On the back of DefenderOfBasic having released the latest homepage for his ORI chapter (2026-04-14 EDIT: Also see Shadow Rebbe's onboarding article), I am continuing from ORI 100 to explain how you, someone who wants to participate in (or even found) an open research institute, can get started.
This is a starting kit that outlines and describes technologies (tools, cookbooks, and frameworks) which you can use to begin creating a community and literature.
None of these items are strictly necessary; your open research institute could simply be a groupchat, or even your own personal notebook. But if you want a simple guide, you have come to the right place.
I will update this page over time; just trying to get this out there, since I have heard that there is a lot of new interest in ORI!
Tools
This is a list of applications which you can use out of the box.
personal knowledge management system (PKMS)
Obsidian
stable permalink social media
X / Twitter
Bluesky - built on AT Protocol
Reddit
walled garden social platform
Discord
Mastodon - built on AT Protocol
Slack
Facebook
static blog post or website creator
Substack
Leaflet - built on AT Protocol
Markdown Space
Cookbooks
Open research institute rules
As I explained in ORI 100, my conception of an open research institute is that it has to at the very least allow use of its materials by all other open research institutes.
You might have a different definition of an ORI, and that is okay! However, I will not consider you an ORI unless you satisfy that requirement. (This is not necessarily a moral judgment, just a descriptor of my own ontology.)
Below are the rules in The Syllabus, as of 2026-04-13. No one can post in the server without agreeing to these rules.
The Syllabus - Canon #canon
Rules
Any message you post in here must either be public domain or be relinquished into the Creative Commons under the Zero license (CC0).
If you cannot do this (either because you do not wish to or because it is not your call to make), please send a link instead of sending it as a message.
Do not break the law or post inculpatory statements.
Guidelines
Consider that this is whole server constitutes a communal document.
Try to not swear.
Keep extended small talk and debate in threads or in your own log.
Hyperlink aggressively.
If you like a thread or train of thought or exchange or post, consider writing a review!
If you make a connection between two different places in this vault, write it down and send it!
Don't be afraid of "necro"! Have fun responding to dusty forgotten messages and rekindling old smoldering threads.
Feel free to forward, screenshot, or remix any message in this server!
Remember: only post if you agree to release the text (and preferably immediate attachments; links to online resources is fine) under CC0.
Join @omarshehata's open research institute (ORI) chapter's Discord to see a larger part of the network! You will get a guest role from using this invite. (Invite redacted.)
A space for your ORI chapter
The usual way of participating in an ORI is to join an existing ORI. However, for many reasons, you might want to start your own ORI. If so, I encourage you! Starting your own ORI does not mean not participating in others' ORI; I am an active member of Defender's ORI, despite having started The Syllabus in part as a criticism of his ORI's shifting public focus.
Starting my own ORI was as simple as starting a Discord server and inviting some initial members (incl. Defender). I did this before I even had any rules written, though I quickly drafted those up to establish my vision of what an open research institute is, and what the Open Research Institute is. Then, I set up some server architecture which has been proven to work, like parallel logs for each "researcher".
Perhaps the most important part of a space, the one that defines its boundaries, is its mission. The exact mission of anything I do as a Christian is going to have to hang on the cross of Jesus, and so it is easier for me to communicate the specific proximate goal of The Syllabus by simply stating instructions, plus my vision of what those instructions will bring about.
Here are some adapted notes from a thread in the Discord for The Syllabus, under a thread titled central command + tech tutorial:
Here are tweets that express what I want to build:
Parallel log channels
Both Defender's ORI and The Syllabus have Discord servers that use a system I created, where every user has their own log. This log acts as a place for them to organize their thoughts or rattle off some musings, and for others to give feedback or contribute.
The logs system should be used in conjunction with more "equal" spaces for conversation; it is simply a way to give more space for different concurrent trains of thought.
In Defender's ORI, the logs (which I have only screenshotted in part, due to there being many) are each named after the researcher who "owns" the space, but are available for everyone to post in. The result is a mix between an office and a mailbox, where anyone can drop into anyone else's space.
Starting up this kind of log system is as easy as making and naming the channels. Everyone's Twitter account is already a log system!
In The Syllabus, the logs (which I have screenshotted in full, showing all active threads as of 2026-04-13) only allow the named researcher to write top-level messages to the channel. Everyone else can send messages in threads, and can start their own threads. The result is a structure much more similar to the blogosphere or YouTube, where one creator has a clear real estate advantage within the space.
Implementation of this system requires that each researcher has their own dedicated space where they can write top level messages that others cannot. On Discord, messing around with permissions a little will get you the desired result:
Create a new category.
In the new category, change the default permissions for
@everyone:No one can send messages in the channel.
Everyone can start threads and send messages in threads.
No one can pin messages in the channel.
Create a log for each researcher.
I recommend only creating a log for each researcher that asks for one; it helps to prevent there being a whole bunch of unused logs that you have to archive later.
Set the permissions for each log:
Make a new set of permissions for the "owner" of the log.
Give them the following permissions:
RESEARCHERcan send messages in the channel.RESEARCHERcan pin messages in the channel.
And there you have it! A functional log system!
NB: If you are the owner of the channel, you are always able to write in every log system. This means that you will have to be careful to not intrude on someone else's space. A way to prevent this is to give ownership of the Discord to a secondary administrator account, and change all permissions from there. This also makes it easy to test if you have set all permissions correctly, without needing someone else to check for you. (Though, you should still get a second pair of eyes!)
zettelkasten on YouTube.) However, you do not have to adhere to any specific system to easily manage your knowledge. Knowledge management systems
Much is written about personal knowledge management systems (PKMS). (If you want proof, look up zettelkasten on YouTube.) However, you do not have to adhere to any specific system to easily manage your knowledge.
In my own day-to-day life, in which I write copious amounts in Obsidian and online in shortform, I do not explicitly try to follow the principles of any specific system like zettelkasten. Rather, I have a few principles which I follow that makes my own life much easier:
Use plaintext.
I have way too much past work that is trapped on proprietary platforms or which is difficult to extract from weird non-human-readable file types. The absolute best part of Obsidian (even better can being able to quickly search through tens of thousands of notes for keywords) is that even if it were gone tomorrow, all your Markdown files are still perfectly readable and even might still work on other platforms. This is why I use Obsidian as a notebook throughout the day. (See this YouTube video by No Boilerplate on plaintext.)
Many other apps which do not create plaintext files still use and display plaintext. Bias toward using platforms that are easy to copy and scrape from.
Use unique file names. No more collisions; embrace the forward march of time.
The easiest way to do this is to use the format
YYYY-MM-DD-HHmm(ss)-hyphenated-slug. The best part of this system is that you always have a title to default to which also gives you a lot of information on when you wrote something (and often in what context you wrote it).This works very well on Obsidian, and I have my note creator and refactor plugins adjusted to create notes with
YYYY-MM-DD-HHmmssprefixes. However, this works equally well with a directory of .txt files or a bunch of blog post drafts.The idea is just to put a title that lets you (1) easily find it among similarly-titled documents, and (2) does not cause annoying namespace collisions.
Gary Gygax, creator of the seminal tabletop roleplaying game (TTRPG) Dungeons and Dragons, wrote in the AD&D Dungeon Master’s Guide: "You cannot have a meaningful campaign if strict time records are not kept." In fact, solo journaling my own roleplaying campaign in a persistent world was one of the first uses of my Obsidian vault. I have since moved on from that, but the idea of Gygaxian time and strict timekeeping has stuck. I find it much more meaningful to "do research" in my own notes when looking at the list of notes shows me a history of my thoughts and interests.
Another way to use "unique names" is to have mnemonic keywords which you put into places in your vault that are easy to look up. These will differ for every person. It helps to keep track of what your most common search queries are.
Have a default place to write in.
For me, the daily note in Obsidian, each named with the
YYYY-MM-DDformat, is where I write almost 99% of my stuff. Only if I realize I can refactor the note (and this usually happens long after I wrote it, rather than right in the moment) do I use a shortcut to trim out an excerpt into its own note, which also leaves behind a link in the original daily note. (On Obsidian, I use the Note Refactor plugin by James Lynch.)The analog paper version of this is to have a scrap notebook from which you are willing to trim out pages to put into a more stable collection. This is how many commonplace books, the pre-digital European personal knowledge management systems, used to work!
Let yourself write off the top of your head. Capture your consciousness right now, as it is.
Think of a jazz musician. Do you hold it against them that they added a funny little lick that no one else would, that only they might fully understand? Of course not. Part of the beauty of jazz is that you are following along someone's train of thought, enjoying the flow of them working in the moment. Treat your own mindstream the same way.
If you are not a fast typist, learn to touch-type and begin letting yourself react to your own words on the page or the words or images in your mind. Alternatively, start dictating with your voice; speech-to-text is already very good, and will only get better and better.
Embrace redundancy. Copy-paste yourself a lot.
I cannot possibly keep count of how many times I have caught myself only hours into shuffling between two notes that I could just have created a third one that combines those two in an easy-to-read format, maybe even with some irrelevant fluffy stuff trimmed out.
When writing something that you have written about before, just copy over the old stuff and rework it from where you left off. Save the new copy alongside the old copy.
Many Obsidian users become obsessed with the idea of reusing the same barebones set of modules over and over, but this is only really useful insofar that it helps you work, whether by saving space or by creating better connections; personally, I find it easier to just make a new note and put into it exactly what I need in the moment from other notes which I link. Then, if I ever need to link to a note from a particular cluster, I use the one that I can most easily think of.
My Obsidian vault is text-only with embeds all hosted online, so even after extensive use for almost three years it is still only about 250MB, including all snapshot backups.
Save links to common digital locations. Treat your vault as map and territory.
"Common" means "shared"; it could be shared across almost all instances, or across only a relevant few interests. Interpret this as appropriate.
When I find myself using a set of websites or specific pages a lot, I put links to them together in my vault. For example, when I hopped back onto Neopets for a while, I had a list of links to daily features and to specific Shop Wizard and Safety Deposit Box search queries.
Anticipate what you will use later. Because I use X almost every day and frequently refer back to old tweets, I have links to almost all the tweets I write on my laptop. I draft my posts in my Obsidian daily notes, copy them over to X, then grab the link and paste it next to my tweet. Sometimes I will never use the link, but I use them often enough that it ends up saving me a lot of time when I am racking my brain for a specific wording I used.
Sometimes I will find myself thinking about a particular note in my vault and will not know why. When I do, I will drop a link to that note and move on, just in case I figure out the connection later.
I have glossary notes in my Obsidian that gather together groups of notes on common themes. Often, looking through them gives me inspiration for new trains of thought, or shows me which notes can be recombined into a new note or disambiguated into two sets of separate notes. Sometimes, instead of finding a specific note, I will just link to the glossary note that I think will help my future self find the needed note. (Almost like giving my future self a fetch quest.).
None of this is specific to Obsidian, necessarily. In fact, a lot of this might still work if you used, say, a Discord server or your microblog as your personal knowledge management system. Remember: the purpose of a tool is to achieve the best end. No matter what you do, think about what is beneficial. There is no rule but the rule of (ultimate) benefit. The end justifies the means.
Frameworks
Knowledge as building blocks. Like a university's syllabus of course prerequisites and corequisites, a literature might have many dependencies which not all interlocutors share.
Save and write clear atomic descriptions of singular ideas which you can reference. LessWrong has a very good example of this. Their page of concepts serves as, within the context of their collective knowledge management system, an index of evergreen notes. Having something like this at your disposal can help your community be on the same page. Using Wikipedia (or another online encyclopedia specific to a field, like the Stanford Encyclopedia of Philosophy) can be a great supplement at first, but building your own community writings can clear up any fault lines that appear.
Social media as a roleplaying game. The exact type of game is decided by the players. Within the ORI literature, the base layer of the game should be positive sum; if the network of hyperlinks is bigger, there is more architecture for everyone to use! (Of course, competition for attention itself might be zero-sum. How do you win there?)
Focus your time and energy on what you want to see more of. There is also a place for polemics and for complaining about things you hate, since it can better clarify what you love. But in the end, a corpus cannot only be about what it is not, but rather about what it is. Existence is goodness, so there is no replacement for building up what is. Think of the parable of the wheat and the tares, where the tares are allowed to grow among the wheat but are thrown into the fire after the harvest. If your enemies are truly wrong and are utterly unrepentant, let them and all their works perish with time.
The public domain is a social media platform. Borrow freely from materials that are available to everyone, and participate in a literature extending far beyond our own timelines.
Remember: everyone's work automatically becomes ORI-compatible when it enters the public domain. This means that everything is, given enough time, part of the Open Research Institute. It is inevitable.