Knowledge Management Workflow Diagram

Hi all, this is my first time posting to the forum having been a long-time admirer from afar :grin:

I’m not sure how much this has been specifically discussed on the forum but right now I’m particularly interested in the idea of knowledge management workflows i.e. how people process different types of information, take notes, and put those into Obsidian.

After watching lots of YouTubers’ videos to understand their workflows (namely Brian Jenks and his video - My Comprehensive Obsidian Workflow For Zettelkasten and Evergreen Notes) I decided to create a diagram using Mermaid.js to outline how I process the media I read, watch, and listen to.

I would be really curious to see if other people have a similar type of diagram they’d be willing to share here to give some insight into their process.

Personally, I’m struggling a tiny bit with how to get useful nuggets of information from Podcasts I listen to so I’m still trying out a few different methods to get this part more concrete.

In most circumstances though, my main “output” from listening to podcasts (aside from what I learn from simply listening) is the resources linked in the show notes which are normally web articles to read; so no issues there.

Anyway here is my diagram (any feedback welcome)

Thanks :slight_smile:

12 Likes

Hi, I saw also the Bryan Jenks Video and it inspired me. Thank you Bryan!

I was working to day with a student of law on checking a legal workflow, where you have to proove several steps.

This is our result:

graph TD
A[A-Klageschrift]
A-->B[B-Rücktritt begründet?]
B-->C{C-Liegt ein Schuldverhältnis vor?}
C-->D1[D1-Vertraglich wg. AGB]
C-->D2[D2-Gesetzlich]
D1-->E{E-Mangelprüfung}
E-->F1{F1-Zeugenvernehmung}
E-->F2{F2-SV-Beweis}
E-->F3{F3-Eigene Kenntnis}
F1-->G1[G1-Ja]
F1-->G2[G2-Nein]
F2-->G3[G3-Ja]
F2-->G4[G4-Nein]
F3-->G5[G5-Ja]
F3-->G6[G6-Nein]
G1-->H1[H1-Rücktritt begründet]
G2-->H2[H2-Rücktritt unbegründet]
G3-->H1[H1-Rücktritt begründet]
G4-->H2[H2-Rücktritt unbegründet]
G5-->H1[H1-Rücktritt begründet]
G6-->H2[H2-Rücktritt unbegründet]

What I found was that it it useful to have a single character for every layer. In the layer it is useful to number each character. I put the characters also in the box with the text. So it is easier to orientate between the characters.

1 Like

Hey all! Glad you all received some value from the video :slight_smile:

This is what my diagram currently looks like these days:

graph TD;
    A((Incoming Media))
	A-->B[raindrop.io]
	A-->C[Research Papers]
	A-->D[Podcasts]
	A-->E[Videos]
	A-->F[Digital Books]
	A-->G[Physical Books]
	
	
	subgraph " "
		B-->M1[Read & file items]
	end
	
	M1-->L6
	
	subgraph " "
		F-->L1
		C-->L1[Gather papers <br>in zotero]
		L1-->L2[give good meta <br>data tags]
		L2-->L3[read & markup]
		L3-->L4[extract with <br>zotfile]
		L4-->L5[run md note on <br>extracted notes]
	end
	
	L5-->L6[put lit notes <br>into obsidian using <br>lit note templates]
	
	subgraph " "
		D-->N0{Listening <br>on the go?}
		N0--Y-->NN[grab Airr quotes]
		NN-->NN0[Caption them <br>with thoughts]
		NN0-->NN1[Export <br>to Markdown <br>with transcript]
		NN1-->NN2[Airdrop <br>to computer]
		NN2-->NN3{only a <br>single quote?}
		NN3--Y-->NNN1[use Airr <br>page template]
		NN3--N-->NNN2[put quotes and <br>links into podcast <br>template, no embedd]
		N0--N-->N1[put podcast player <br>into obsidian note]
		N1-->N2[2 copies of note open <br>listen and noteate]
	end
	
	NNN1-->L6
	NNN2-->L6
	N2-->L6
	
	subgraph " "
		E-->O1[watch and notate <br>with Yinote]
		O1-->O2[get output from <br>google drive]
		O2-->O3[clean output]
	end
	
	O3-->L6
	
	subgraph " "
		G-->P1[read and hand <br>write notes]
		P1-->P2{Lengthy <br>or<br> Complex?}
		P2--Y-->Q1[Transcribe <br>each chapter]
		P2--N-->Q2[Transcribe <br>whole batch]
	end
	
	Q1-->L6
	Q2-->L6
	
	%% The final obsidian note generation workflow:
	L6-->L7[Lit notes into seedbox]
	L7-->L8[Review lit notes <br>and generate seedlings]
	L8-->L9[Incubate seedlings <br>with thought <br>and linking]
	L9-->L10>Plant seedlings <br>into Evergreen forest]

you also may find some of the mermaid.js code in my map of moc interesting for future diagraming/refactoring your workflow diagrams :slight_smile: i might need to re-look at my own :rofl:


graph RL;
	%%==============%%
	%% Declarations %%
	%%==============%%
		A[INDEX]
			click A INDEX;
		B[Interests]	
			click B Interests;
		C[Statistics]
			click C Statistics;
		D[Programming]
			click D Programming;
		E[Vim]
			click E Vim;
		F[Bash]
			click F Bash;
		G[Python]
			click G Python;
		H[R]
			click H R;
		I[C++]
			click I "C++";
		J[LaTeX]
			click J LaTeX;
		K[Groff]
			click K Groff;
		L[Business]
			click L Business;
		M[YouTube]
			click M YouTube;
		N[PKM]
			click N PKM;
		O[Religion]
			click O Religion;
		P[Heathenism]
			click P Heathenism;
		Q[T-SQL]
			click Q T-SQL;
		R[Markdown]
			click R Markdown;
		S[Twitch]
			click S Twitch;
		T[Twitter]
			click T Twitter;
		U[Zettelkasten]
			click U Zettelkasten;
		V[Spaced Repetition]
			click V "Spaced Repetition";
		W[Technology]
			click W Technology;
		X[GitHub]
			click X GitHub;
		Y[JavaScript]
			click Y JavaScript;
		Z[Web Development]
			click Z "Web Development";
		A1[CSS]
			click A1 CSS;
		B1[HTML]
			click B1 HTML;
		C1[Academia]
			click C1 Academia;
		D1[School]
			click D1 School;
			
		class A,B,C,D,E,F,G,H,I,J,K,L,M,N,O,P,Q,R,S,T,U,V,W,X,Y,Z internal-link;
		class A1,B1,C1,D1 internal-link;
	%%=========%%
	%% Linking %%
	%%=========%%
	%% Index
	A --> B & L & N & O & C1 & D1
		%% Business
		L --> M & S & T
		%% Interests
		B --> C & D & W
			%% Programming
			D --> F & G & H & I & J & K & Q & R & Z
				%% Web Development
					Z --> Y & A1 & B1
			%% Technology
			W --> E & X
		%% PKM
		N --> U & V
		%% Religion
		O --> P

cheers!

10 Likes

I learned four things from your post:

  1. round boxes with (( ))
  2. Existance of Subgraphs
  3. Yes/No Marking after Decision Boxes seems useful
  4. linebreaks with

THANK YOU!!!

2 Likes

and theres more here:

:slight_smile:

1 Like

@argually
thanks for your insight, your naming convention seems much simpler than my initial approach :+1:

@tallguyjenks
very helpful! thank you for sharing the code behind your graphs too, there are some concepts in there I hadn’t considered :thinking: subgraphs seem quite useful too.

Definitely some concepts I want to integrate to a v2 of my diagram.

I also like your idea of a map of mocs, that’s something that would be helpful for me to build out as my vault develops :smiley:

Appreciate the video link also!

3 Likes

Cool overview, I’m also working towards a Unified research & knowledge management workflow


I also use PlayBooks for all book highlights, wrote a post on how to get them out of Google Docs


As also shown in the video by @tallguyjenks, Airr.io for “airrquotes” of podcasts is great (currently only iOS; bulk export to md available but no automatic / scheduling yet)


Memex for webhighlighting and export to markdown (with custom templates and variable i.e. {{PageURL}}, {{PageTitle}} to fill metadata of note for Obsidian)

2 Likes