so they’re out of sight).

Your method of handling client projects with longer IDs is instersting, and I’ll definitely keep it in mind. Right now, I use the ID from my billing system as the project ID (so I can find a project even if the only detail I have is the invoice number). Now I wonder if it will be technically possible to adapt my billing system to issue a JD ID as the project number so everything will match. That’s intriguing—thank you.

If you check my reply to Fender above, you’ll see more details about how my system is structured. This structure has served me well for a long time, so I’m hesitant to change it just for the sake of change. For many years, I sorted projects by year, month, and then project ID + client name, which over time led to fragmented client data and indecision about where to place administrative documents (agreements, signed forms, style guides, and every other piece of information that is not strictly a project or related to a specific project). Switching to organizing projects by client (with names sorted alphabetically) and then by year, and project ID reduced friction, and having all client-related information in one place has made things easier.

That said, I’m not set in my ways, and figuring out how to fit the clients system into the JD system is something want to explore further. I’ll need to revisit the information on the website, workbook, and forum, and I’ll probably create a separate post to get more advice and inspiration.

Regarding media management, I’m still undecided. There’s personal media (family photos), business-related media (marketing), and potentially shared media (which often overlaps with personal media). This gave me pause when I considered organizing media in its own separate system because I could almost envision the chaos creeping in as result of not having one clear place to put a piece of media. In reality, only the shared media (which could be considered personal media at this point, i.e. family photos and videos) would go there, and I doubt that justifies an entire system. An area within the system would likely suffice since to me it makes more sense to sort media by name or date rather than a JD ID.

I need to think this through some more, but this is where I’m currently leaning.

Thanks again for all your advice and experiences. They’ve really helped clarify my next steps.

Cheers,
Shai

2 Likes