"At the June 2019 conference of the American Library Association the Council voted to remove Dewey's name from its top honor, the Melvil Dewey Medal; the resolution cited Dewey's history of racism, anti-Semitism, and sexual harassment. The resolution was passed overwhelmingly with no debate. The award was renamed the ALA Medal of Excellence at the Association's January 2020 conference"
I’m not a historian, my friend. Dewey’s kinda famous, my mate basically said it out loud one day, my name is Johnny, the system has a decimal in it, boom, that was it. At the time I was the only person using it and the idea of a website was years away. Let’s not read too much in to it eh.
I remember when everything was on network drives, and no-one understood the folders other people had created, so they created their own. So you end up with three 'IT Projects' folders, in three different places. And worse, multiple copies of the same document in different places, all slightly different.
Sharepoint should have been able to fix all that, but somehow it just made it the same, but slower. Even after 20 years, I never know where the f__k I am in sharepoint, or when someone sends me a link to a sharepoint doc, I can open the doc, but I don't know where the f__k the document _is_. I dont know quite why sharepoint is so confusing, I think its because it tries hard to pretend to be something other than the document store that it should so obviously be. Just when I get the hang of it, they redesign it all, and add some new front end like Delve.
If I join a project, something that rings alarm bells is if the Project Manager has created sixty empty folders in Sharepoint arranged into a three level hierarchy with different folders for every stage and stream of the project. Most of the time, most of the folders remain empty for the duration.
So yep I really like this Johnny.Decimal idea, simple and workable.
> So yep I really like this Johnny.Decimal idea, simple and workable.
Seems to be no different than this:
>something that rings alarm bells is if the Project Manager has created sixty empty folders in Sharepoint arranged into a three level hierarchy
Except that it's a hundred empty folders in a two level hierarchy replicated by convention across different systems, including Sharepoint if your organization uses it.
Those here arguing that things can, in fact, be found, or saying that my claim that "nobody can find anything any more" is absurd, clearly haven't worked in the ~5 large enterprises that I've worked in over the last decade.
Internal organisation in enterprises is shockingly bad. Remember, these people aren't techs. They have no idea what `find / -iname file | grep whatever` means. They crack open SharePoint and rummage around. When they want to save a thing, they just put it wherever. It's chaos.
I'm just implementing this for a few new projects at my current place and my boss has messaged me a number of times expressing his joy at just knowing where stuff is.
> “Hey Kristy, where can I find the payroll schedule?”
> “Twelve dot oh-three.”
Or, you know, don't encumber the brains of people with useless numbers, and use computers for what they are actually good are, at least better than people, e.g. use the search feature of your file browser for "payroll schedule".
That's cool if you know exactly what you're searching for. It gets less useful as the amount of information you have drops - you basically have to do multiple searches and hope one of the terms you're guessing will be a hit. Having categorised folders on the other hand means you can narrow it down and be sure that your search process is "exhaustive".
In practice it looks like this: here's stuff matching "payroll schedule" - "payroll schedule 2018", "payroll schedule template", "payroll schedule for contractors", "payroll to reschedule July 2017", etc.
It's a common issue with intranets. There's lots of content, but not enough traffic to figure out what you may be interested in. And even if there was, you may not be after the "most popular" item.
Don't forget that the document you may be looking for may actually be called "payment schedule" instead of payroll schedule because someone wasn't good at following the implicit, unwritten naming convention.
This is a real example from a past life, and really works when you're shouting across an office or talking down the phone.
The longer original, edited down for brevity, went something like:
"Hey Kristy, where can I find the payroll schedule?"
"Okay go to the Finance folder. No, the one in the G: drive. Yeah now go to Payroll ... oh sorry, it's Payroll 2020. Yeah no not the Payroll 2020 in the Payroll folder, the one at the root. Yeah. Now in there ... hang on let me have a look ... okay there's a 'Janice' folder and in there..." ... and so on, ad. absurdium.
Search is not reliable or reproducible. People search for things in different ways, and the content of a file is separate from the location while both are important.
And search systems today are still terrible. Different interfaces, searching different locations and content, with different ranking and relevance and operators. Much faster to just direct someone where something is and let their brain figure out the rest, because we're actually very good at that.
Search is awesome. I agree, gmail's is great. Apple's Spotlight is also fabulous. So why does the article say "search doesn't fix this?"
The problem is when search extends across domains. I can search files with Spotlight, but not my google emails. On my phone, it's worse because there are so many different comms systems. I can search my Whatsapp messages; my SMS messages; my email; Facebook messenger; Skype.... Each individual search is fine, but no one tool does them all. Hence, the idea of using manual labels to classify stuff - including emails - can still be useful. Crazy that we're forced back to this.
There is a product who's name I cannot remember that indexes all your stuff (local, cloud) to make it all searchable - it's not Google but your index is in their system.
Edit: Seva and Cloudtenna
I wish I had a local one that just packed it all into Solr
This is why fzf is so great in my mind. The longer a file path is, the better chance I have of finding it later (whereas deeply nested file paths make it harder to find files using something like Finder or ls)
These days I hardly ever bring up file naming in code reviews—fzf just makes it so easy to find everything.
Yep, real time fuzzy finding makes it so easy to find stuff quickly.
You can hit CTRL+t from the terminal after installing FZF and it will let you fuzzy search for any file you have. So you can do something like run `vim [CTRL+t]` to quickly find and open any file you want, or `cd [CTRL+t]` to switch into some deeply nested directory.
Even inside of WSL 2 with a 6 year old workstation on a first generation SSD it takes ~3 seconds to index 190,000 files and once that index is built, narrowing down results while typing is close to instant when using ripgrep as FZF's search back-end.
The amazing thing about all of this is it's dynamic. There's no having to create aliases or pre-defined directory structures. I only discovered CTRL+t from FZF a few weeks ago but once I did it was almost as good as discovering CTRL+r for the first time.
It seems that for this to work, existing directories can't be renamed. However with time the initial structure might turn out (or become) not optimal. In general I like the idea but not having a way to rotate the structure is a significant downside to me.
This idea looks great, and it would be great if it worked: I've been struggling to keep all my documents organized many times, and I think I've followed basically the same system (except for the number.mapping).
This is my gotcha: how do you decide where a document go? The categories might be clearly split, but documents typically won't fall in exactly one category.
How do you manage temporary tags? That is: how do you operate on the documents? You might be having some documents for your "2020 tax return" in several categories of your system. And what now? Do you duplicate them in another category (tax return?)? Do you tag them? How do you keep track that you moved them? (In case they're real objects, like pieces of paper).
> This is my gotcha: how do you decide where a document go? The categories might be clearly split, but documents typically won't fall in exactly one category.
Symlinks!
> How do you manage temporary tags?
Symlinks again!
Basically you put everything in an incoming folder, maybe organized by year or month/year or even type. Then you symlink to all of the categories it would be interesting to you.
Your temporary tags would just all be a folder of symlinks, so you can kill the tag and retain your document.
My advice to people is usually, though, that your brain will sort it out for you. And it really does.
The canonical example is "home insurance". Say you have a category of "home" and a category elsewhere of "insurance". Perfectly possible.
Which one does it go in? Well, you designed this system. Your brain came up with it. I've found that in nearly all cases, you just know.
I see where this may fall down in a shared environment (work), which is why I advocate (in one of the many unwritten posts on my as-yet-to-exist blog) for the role of a 'librarian' at work. We seem to think that 20 people can work on a project together and all just magically decide where everything goes. That has been shown again and again to be a fallacy that we believe search will help us with. My argument -- from experience -- is that it does not, and that we need to go back to a world where we organise things, and that people are responsible for this organisation.
I do not like your organizational system (nested folders are a great way to collapse conceptual complexity
IMHO) but I thoroughly agree with the need for someone to dictate organizational systems for a group project - my time in the animation world taught me the value of everyone who touches a project having a little cheat sheet or two stuck on the wall near their desk that tells them how to name things so as to not cause conflicts when merging with others’ work, and to make it easy for other folks to find stuff!
My addition to this: I add “Destroy after {reasonable future date}” to documents that I need to “keep”. I know, for example, that I need to keep my medical receipts for 12 months, but will never reference them after that.
When I go through old documents, I don’t even skim through the contents. I just check the date and remove them if necessary.
Eh, I'm not convinced. This might be nice for a static hierarchy of not too many office documents, but for the very heterogenous files I deal with, I'm happy with the first level of division being a "project". There are only ever a handful of projects I'm working on at a time, and different projects' files are usually suited to very different directory structures.
That said, the only documents I need to keep in long-term storage are invoices and contracts, which I can just put into two directories. If I ever get too many of such official documents, I might look into a structure like this.
When I discovered this before, I gave it a try and organized my email account.
What I can say is that I like it for helping you choose levels of hierarchy. So far I've squeezed 10 or fewer sub-categories into each chosen category. (I use the 0s, so within 10-19, I start with a sub-folder of 10...)
What I do not do is any of the other suggestions - searching by it, memorizing numbers (though I have an intuitive sense now when organizing email or looking through it.)
I have not tried this folder structure yet for files.
I don't find tags useful for many of the same reasons that I invented this system.
What's the tag for the payroll schedule? Is it 'Payroll'? 'Finance'? Something else?
Who manages the tags? Can anyone add one?
Tags need managing just as much as -- if not more than -- a well-organised folder structure. Tags aren't just some free magical way to organise stuff ... otherwise I dare say they'd be more popular.
"What is the tag?" is a category error. It's "what are the tags?" Tags are not paths.
Is it Payroll? Yes. Is it Finance? Yes, if that's how your org does things. Is it other things besides? Yes. A "superseded" tag would likely be useful to many organisations.
The idea of tags as opposed to paths is that a leaf object can have lots of them.
Of course tags are not free. The question is, does the value they bring outweigh their costs? Like you, I think the answer is usually "no". People suck at tagging.
If tags are to be used, "who manages the tags" is a key question that needs to be explicitly asked and answered at the same time as, or before, deciding your categories.
However the same applies to managing the "chart of accounts" that Johnny Decimal requires. The structure that is best for the organisation almost certainly won't be optimal for any given staff member. So management is required.
These are both stewardship problems.
The Johnny Decimal website is silent on the subject of stewardship. Here I want to focus on two aspects: keeping things working, and managing succession.
Keeping things working: using a gardening metaphor, is planting and weeding, re-potting, trimming and pruning and harvesting, turning over the garden beds, and so on. Like gardening, this part of records stewardship is best carried out regularly and frequently.
Obsoleting things is a key part of this. Contra the poster above who says "just delete it"[1], often you have to keep no-longer-current versions of filed objects around for a long time, for some combination of reasons: the maintenance and repair manual for a machine tool model we stopped selling 20 years ago, minutes of board meetings from an equivalent time ago, commemorative photos, old logo artwork, client records (if, say, you're a dentist) for the life of the client, even if you haven't seen them for a decade, ...
Succession is about ensuring groups have the right access, new people belong to the right groups and are properly indoctrinated, and things are handed over when staff leave or change roles, and ex-staff have their access removed; and about revising the structure, ontology, and access rights so that they stay relevant as the organisation they serve changes.
Records management is somewhat like plumbing--good drains, no-one notices; with bad ones, everyone is miserable. And keeping the drains "good" requires work.
The problems that are now universal ("Acme Corp Contract", "Acme Corp Contract - final version", "Acme Corp Contract - final with client revisions", "Acme Corp Contract - senior management approval", "Acme Corp Contract Brian", etc.) have indeed come about because computer-illiterate and filing-illiterate office workers thought the FAT file system (and then NTFS and/or Sharepoint) was a magic, zero-effort substitute for having a system both for keeping records and for training/indoctrinating new staff to keep records properly.
We have had two generations of office workers who have never experienced properly functioning filing systems (except possibly at a university library). Just like people who browse the web without an ad blocker, they don't know any different, so this is an invisible problem to them.
Johnny Decimal is a step towards opening people's eyes (or noses). Thanks for publishing your website!
1. Deleting things that should be deleted is an important skill, but another one is knowing what should not be deleted, but instead moved out of the main stream.
> The Johnny Decimal website is silent on the subject of stewardship. Here I want to focus on two aspects: keeping things working, and managing succession.
An error I intend to correct! On sufficiently large projects I believe you need a ‘librarian’ who has ownership of the system. I’ll expand on this on the site eventually.
Ah, if only I didn’t have an actual job. :-/
> We have had two generations of office workers who have never experienced properly functioning filing systems
This sentiment inspired the opening paragraphs of the site. We literally used to have people - women - called ‘administrators’. Their job would be to file things so that people could find them again. Then we all got a computer, the number of files exploded 100x, and we pretended like we could just search for what we needed.
Sorry if the comment came off as critical to what you've done; it was not meant to be. I just find the ability to tag data to be useful for organization and discoverability and I find it interesting that in 2020 there's no widely used file systems that have it out of the box.
I'd be more interested in a tagging file system for personal data like music, photos, documents etc. I orgainize my data in a hierarchical file system like most but I think tagging would be better for discoverability.
`find |grep -i ...` and `find -type f |xargs grep -i ...` have become old favourites.
This seems like an interesting system, provided you don't have more categories than you can easily remember.
If you use it in a team setting, there must not be more categories than the team as a whole can remember which is, perhaps counter-intuitively, fewer than you can probably remember on your own.
I usually put a few paths in between "find" and "-type", use the other predicates of "find" and often rewrite if from a just executed `find ... | grep`.
One particular modification is to extend the pipeline between the "find" and the "xargs" to filter the file list, especially `grep -v .git` ;-)
I could see this working better for a shared folder structure. I tried it for about six months with my personal documents and it went horribly. Between maintaining the index, constantly trying to decide where something should go, and running out of IDs once and having to rename a bunch of stuff, it was just a huge pain.
- It only solves the problem of communicating categories to another person/party with its brevity. 12.03, easy to tell others. But all this at the expense of the following things.
- There is no way to immediately know what 12.03 means. One has to carry a look up table. You can replace it with 2 char code. For example, Finance.Taxes to FN.TX. Why use numbers? Is the ordering important? Why not use char?
- What if categories do overlap? GMail solved this problem in 2005 with labels instead of folders. You can apply multiple tags. Tagging should be enum-like, while tagging it should show what tags already exist so that you dont create two seperate but similar tags, for e.g. Finance and Finances.
- The author dismisses search without giving any reason. Search is amazing for digital documents that is not possible in physical analog documents. The focus should be on tags + keywords or description of the document. Then use something like Algolia search to find stuff - it is extremely forgiving and powerful. Ofcourse, I am just speaking in general about data organization, not specifically about OS-level folders. For that, I think we're stuck with what the OS provides as a search engine. On MacOS, I think Alfred does a better job of searching, haven't used it personally though.
- You can still assign a unique ID to the document, after all thats all the 12.03 scheme does. You can communicate precisely to Jane, "Hey Jane, the document is B75AE2". Jane types that in the search engine and there is no need to weed through folders.
- Additional metadata such as year, author, owner, etc. would help with search.
Instead of making search powerful and contextualizing it, the author expends his arguments on frivolous pursuit of Johnny.Decimal.
It seems to me that the purpose of numbers over character codes is to constrain the available identifiers to force you to define a fairly small set of broad areas (you only have ten) and drill down within that. When we release that restriction and go with categories I think that would lead to making more and more categories with fuzzy overlap -- and that overlap is precisely what you don't want in a filing system where (unlike labels or tags) each item must have one and only one location.
You went on to mention labels and tags as a solution there -- but the point here is to implement the structure in a common filing system, isn't it? Your point works fine for email, but for documents on my drive tags don't seem like a great approach because at least on my machine even though tagging is an option, it's not robust and reliable.
It didn't feel totally fair to say that the author needed to address search more; the ability to search doesn't go away when you implement a system like this, it's still a powerful tool available to you if you know what you're searching for. If you don't know, then a system like this is a good way to explore the topics and assets in your file tree in a way that allows you to discover collections of related files.
Lastly, true that you can assign a unique ID -- but again, isn't that introducing and solving a different problem? Uniquely addressing every file might be a really useful thing, but as you mentioned, there's no context clue to help you decode B75AE2 -- you would need a lookup table. Even then, presumably you would want to know which categories and families that file should belong to.
Overall, everything you listed here is thought provoking and I think points out some of the limitations of the way our digital file systems emulate physical ones, but this collection of objections isn't persuasive that the Johnny decimal system isn't a good idea.
I'm not sure it even works there. I have emails both categorized and tagged and still can't find anything between the thousands of work emails I get a week from coworkers, mailing lists, and automated systems. Any search brings up dozens to thousands of results.
> but for documents on my drive tags don't seem like a great approach because at least on my machine even though tagging is an option, it's not robust and reliable.
Not to mention that most users aren't interested in tagging every file they have. I never tag files. My emails can get auto-tagged by mail rules, but I don't have time to add tags to all of my files.
> Instead of making search powerful and contextualizing it, the author expends his arguments on frivolous pursuit of Johnny.Decimal.
What do you mean? How do you expect the author to "make search powerful"? This system is clearly in the context of a desktop/NAS filesystem, so they're stuck with whatever search systems exist. I honestly don't plan to use this system, and still feel the need to respond to this critique of it.
> Why use numbers?
Is answered here[0], to intentionally preserve ordering. You don't need to know what a plain ID means, because the only reason to use it is in context.
> What if categories do overlap?
There's nothing to stop you from tagging things in addition to using this system, if you want.
> dismisses search without giving any reason
Again, the author is not building their own document storage system.
> Additional metadata
This system is trying to solve the problem of having to supply a bunch of information to find a specific document.
Overall, there are lots of anecdotes here about how this system is useful in some situations. There's really no reason to sound so incredulous about it, even if it's not for you.
Thanks for the color. I'll add one more thing against my argument:
- Discoverability: How many times have you opened glossary of a book just to explore and discover what else is in this book? Or open up a telephone directory with no specific person or business in mind, but a general category - "Shoe repair"? I think when things are organized by category, it allows for discoverability which would be difficult in a flat-searchable structure. Search works great when you know what you're looking for.
This can be half-solved by listing all items in a tag. Just that they'd appear in multiple tag archives as duplicates.
This magical search system doesn't exist. In reality we have dozens of different search backends searching different stuff in different places with different relevance and UIs and results. And if you're going to use labels then this is the same thing, except picking 1 major label to organize with.
It's why we have street numbers to find buildings. It's why libraries use catalogs to locate books. It's why financial systems use a chart of accounts with numbers and prefixes to organize transactions.
Location is important, and abstracting that away into a giant file share with search only makes things worse.
I've seriously considered making a file/folder organizer based on an old program called "animal" that I saw on some mainframe in the '80s.
Animal asked you to think of an animal, then it tried to identify it by asking you yes/no questions. Let's say it asked you if it has four legs (no), does it have wings (yes), and then it guesses that you are thinking of a hummingbird.
That's wrong, because you were thinking of a penguin. It would then ask you what yes/no question it could have asked to tell a penguin from a hummingbird, and you could tell it to ask "Does it fly?" and the answer is "no" for a penguin.
It would add that question to its decision tree, so if anyone else thought of a penguin it would get it.
For much of my stuff at work, I use an unfiled filing system. I started this one day when I was asked to write some one time report to solve some mystery.
I spent a half hour with the line "mkdir " typed into my terminal window unable to think of a good name for the directory to do this task in. Finally I gave up on naming the thing, and did "mkdir unfiled". Then in unfiled I did "mkdir 1", and "vi directory.txt". In "directory.txt" I added something like this:
"1 Report for Alice and Bob to identify source of unexpected load on main website"
and then went into the "1" directory and wrote the code to gather the data and analyze it and produce that report.
I intended to figure out a real name for the thing later and rename/move it.
I never got around to figuring out a real name, and unfiled/1 is still there. Along with unfiled/2, unfiled/3, ..., unfiled/314.
This has actually worked out well for me. Suppose I run into some kind of issue with UTF-8 handling in MySQL. I remember I've dealt with that before, and grep in directory.txt for terms like mysql and utf. That turns up that the work for those earlier problems is in unfiled/258, and so that's where I go to find the programs I wrote for investigating that kind of issue.
Not everything goes in unfiled, of course. A good rule of thumb is that if it is something my replacement would need to takeover if I got hit by a bus, it should not be in unfiled. It should be in a well named repo on my department's git server.
>When we kept everything on paper, organised people had these things called filing cabinets. They stored all of their documents in them in a structured way so that they could find them again.
>Now those same people store all of their files in arbitrarily named folders on their company’s shared drive and wonder why they can’t find anything.
This is the most outrageously false premise on which to base an argument. It's staggering.
Well, this was bound to happen one day. This is my site. I guess I’ll read the comments here then go take a bath with a bottle of gin. :-)
Edit: yep. As suspected.
Let me just say this, if I may:
1. You don’t have to like every idea, but please consider that some people do.
2. I’ve been using this successfully for a decade. I’m using it right now to organise a project. My boss loves me for it, because guess what? We can find things!
3. I’m not alone. I’ve done essentially nothing to promote this yet still I get multiple emails a week from strangers telling me that they love it and asking for guidance.
Okay, carry on. I’ll answer questions here where doing so feels productive.
I had a project I'm working on (which I already knew was going to be contentious) hit the front page a few months ago in a big way. After about 5 minutes I had to block the ID in my browser. It wasn't great.
Thanks for posting. If nothing else the good thing about your system is getting people to think about how to name and nest folders. I quite like one note in that it forces a 2 level heirarchy
The HN comments always sting a little. My guess is the cynics are the ones who usually comment while people who are more positive will probably be more likely to just upvote? So my recommendation (having had a couple HN-front pagers) would be to be encouraged by the upvotes and the attention, while keeping your skin thick.. But also learning from the criticism what you can improve.
Same here. I like the idea of JD very much - and I've been meaning to reorganize my notes by it, but can't ever get started on it - so there's very little of value I can add to conversation beyond an upvote.
On GP's:
> My guess is the cynics are the ones who usually comment while people who are more positive will probably be more likely to just upvote?
It irks me to see this comment pattern regularly being called out as "HN specific", because it isn't. It's universal in on-line discussions - particularly those on platforms with upvote/reaction buttons. There is nothing interesting in "I like it" comments, so you always get disproportionately more of critical comments ("I don't like it because", "I had a problem with it"), tangents, and low-quality humor. HN is perhaps different than most places in its strong dislike of low-effort jokes - which leaves us with just criticism and tangents. My point thus is: it's not HN, it's the Internet. It's not a mutual admiration society like you get when you go out with friends for a beer. I would think people would've gotten used to it in 2020.
>> My guess is the cynics are the ones who usually comment while people who are more positive will probably be more likely to just upvote?
> It irks me to see this comment pattern regularly being called out as "HN specific"
I've heard this general rule vocalized on social media as "Fans like, haters comment".
If you get loads of upvotes for an idea and a lot of flack, those pushing back is not your audience and can safely be ignored. You are not doing anyone any favors by trying to appease everyone.
Yeah, downvotes are weird in that way. Comment sections that marks posts and comments with more downvotes than upvotes as "low quality" also seem to encourage "pile-on" behavior.
This is a bit off-topic, but by any chance did you draw inspiration for the design of the site from Butterick's Practical Typography[0]? I _love_ the beautiful simplicity of this page.
Perhaps you could, rather than taking criticism personally, use these comments as an opportunity to get a broader understanding of how others perceive your work.
I imagine it's a minority view here (well, pretty much everywhere), but I relish getting constructive criticism of my works, whether it's on a site like this or a review by my peers/customers/employers.
That's because I have my own perspective, point of view and view of the world. Others have different ones.
When other folks share their opinions/perspective, it gives me an opportunity to examine my own preconceptions -- which usually winds up giving me a better understanding of the topic at hand.
Granted, in a setting like HN, there will be those who will be uncharitable or dismissive, and those voices can be ignored.
However, those who criticize specific stuff and provide the assumptions/reasoning/thought processes behind those criticisms can provide valuable insight into how you and/or your product/service are perceived outside of your small circle of shared views and assumptions.
I hope the bulk of the comments here are of the latter kind rather than the former, and that you get new perspectives that can drive positive growth for your activities.
Indeed -- and this is why, really, we all want our pet thing to be up here on the front page at some point. Feedback from this audience is truly invaluable...
I believe the trick is to remove the emotion from it. It can be very hard to not perceive criticism of the thing as criticism of oneself. Especially if the thing is something we've worked hard at and are proud of.
I use the "feedback bucket" mental trick: all feedback is shit (whether supportive or critical). All the feedback goes in the bucket, where it nourishes the growth of the beautiful roses of customer insight.
The point is to remove the immediacy of the feedback, detach it from the thing (and my feelings about the thing). But still allow it to inform my thinking about the thing. Instead of taking a criticism to heart, I can mentally "put it in the bucket". If it's useful, thoughtful, considered, constructive criticism then it will produce some insight later. If it's just spite, then it'll produce nothing.
I agree, except for the part where they immediately contradict themselves:
> If you choose to give me your email address, I will treat it with the utmost respect. I’ll never spam you, or give it to anyone else.
> In 2019 I moved from Mailchimp, whose business model started to make me uncomfortable, to Buttondown. It’s run by one guy who I trust to do the right thing.
Are you arguing that using a service provider to send email is "giving email adresses away"? If so, how do you send your email? (I guess you could host your own mail server on-premises, but is that the bar?)
Looks not valid in terms of GDPR. What are my rights and in which paragraph are those rights stated? How long is data stored? Which log files of my visit arise (just my IP and date of visit? Who hosts that site, if it's not a machine you physically own? Do they store that data? More data, like browser or OS version?)
You are usually allowed to use that data for debugging and intrusion detection etc. but you still need to mention it, if that data is collected, even by a third party you hired.
You could add a clause: "If your email address contains the string "a3w" it will be immediately forgotten." :)
I think if[1] I ever had a website, and if[2] I had advertising I would choose one advertiser per month, display a static image at the top of each page, and clicking on it would take you to a static page on an advertiser's website. and it would have a privacy policy like your good example
There's not much to change there to actually make it "more GDPR-like". It already does what GDPR wants to cause, i.e. you don't gather Personal Data, with the only exception being the email, and given that it's not operating as an organisation it's probably in the clear.
Maaaaaybeee you might want to strip last octet in IP number from logs, but that's pretty much it.
Complex GDPR policies are only necessary when you want to store and process Personal Data.
I've met your site via your comment on HN about a year ago, on the topic about organizing...something...that I can't remember now.
The idea intrigue me! It's not often you see kind of thing that seems mundane(to general interwebz population) like a system to organizing files and folder. There is some method floating around, but they either too basic or too complex for my need.
That said, back then I give Johnny Decimal a try a bit, but works took toll on me and I eventually give in to my old way of filing things. Maybe it's time to revisited. :)
I really liked your project. It may or may not be useful in my case, but I want to just say that you've put a lot of effort into this and it certainly shows. I have a problem with naming things as well, usually prefer some kind of a systemic way of organizing things. Perhaps, peculiarly I have a fascinationg for how companies name their products! It is similar in a way - organizing information for best possible internal and external communication.
Website is clear cut and to the point, its gorgeous btw. :-)
Thanks for your comments here. They made me see that this is meant for a pretty particular context.
My first reaction was creeping horror, because this feels like a solution to a problem people shouldn't have. But now that I read you talking about the situations you've made this for, I get your point: it's a problem people do have, and one that's not going away any time soon.
Just to save future bathub wear and tear, I'd suggest adding a bit to the top of the website to say who this is for. Just a couple sentences telling the story of your archetypal user would do it. You might also say who it probably isn't for, just so we can get to the dogs wearing clothes images sooner.
It does feel entertaining to read the comments from people who have so obviously never worked for a large (or old) organisation doing an admin related job who think that spotlight or fuzzy search will save them.
In a previous job I had to respond to freedom of information requests for a public sector organisation with more than 20,000 employees, which meant deep diving different departments shared drives and shared inboxes. The vast majority had no real structure at all and good luck finding a system that can search dozens of silo'd shared drives for info in .pdf and .doc files with permission systems that have evolved over decades. Discovering that the R&I division (which had several thousand employees on its own) mostly used johnny.decimal on their shared drives and often in their personal spaces was like discovering a tribe of ray gun wielding supermen in the depths of stone age Norfolk. It just makes life easy.
I idealize / fantasize about archiving and organizing work sometimes, not going to lie. I think a lot of people do as well for their own notes, but if they're anything like me, they don't have nearly as much stuff, or they intentionally produce and hoard stuff for the satisfaction, not because they actually need it.
There being new filing / organizing systems and apps (zettelkasten is the new hip thing nowadays it seems) popping up on HN every once in a while isn't helping either.
Tl;dr, I think it's the idealization of organizing, the feeling of productivity, etc that is fueling a lot of people.
Meanwhile I've gone through three or four different note keeping / taking / managing apps and still haven't settled on anything, lol. I also lost the need for it, now that I've gone from the exploratory phase to actually building software.
Sometimes unfortunately you have no idea of the keywords to search for...
... or you drown under the ocean of unrelated finds, which is my usual experience with mass-applied windows indexing (in the form of massive corporate sharepoint)
Ediscovery software, and I used a bespoke solution in that job, takes a job that can take weeks or months the and makes it possible to do it in days. Or y'know, you could use a filing system and do it in minutes?
Love this idea, and I want to adopt it for our organisation.
Could you advise what you would do in the following situation though?
I imagine one folder called "1 Projects" and immediately stumble upon an issue. We have 100's projects, not 10's.
Do I introduce another decimal point? Like so: "1.123.0"
My top-priority job as soon as I get time over Xmas is to get (an updated, better) version of this up on the main site. It’s the #1 question I get asked.
Took me a few years to figure this one out but I’ve been running it in anger now for a year and I can say that it works.
Re: Dewey, no. You make your own category numbers up for each system, every one is different (unless you create a standard, e.g. if you do lots of similar projects).
How have I never seen that link?! Yeah the issues described therein sound familiar...
Really great site! Uncluttered and easy to follow.
I have one question: is there JD best practice on how to handle file sets that are generated frequently and in large-ish numbers? I’m thinking specifically of camera image files. My current approach is to have a directory per year, and then subdirectories (with a YYYMMDD-subject name) for each distinct shoot (typically between 30 and 50) in that year, together containing between 5000 and 20000 new images (depending on how many shoots I do). I maintain a parallel folder structure for the processed images. My current thinking is to keep this photo-library as-is, but put all of my ‘normal’ files into a JD structure.
By the way, and picking up memetherapy’s comment, I’ve also experienced knowledge management problems in large corporations for which JD would be a great answer. The last place I worked moved several years ago from a ‘classic’ system based on servers and file systems to a system based upon numerous per-project sharepoint sites. It didn’t help that the template for the sites was developed from a project lifecycle perspective and had about 70 folders covering every stage from initiating a bid to project shutdown. There was pretty-much a single ‘dev’ folder for the actual work, that techies like myself inevitably populated in inconsistent ways. Something that was never resolved was that all of these silos made searching for old reports, etc really really hard and fundamentally broke knowledge management in the company. Especially when most of the silos had per-project access permissions.
On the point made elsewhere about tagging, in my experience it is extremely hard to get people across a large company to add metadata to content, let alone do it consistently.
> My current thinking is to keep this photo-library as-is, but put all of my ‘normal’ files into a JD structure.
Yes. This sort of data set already sorts itself pretty well: by date. There’s no need to mess with that.
See also: your music library, if you don’t have iTunes do it for you.
> It didn’t help that the template for the sites was developed from a project lifecycle perspective and had about 70 folders covering every stage from initiating a bid to project shutdown.
Aah, when the PMO tries to be helpful. I see this all the time - a ‘baseline’ set of folders that you’re meant to understand know what to do with but literally nobody does so they just get ignored.
Worse, I frequently see them with numbers at the beginning. Worse even, those numbers just make no sense at all. I’ve recently experienced a repeating pattern which involves folders 001-015 being something that I can’t remember, and then `016 TO BE DELETED` (because this is in a document management system where you can’t actually delete a file).
Hi, I'm currently reading the site/guide, my question is, how viable is this for organizing personal files? Is this strictly for business only? I'm currently doing a personal files reorg and this hackernews post came in perfect timing.
Also on my to-do list: get a few example structures up there. But drop me a mail and I’ll send you my home structure for inspiration. hello <at> the domain.
Elsewhere someone suggested going 36-bit. Or hex, or whatever. So after 99 you’d go A0 and so on.
I don’t love that — the neatness gene is making me itch — but it’d work.
I’d say perhaps you’re being too granular if you have 100+ of a thing in a category. I’d love to know what it is, if you want to mail me. All this info adds to the system.
I had a go at this today and was instantly reminded of why all my past organization efforts fell flat on their face: I don't know how to categorize things, and I can't seem to figure it out.
If there is an obvious delineation, I will accept it and use it. If there is a line separating categories and that line is blurry at all, I find it impossible to both confidently choose where it goes and know where I filed it later on. I can know where to put something or I can know where to pull it from. Never both, somehow. (I am as confused by this as you are, but it is true for me.)
So everything winds up in piles, both physically in real life and in my filesystems. The only way I can remember where something is, is to remember what the pile looked like when I placed that thing on the pile. Then I know exactly where it is.
In network file folders I can find something by remembering my own history in that folder and remembering about how many items were there when I created the thing I'm looking for. Then I sort by creation time and I count that many down, and it's usually one or two away from that number, in one direction or the other.
My windows desktop is organized autobiographically as well. I can never remember what I named a folder, but I know exactly where it is on the desktop.
I'm basically a nightmare for anyone that needs a tidy environment. If you clean up my stuff, everything you organized is lost to me forever.
Thanks for the great site, it's a great reference for new clients when setting up there new file servers etc.
I'm currently doing a Zettlekasten (note/idea organisation) which uses a _similar_ principle, I think between this and johnny.decimal you could build a very custom but efficient system. Will have to ponder further :)
Cheers
Thank you for this! I have been working at the same company for 8+ years, and I have accumulated a lot of stuff over the years. I have trouble finding anything these days. The stuff I /can/ find is because I have multiple copies sprinkled in different directories.
After reading this post, I have started to organize my documents and projects according to the description. I feel like this is going to be helpful not only so that I can find what I'm looking for, but so that I don't have 4 copies of every giant source code repo taking up my disk space.
I do have one question: in the examples, the category shows "40-49 BlahBlah", but the subcategory goes "41 Taxes, 42 Expenses, etc." Is there a reason why X0 is skipped? Is this so you have room for something you might have forgotten?
> In the example we’ve been using, 20-29 was Administration but our first category was 21 Company registration. What happened to 20?
> We reserve these numbers for ‘management & meta’ information about the category. Although I hardly ever use them, it’s rare that I have more than nine categories in an area. I don’t miss the wasted number, but I always appreciate it when I need it.
Also recently I’ve been starting my IDs at `.11`. Same idea — sometimes you really want stuff to always appear ‘at the front’ and this lets you achieve that.
In the project system I’m setting up now, for example, we’re using the PMP methodology [0]. This very neatly divides up in to:
21 Scope
22 Schedule
23 Cost
...
29 Stakeholder
For each of these, you have a [x] Management Plan. They always sit at `2x.01`. Other mandatory artefacts will live in the zeros.
Then I won’t fill up the zeros, I’ll skip straight to .11 in each category for the things that are ‘personal’ to this particular project. I’m expecting that they will not be standard across projects.
and I'll add, It's not just fast, its incredibly beautiful and a joy to browse. Thankyou :) Its added to my list of inspiration for beautiful simple websites.
I'd love to know if you have any tips on your images like were they preprocessed or scaled down? They're really well done and still high quality despite loading quick?
Blood sweat and tears? I tried really really hard to get them as small as possible while still being original-sharp.
It’s a real PITA to be honest. I’m typing this from bed and can’t remember the exact details but chuck me an email and I’ll have a look over the weekend. (Or just scan the repo if you can be bothered, but I’m happy to try to remember what I did and put it in a mail.)
Okay so the pain I had was taking retina screenshots and having them show up in retina resolution.
The site is written in MDX and then `gatsby-plugin-mdx` does the heavy lifting. What I could not get to work was Markdown formatting for inline images, especially not for smaller images. They always ended up being shown at 2x.
So now it's just an image, imported in to the file at the top, and displayed further down.
import outlook_folder_structure_1232_expanded from "./Outlook__folder-structure-12-32-expanded-min.png"
And then just a straight `<img src={outlook_folder... }` where it needs to appear.
As for making them sharp and small, they're just screenshots from my Mac run through one of the many online PNG resizers. From memory that gave me a file with a 78% saving and if it wasn't near magic then I couldn't tell you how it did it because there was zero difference to my eye. I can't remember which one I used, just try a few.
190 comments
[ 0.19 ms ] story [ 315 ms ] threadhttps://en.wikipedia.org/wiki/List_of_Dewey_Decimal_classes
You probably don't know much about Dewey, he's got a pretty "interesting" life story:
https://en.wikipedia.org/wiki/Melvil_Dewey
"At the June 2019 conference of the American Library Association the Council voted to remove Dewey's name from its top honor, the Melvil Dewey Medal; the resolution cited Dewey's history of racism, anti-Semitism, and sexual harassment. The resolution was passed overwhelmingly with no debate. The award was renamed the ALA Medal of Excellence at the Association's January 2020 conference"
Yep, a friend of mine named it in 'honour' of Dewey Decimal. Kinda genius, which I can say because that wasn't my idea. :-)
But yeah had no idea he was a terrible human, that's a shame.
> He resigned from the American Library Association due to allegations of sexual harassment, racism, and antisemitism in 1905.
Sharepoint should have been able to fix all that, but somehow it just made it the same, but slower. Even after 20 years, I never know where the f__k I am in sharepoint, or when someone sends me a link to a sharepoint doc, I can open the doc, but I don't know where the f__k the document _is_. I dont know quite why sharepoint is so confusing, I think its because it tries hard to pretend to be something other than the document store that it should so obviously be. Just when I get the hang of it, they redesign it all, and add some new front end like Delve.
If I join a project, something that rings alarm bells is if the Project Manager has created sixty empty folders in Sharepoint arranged into a three level hierarchy with different folders for every stage and stream of the project. Most of the time, most of the folders remain empty for the duration.
So yep I really like this Johnny.Decimal idea, simple and workable.
Seems to be no different than this:
>something that rings alarm bells is if the Project Manager has created sixty empty folders in Sharepoint arranged into a three level hierarchy
Except that it's a hundred empty folders in a two level hierarchy replicated by convention across different systems, including Sharepoint if your organization uses it.
This. One hundred times this.
Those here arguing that things can, in fact, be found, or saying that my claim that "nobody can find anything any more" is absurd, clearly haven't worked in the ~5 large enterprises that I've worked in over the last decade.
Internal organisation in enterprises is shockingly bad. Remember, these people aren't techs. They have no idea what `find / -iname file | grep whatever` means. They crack open SharePoint and rummage around. When they want to save a thing, they just put it wherever. It's chaos.
I'm just implementing this for a few new projects at my current place and my boss has messaged me a number of times expressing his joy at just knowing where stuff is.
Or, you know, don't encumber the brains of people with useless numbers, and use computers for what they are actually good are, at least better than people, e.g. use the search feature of your file browser for "payroll schedule".
It's a common issue with intranets. There's lots of content, but not enough traffic to figure out what you may be interested in. And even if there was, you may not be after the "most popular" item.
This is a real example from a past life, and really works when you're shouting across an office or talking down the phone.
The longer original, edited down for brevity, went something like:
"Hey Kristy, where can I find the payroll schedule?"
"Okay go to the Finance folder. No, the one in the G: drive. Yeah now go to Payroll ... oh sorry, it's Payroll 2020. Yeah no not the Payroll 2020 in the Payroll folder, the one at the root. Yeah. Now in there ... hang on let me have a look ... okay there's a 'Janice' folder and in there..." ... and so on, ad. absurdium.
And search systems today are still terrible. Different interfaces, searching different locations and content, with different ranking and relevance and operators. Much faster to just direct someone where something is and let their brain figure out the rest, because we're actually very good at that.
This scheme is the best I’ve found (some tagging software used it): `report[important,sales,veronica].pdf`.
I search "driveway" and get all the photos my wife and I ever took that shows our driveway. Ah yes the cracks are getting worse.
I search email for "driveway repair" and find that old recommendation from a friend despite the word driveway having never been mentioned.
I don't need a sorting system. I need a big pile of all my data ever with a fantastic search engine.
I'd prefer if it wasn't google.
The problem is when search extends across domains. I can search files with Spotlight, but not my google emails. On my phone, it's worse because there are so many different comms systems. I can search my Whatsapp messages; my SMS messages; my email; Facebook messenger; Skype.... Each individual search is fine, but no one tool does them all. Hence, the idea of using manual labels to classify stuff - including emails - can still be useful. Crazy that we're forced back to this.
Edit: Seva and Cloudtenna
I wish I had a local one that just packed it all into Solr
Yep, search is great in Google-world or on your Mac. But try it at work where everything is in disparate SharePoints if you're lucky.
> Hence, the idea of using manual labels to classify stuff - including emails - can still be useful. Crazy that we're forced back to this.
I know!
These days I hardly ever bring up file naming in code reviews—fzf just makes it so easy to find everything.
https://github.com/junegunn/fzf
You can hit CTRL+t from the terminal after installing FZF and it will let you fuzzy search for any file you have. So you can do something like run `vim [CTRL+t]` to quickly find and open any file you want, or `cd [CTRL+t]` to switch into some deeply nested directory.
Even inside of WSL 2 with a 6 year old workstation on a first generation SSD it takes ~3 seconds to index 190,000 files and once that index is built, narrowing down results while typing is close to instant when using ripgrep as FZF's search back-end.
The amazing thing about all of this is it's dynamic. There's no having to create aliases or pre-defined directory structures. I only discovered CTRL+t from FZF a few weeks ago but once I did it was almost as good as discovering CTRL+r for the first time.
This is my gotcha: how do you decide where a document go? The categories might be clearly split, but documents typically won't fall in exactly one category.
How do you manage temporary tags? That is: how do you operate on the documents? You might be having some documents for your "2020 tax return" in several categories of your system. And what now? Do you duplicate them in another category (tax return?)? Do you tag them? How do you keep track that you moved them? (In case they're real objects, like pieces of paper).
Symlinks!
> How do you manage temporary tags?
Symlinks again!
Basically you put everything in an incoming folder, maybe organized by year or month/year or even type. Then you symlink to all of the categories it would be interesting to you.
Your temporary tags would just all be a folder of symlinks, so you can kill the tag and retain your document.
Symlinks can work, and I've used them.
My advice to people is usually, though, that your brain will sort it out for you. And it really does.
The canonical example is "home insurance". Say you have a category of "home" and a category elsewhere of "insurance". Perfectly possible.
Which one does it go in? Well, you designed this system. Your brain came up with it. I've found that in nearly all cases, you just know.
I see where this may fall down in a shared environment (work), which is why I advocate (in one of the many unwritten posts on my as-yet-to-exist blog) for the role of a 'librarian' at work. We seem to think that 20 people can work on a project together and all just magically decide where everything goes. That has been shown again and again to be a fallacy that we believe search will help us with. My argument -- from experience -- is that it does not, and that we need to go back to a world where we organise things, and that people are responsible for this organisation.
When I go through old documents, I don’t even skim through the contents. I just check the date and remove them if necessary.
A whole book on this is "Delete: The Virtue of Forgetting in the Digital Age" by Viktor Mayer-Schönberger (2011).
A longer piece with the author in The Guardian: https://www.theguardian.com/technology/2011/jun/30/remember-...
And the book: https://www.jstor.org/stable/j.ctt7t09g
Exactly unlike actual documents. Where do I put the sales figures we use for marketing?
This is what basically kills the whole idea. It's what made people want to switch from folders to search, which this site claims doesn't work.
Yep, this can be an issue. See my sibling comment here:
https://news.ycombinator.com/item?id=25412455
so something like:
That said, the only documents I need to keep in long-term storage are invoices and contracts, which I can just put into two directories. If I ever get too many of such official documents, I might look into a structure like this.
What I can say is that I like it for helping you choose levels of hierarchy. So far I've squeezed 10 or fewer sub-categories into each chosen category. (I use the 0s, so within 10-19, I start with a sub-folder of 10...)
What I do not do is any of the other suggestions - searching by it, memorizing numbers (though I have an intuitive sense now when organizing email or looking through it.)
I have not tried this folder structure yet for files.
[0] https://github.com/slowernews/hamster-system
Oh yeah: Yahoo directory vs. Google.
I've given up on organizing documents in hierarchies a long time ago. Specifically because many things actually belong in many places.
Search is the way to go.
[edit]: and by "Search" and mean "indexed search"
I don't find tags useful for many of the same reasons that I invented this system.
What's the tag for the payroll schedule? Is it 'Payroll'? 'Finance'? Something else?
Who manages the tags? Can anyone add one?
Tags need managing just as much as -- if not more than -- a well-organised folder structure. Tags aren't just some free magical way to organise stuff ... otherwise I dare say they'd be more popular.
"What is the tag?" is a category error. It's "what are the tags?" Tags are not paths.
Is it Payroll? Yes. Is it Finance? Yes, if that's how your org does things. Is it other things besides? Yes. A "superseded" tag would likely be useful to many organisations.
The idea of tags as opposed to paths is that a leaf object can have lots of them.
Of course tags are not free. The question is, does the value they bring outweigh their costs? Like you, I think the answer is usually "no". People suck at tagging.
If tags are to be used, "who manages the tags" is a key question that needs to be explicitly asked and answered at the same time as, or before, deciding your categories.
However the same applies to managing the "chart of accounts" that Johnny Decimal requires. The structure that is best for the organisation almost certainly won't be optimal for any given staff member. So management is required.
These are both stewardship problems.
The Johnny Decimal website is silent on the subject of stewardship. Here I want to focus on two aspects: keeping things working, and managing succession.
Keeping things working: using a gardening metaphor, is planting and weeding, re-potting, trimming and pruning and harvesting, turning over the garden beds, and so on. Like gardening, this part of records stewardship is best carried out regularly and frequently.
Obsoleting things is a key part of this. Contra the poster above who says "just delete it"[1], often you have to keep no-longer-current versions of filed objects around for a long time, for some combination of reasons: the maintenance and repair manual for a machine tool model we stopped selling 20 years ago, minutes of board meetings from an equivalent time ago, commemorative photos, old logo artwork, client records (if, say, you're a dentist) for the life of the client, even if you haven't seen them for a decade, ...
Succession is about ensuring groups have the right access, new people belong to the right groups and are properly indoctrinated, and things are handed over when staff leave or change roles, and ex-staff have their access removed; and about revising the structure, ontology, and access rights so that they stay relevant as the organisation they serve changes.
Records management is somewhat like plumbing--good drains, no-one notices; with bad ones, everyone is miserable. And keeping the drains "good" requires work.
The problems that are now universal ("Acme Corp Contract", "Acme Corp Contract - final version", "Acme Corp Contract - final with client revisions", "Acme Corp Contract - senior management approval", "Acme Corp Contract Brian", etc.) have indeed come about because computer-illiterate and filing-illiterate office workers thought the FAT file system (and then NTFS and/or Sharepoint) was a magic, zero-effort substitute for having a system both for keeping records and for training/indoctrinating new staff to keep records properly.
We have had two generations of office workers who have never experienced properly functioning filing systems (except possibly at a university library). Just like people who browse the web without an ad blocker, they don't know any different, so this is an invisible problem to them.
Johnny Decimal is a step towards opening people's eyes (or noses). Thanks for publishing your website!
1. Deleting things that should be deleted is an important skill, but another one is knowing what should not be deleted, but instead moved out of the main stream.
An error I intend to correct! On sufficiently large projects I believe you need a ‘librarian’ who has ownership of the system. I’ll expand on this on the site eventually.
Ah, if only I didn’t have an actual job. :-/
> We have had two generations of office workers who have never experienced properly functioning filing systems
This sentiment inspired the opening paragraphs of the site. We literally used to have people - women - called ‘administrators’. Their job would be to file things so that people could find them again. Then we all got a computer, the number of files exploded 100x, and we pretended like we could just search for what we needed.
* over 5,000 entries * over 3,000 tags
And I still have issues finding posts I've known I made via the tags.
[1] https://github.com/spc476/mod_blog
This seems like an interesting system, provided you don't have more categories than you can easily remember.
If you use it in a team setting, there must not be more categories than the team as a whole can remember which is, perhaps counter-intuitively, fewer than you can probably remember on your own.
As with all classification systems, one should beware the Celestial Emporium of Benevolent Knowledge: https://en.wikipedia.org/wiki/Celestial_Emporium_of_Benevole... or be doomed to repeat it.
One particular modification is to extend the pipeline between the "find" and the "xargs" to filter the file list, especially `grep -v .git` ;-)
Now I just name things verbosely and use Everything (https://www.voidtools.com/support/everything/).
- It only solves the problem of communicating categories to another person/party with its brevity. 12.03, easy to tell others. But all this at the expense of the following things.
- There is no way to immediately know what 12.03 means. One has to carry a look up table. You can replace it with 2 char code. For example, Finance.Taxes to FN.TX. Why use numbers? Is the ordering important? Why not use char?
- What if categories do overlap? GMail solved this problem in 2005 with labels instead of folders. You can apply multiple tags. Tagging should be enum-like, while tagging it should show what tags already exist so that you dont create two seperate but similar tags, for e.g. Finance and Finances.
- The author dismisses search without giving any reason. Search is amazing for digital documents that is not possible in physical analog documents. The focus should be on tags + keywords or description of the document. Then use something like Algolia search to find stuff - it is extremely forgiving and powerful. Ofcourse, I am just speaking in general about data organization, not specifically about OS-level folders. For that, I think we're stuck with what the OS provides as a search engine. On MacOS, I think Alfred does a better job of searching, haven't used it personally though.
- You can still assign a unique ID to the document, after all thats all the 12.03 scheme does. You can communicate precisely to Jane, "Hey Jane, the document is B75AE2". Jane types that in the search engine and there is no need to weed through folders.
- Additional metadata such as year, author, owner, etc. would help with search.
Instead of making search powerful and contextualizing it, the author expends his arguments on frivolous pursuit of Johnny.Decimal.
You went on to mention labels and tags as a solution there -- but the point here is to implement the structure in a common filing system, isn't it? Your point works fine for email, but for documents on my drive tags don't seem like a great approach because at least on my machine even though tagging is an option, it's not robust and reliable.
It didn't feel totally fair to say that the author needed to address search more; the ability to search doesn't go away when you implement a system like this, it's still a powerful tool available to you if you know what you're searching for. If you don't know, then a system like this is a good way to explore the topics and assets in your file tree in a way that allows you to discover collections of related files.
Lastly, true that you can assign a unique ID -- but again, isn't that introducing and solving a different problem? Uniquely addressing every file might be a really useful thing, but as you mentioned, there's no context clue to help you decode B75AE2 -- you would need a lookup table. Even then, presumably you would want to know which categories and families that file should belong to.
Overall, everything you listed here is thought provoking and I think points out some of the limitations of the way our digital file systems emulate physical ones, but this collection of objections isn't persuasive that the Johnny decimal system isn't a good idea.
I'm not sure it even works there. I have emails both categorized and tagged and still can't find anything between the thousands of work emails I get a week from coworkers, mailing lists, and automated systems. Any search brings up dozens to thousands of results.
> but for documents on my drive tags don't seem like a great approach because at least on my machine even though tagging is an option, it's not robust and reliable.
Not to mention that most users aren't interested in tagging every file they have. I never tag files. My emails can get auto-tagged by mail rules, but I don't have time to add tags to all of my files.
What do you mean? How do you expect the author to "make search powerful"? This system is clearly in the context of a desktop/NAS filesystem, so they're stuck with whatever search systems exist. I honestly don't plan to use this system, and still feel the need to respond to this critique of it.
> Why use numbers?
Is answered here[0], to intentionally preserve ordering. You don't need to know what a plain ID means, because the only reason to use it is in context.
> What if categories do overlap?
There's nothing to stop you from tagging things in addition to using this system, if you want.
> dismisses search without giving any reason
Again, the author is not building their own document storage system.
> Additional metadata
This system is trying to solve the problem of having to supply a bunch of information to find a specific document.
Overall, there are lots of anecdotes here about how this system is useful in some situations. There's really no reason to sound so incredulous about it, even if it's not for you.
[0]: https://johnnydecimal.com/concepts/areas-categories/
- Discoverability: How many times have you opened glossary of a book just to explore and discover what else is in this book? Or open up a telephone directory with no specific person or business in mind, but a general category - "Shoe repair"? I think when things are organized by category, it allows for discoverability which would be difficult in a flat-searchable structure. Search works great when you know what you're looking for.
This can be half-solved by listing all items in a tag. Just that they'd appear in multiple tag archives as duplicates.
I've tried this (in the context of multiple projects [0]). Believe me, I've tried everything!
The reason the JD site is the way it is is that I've been doing this myself for a decade and this is what I consider to be The Best Way.
[0]: https://play.johnnydecimal.com/johnny.decimal/projects
It's why we have street numbers to find buildings. It's why libraries use catalogs to locate books. It's why financial systems use a chart of accounts with numbers and prefixes to organize transactions.
Location is important, and abstracting that away into a giant file share with search only makes things worse.
Animal asked you to think of an animal, then it tried to identify it by asking you yes/no questions. Let's say it asked you if it has four legs (no), does it have wings (yes), and then it guesses that you are thinking of a hummingbird.
That's wrong, because you were thinking of a penguin. It would then ask you what yes/no question it could have asked to tell a penguin from a hummingbird, and you could tell it to ask "Does it fly?" and the answer is "no" for a penguin.
It would add that question to its decision tree, so if anyone else thought of a penguin it would get it.
For much of my stuff at work, I use an unfiled filing system. I started this one day when I was asked to write some one time report to solve some mystery.
I spent a half hour with the line "mkdir " typed into my terminal window unable to think of a good name for the directory to do this task in. Finally I gave up on naming the thing, and did "mkdir unfiled". Then in unfiled I did "mkdir 1", and "vi directory.txt". In "directory.txt" I added something like this:
"1 Report for Alice and Bob to identify source of unexpected load on main website"
and then went into the "1" directory and wrote the code to gather the data and analyze it and produce that report.
I intended to figure out a real name for the thing later and rename/move it.
I never got around to figuring out a real name, and unfiled/1 is still there. Along with unfiled/2, unfiled/3, ..., unfiled/314.
This has actually worked out well for me. Suppose I run into some kind of issue with UTF-8 handling in MySQL. I remember I've dealt with that before, and grep in directory.txt for terms like mysql and utf. That turns up that the work for those earlier problems is in unfiled/258, and so that's where I go to find the programs I wrote for investigating that kind of issue.
Not everything goes in unfiled, of course. A good rule of thumb is that if it is something my replacement would need to takeover if I got hit by a bus, it should not be in unfiled. It should be in a well named repo on my department's git server.
>Now those same people store all of their files in arbitrarily named folders on their company’s shared drive and wonder why they can’t find anything.
This is the most outrageously false premise on which to base an argument. It's staggering.
Edit: yep. As suspected.
Let me just say this, if I may:
1. You don’t have to like every idea, but please consider that some people do.
2. I’ve been using this successfully for a decade. I’m using it right now to organise a project. My boss loves me for it, because guess what? We can find things!
3. I’m not alone. I’ve done essentially nothing to promote this yet still I get multiple emails a week from strangers telling me that they love it and asking for guidance.
Okay, carry on. I’ll answer questions here where doing so feels productive.
if on the other hand i see something that's problematic or could be improved, i should say that, instead of just downvoting.
i think the problem comes from the tone people use to voice their disagreements
On GP's:
> My guess is the cynics are the ones who usually comment while people who are more positive will probably be more likely to just upvote?
It irks me to see this comment pattern regularly being called out as "HN specific", because it isn't. It's universal in on-line discussions - particularly those on platforms with upvote/reaction buttons. There is nothing interesting in "I like it" comments, so you always get disproportionately more of critical comments ("I don't like it because", "I had a problem with it"), tangents, and low-quality humor. HN is perhaps different than most places in its strong dislike of low-effort jokes - which leaves us with just criticism and tangents. My point thus is: it's not HN, it's the Internet. It's not a mutual admiration society like you get when you go out with friends for a beer. I would think people would've gotten used to it in 2020.
> It irks me to see this comment pattern regularly being called out as "HN specific"
I've heard this general rule vocalized on social media as "Fans like, haters comment".
If you get loads of upvotes for an idea and a lot of flack, those pushing back is not your audience and can safely be ignored. You are not doing anyone any favors by trying to appease everyone.
[0]: https://practicaltypography.com/
I’ll say that Tailwind CSS is a godsend for someone like me. Being able to semantically describe what a thing should look like, inline, is miraculous.
Edit: /s/tailscale/tailwind. Tailscale is also amazing.
I imagine it's a minority view here (well, pretty much everywhere), but I relish getting constructive criticism of my works, whether it's on a site like this or a review by my peers/customers/employers.
That's because I have my own perspective, point of view and view of the world. Others have different ones.
When other folks share their opinions/perspective, it gives me an opportunity to examine my own preconceptions -- which usually winds up giving me a better understanding of the topic at hand.
Granted, in a setting like HN, there will be those who will be uncharitable or dismissive, and those voices can be ignored.
However, those who criticize specific stuff and provide the assumptions/reasoning/thought processes behind those criticisms can provide valuable insight into how you and/or your product/service are perceived outside of your small circle of shared views and assumptions.
I hope the bulk of the comments here are of the latter kind rather than the former, and that you get new perspectives that can drive positive growth for your activities.
...sometimes. ;-)
I use the "feedback bucket" mental trick: all feedback is shit (whether supportive or critical). All the feedback goes in the bucket, where it nourishes the growth of the beautiful roses of customer insight.
The point is to remove the immediacy of the feedback, detach it from the thing (and my feelings about the thing). But still allow it to inform my thinking about the thing. Instead of taking a criticism to heart, I can mentally "put it in the bucket". If it's useful, thoughtful, considered, constructive criticism then it will produce some insight later. If it's just spite, then it'll produce nothing.
https://johnnydecimal.com/privacy/
> If you choose to give me your email address, I will treat it with the utmost respect. I’ll never spam you, or give it to anyone else.
> In 2019 I moved from Mailchimp, whose business model started to make me uncomfortable, to Buttondown. It’s run by one guy who I trust to do the right thing.
If that privacy policy doesn’t make it screamingly obvious that I’m not tracking you in any way then holy moly I don’t know what to do.
I think if[1] I ever had a website, and if[2] I had advertising I would choose one advertiser per month, display a static image at the top of each page, and clicking on it would take you to a static page on an advertiser's website. and it would have a privacy policy like your good example
[1] big if
[2] bigger if
Maaaaaybeee you might want to strip last octet in IP number from logs, but that's pretty much it.
Complex GDPR policies are only necessary when you want to store and process Personal Data.
It’s hosted on Netlify, so I don’t even have logs. :-)
GDPR compliance is overblown, when the simplest way to be compliant is to "just don't store nor process personal data"
The idea intrigue me! It's not often you see kind of thing that seems mundane(to general interwebz population) like a system to organizing files and folder. There is some method floating around, but they either too basic or too complex for my need.
That said, back then I give Johnny Decimal a try a bit, but works took toll on me and I eventually give in to my old way of filing things. Maybe it's time to revisited. :)
Website is clear cut and to the point, its gorgeous btw. :-)
My first reaction was creeping horror, because this feels like a solution to a problem people shouldn't have. But now that I read you talking about the situations you've made this for, I get your point: it's a problem people do have, and one that's not going away any time soon.
Just to save future bathub wear and tear, I'd suggest adding a bit to the top of the website to say who this is for. Just a couple sentences telling the story of your archetypal user would do it. You might also say who it probably isn't for, just so we can get to the dogs wearing clothes images sooner.
“It isn’t for everyone” is absolutely right, and if I can make that separation higher up then it’s better for everyone.
https://news.ycombinator.com/item?id=25415022
There being new filing / organizing systems and apps (zettelkasten is the new hip thing nowadays it seems) popping up on HN every once in a while isn't helping either.
Tl;dr, I think it's the idealization of organizing, the feeling of productivity, etc that is fueling a lot of people.
Meanwhile I've gone through three or four different note keeping / taking / managing apps and still haven't settled on anything, lol. I also lost the need for it, now that I've gone from the exploratory phase to actually building software.
https://www.google.com/search?q=ediscovery+software
Not suggesting it though :P
... or you drown under the ocean of unrelated finds, which is my usual experience with mass-applied windows indexing (in the form of massive corporate sharepoint)
Do I introduce another decimal point? Like so: "1.123.0"
My top-priority job as soon as I get time over Xmas is to get (an updated, better) version of this up on the main site. It’s the #1 question I get asked.
Took me a few years to figure this one out but I’ve been running it in anger now for a year and I can say that it works.
Second, have you experienced any of the problems outlined in MetaCrap https://people.well.com/user/doctorow/metacrap.htm
How have I never seen that link?! Yeah the issues described therein sound familiar...
I have one question: is there JD best practice on how to handle file sets that are generated frequently and in large-ish numbers? I’m thinking specifically of camera image files. My current approach is to have a directory per year, and then subdirectories (with a YYYMMDD-subject name) for each distinct shoot (typically between 30 and 50) in that year, together containing between 5000 and 20000 new images (depending on how many shoots I do). I maintain a parallel folder structure for the processed images. My current thinking is to keep this photo-library as-is, but put all of my ‘normal’ files into a JD structure.
By the way, and picking up memetherapy’s comment, I’ve also experienced knowledge management problems in large corporations for which JD would be a great answer. The last place I worked moved several years ago from a ‘classic’ system based on servers and file systems to a system based upon numerous per-project sharepoint sites. It didn’t help that the template for the sites was developed from a project lifecycle perspective and had about 70 folders covering every stage from initiating a bid to project shutdown. There was pretty-much a single ‘dev’ folder for the actual work, that techies like myself inevitably populated in inconsistent ways. Something that was never resolved was that all of these silos made searching for old reports, etc really really hard and fundamentally broke knowledge management in the company. Especially when most of the silos had per-project access permissions.
On the point made elsewhere about tagging, in my experience it is extremely hard to get people across a large company to add metadata to content, let alone do it consistently.
Yes. This sort of data set already sorts itself pretty well: by date. There’s no need to mess with that.
See also: your music library, if you don’t have iTunes do it for you.
> It didn’t help that the template for the sites was developed from a project lifecycle perspective and had about 70 folders covering every stage from initiating a bid to project shutdown.
Aah, when the PMO tries to be helpful. I see this all the time - a ‘baseline’ set of folders that you’re meant to understand know what to do with but literally nobody does so they just get ignored.
Worse, I frequently see them with numbers at the beginning. Worse even, those numbers just make no sense at all. I’ve recently experienced a repeating pattern which involves folders 001-015 being something that I can’t remember, and then `016 TO BE DELETED` (because this is in a document management system where you can’t actually delete a file).
Given how the system is setup, I find this amusing in a meta-way.
Also on my to-do list: get a few example structures up there. But drop me a mail and I’ll send you my home structure for inspiration. hello <at> the domain.
I'm gonna organize my 2TB thumb drive this way tomorrow (I'll at least start) and see how it goes...
Elsewhere someone suggested going 36-bit. Or hex, or whatever. So after 99 you’d go A0 and so on.
I don’t love that — the neatness gene is making me itch — but it’d work.
I’d say perhaps you’re being too granular if you have 100+ of a thing in a category. I’d love to know what it is, if you want to mail me. All this info adds to the system.
If there is an obvious delineation, I will accept it and use it. If there is a line separating categories and that line is blurry at all, I find it impossible to both confidently choose where it goes and know where I filed it later on. I can know where to put something or I can know where to pull it from. Never both, somehow. (I am as confused by this as you are, but it is true for me.)
So everything winds up in piles, both physically in real life and in my filesystems. The only way I can remember where something is, is to remember what the pile looked like when I placed that thing on the pile. Then I know exactly where it is.
In network file folders I can find something by remembering my own history in that folder and remembering about how many items were there when I created the thing I'm looking for. Then I sort by creation time and I count that many down, and it's usually one or two away from that number, in one direction or the other.
My windows desktop is organized autobiographically as well. I can never remember what I named a folder, but I know exactly where it is on the desktop.
I'm basically a nightmare for anyone that needs a tidy environment. If you clean up my stuff, everything you organized is lost to me forever.
After reading this post, I have started to organize my documents and projects according to the description. I feel like this is going to be helpful not only so that I can find what I'm looking for, but so that I don't have 4 copies of every giant source code repo taking up my disk space.
I do have one question: in the examples, the category shows "40-49 BlahBlah", but the subcategory goes "41 Taxes, 42 Expenses, etc." Is there a reason why X0 is skipped? Is this so you have room for something you might have forgotten?
> We reserve these numbers for ‘management & meta’ information about the category. Although I hardly ever use them, it’s rare that I have more than nine categories in an area. I don’t miss the wasted number, but I always appreciate it when I need it.
[0]: https://johnnydecimal.com/concepts/what-about-00-09/
Also recently I’ve been starting my IDs at `.11`. Same idea — sometimes you really want stuff to always appear ‘at the front’ and this lets you achieve that.
In the project system I’m setting up now, for example, we’re using the PMP methodology [0]. This very neatly divides up in to:
For each of these, you have a [x] Management Plan. They always sit at `2x.01`. Other mandatory artefacts will live in the zeros.Then I won’t fill up the zeros, I’ll skip straight to .11 in each category for the things that are ‘personal’ to this particular project. I’m expecting that they will not be standard across projects.
[0]: https://www.pmi.org/certifications/project-management-pmp
Thanks! I put a lot of work in to it and I'm no designer let me tell you.
FWIW it's running on Gatsby. SHOCK HORROR JAVASCRIPT. ;-)
Helps a lot that it's very light on images.
It’s a real PITA to be honest. I’m typing this from bed and can’t remember the exact details but chuck me an email and I’ll have a look over the weekend. (Or just scan the repo if you can be bothered, but I’m happy to try to remember what I did and put it in a mail.)
https://github.com/johnnydecimal/johnnydecimal.com or hello <at> the domain.
The site is written in MDX and then `gatsby-plugin-mdx` does the heavy lifting. What I could not get to work was Markdown formatting for inline images, especially not for smaller images. They always ended up being shown at 2x.
So now it's just an image, imported in to the file at the top, and displayed further down.
And then just a straight `<img src={outlook_folder... }` where it needs to appear.As for making them sharp and small, they're just screenshots from my Mac run through one of the many online PNG resizers. From memory that gave me a file with a 78% saving and if it wasn't near magic then I couldn't tell you how it did it because there was zero difference to my eye. I can't remember which one I used, just try a few.