If you use a personal knowledge management system (PKM) like Obsidian, RoamResearch, LogSeq or RemNote, you have a graph of your ideas stored as a whole. While some of those ideas are private, you might want to share a part of your knowledge graph that you want to make public. This way you can share them with others and analyze them using InfraNodus PKM Knowledge Graph Analysis Tool, to reveal main ideas, the relations between them, and expose the gaps in your knowledge, which you can fulfil with the help of GPT-3 AI plugin inside InfraNodus or with the help of your readers.
The best software to export your graphs is the open-source LogSeq PKM tool, because you can choose just one folder on your hard drive that you'd like to visualize as a graph. Then you can export that whole folder using LogSeq and make it public.
So the only question is to get your data into LogSeq. What you need to do is to export your private graph from RoamResearch to Obsidian, use Obsidian's folder structure editor to move all your public pages into a separate folder. As this folder is on your hard drive anyway (or a synced cloud like Dropbox or Google Drive), you can then import that folder using LogSeq, then export those pages from LogSeq in the MD (markdown) format.
Here are the instructions step by step:
Step 1: Choose a Knowledge Graph to Share
At this point, you need to decide which knowledge graph you would like to share. In my case, it's a graph I have on RoamResearch which I started writing on fractal dynamics.
Step 2: Export that Whole Graph (from RoamResearch)
Go to Graph > Export Whole Graph (or Export All) and save your graph on a local hard drive in the MD (markdown) format.
Your files will be named as your pages. The references to the other pages are contained within those files using the [[backlink syntax]].
Step 3: Open your Graph in Obsidian, Create a Public Folder
Once you open your exported folder, you will have a lot of pages and probably no folder structure.
You can move your files around into folders, making sure all your public ones are in a separate folder.
Perhaps, it might seem a bit difficult, but in fact it is not. What you will probably want to expose is something that's more or less finished. It will be a different kind of notes, which are more defined. They will have links to the other pages but as those pages are not public, you will not get any backlinked content to them, so that part stays private.
Once you have your public files in a separate folder, close Obsidian.
Step 4: Open the Public Folder in LogSeq
Go to LogSeq.Com and create a new graph session. Choose your local Public folder that you created in Step 3. It will be visualized and shown by LogSeq as a graph. Go to Export Graph > MD (Markdown) format. Export that whole graph, which is, in fact, a public folder.
You can now expose just this public graph using any software you like (Obsidian / RoamResearch / LogSeq).
Step 5: Let Others Explore Your Graph
Your knowledge graph becomes a public ideogram that can be used to explore a certain aspect of knowledge.
If you leave your original files open, you can then also send them to InfraNodus. It will perform text mining on your semantic data as well as network analysis on your backlink knowledge graph structure. InfraNodus will combine the both and show you a general picture. The main topics, the key concepts, the connections between them.
Anyone will be able to use this visual overview of your knowledge and to query it using the network graph and the AI. The built-in GPT-3 language generator will detect structural gaps and interesting patterns in your ideas. It will then propose a user to generate a research question or an idea that would link those gaps and rewire the existing patterns in interesting ways. So your ideas are having a conversation with another person and the whole society at large via AI (as AI is built on the basis of typical human knowledge). Which is an interesting way to make them alive again and to propose a different perspective on whatever it is that you are writing or thinking. A text is not anymore something you read from A to Z. It is a non-linear network entity that you can explore through the relations. Playing the graph of knowledge using the edges as strings of the web.
Please sign in to leave a comment.