One point that is probably not your problem, but might be useful to address at some point.
$Name(parent)=="Synthesis Map"
consider using inside(/path/to/Synthesis Map). The latter is potentially more efficient, and has some advantages for the semantics of aliases.
Also (from brief scan-not time to test ATM) the Synthesis Map agent is off, so the agent action wonāt be firing. If Evidence 6ās āOpposesā link was added after the item was added to the agent map, I donāt think we expect the agent action (IOW, the agent OnAdd equivalent) to fire until the agent is active again.
If you donāt want agent updateās to re-organise the agent map, see Re-arrangeable Agent Maps.
I think you are acting accordingly but Iād note to be aware aliases may have different basic (i.e. note-to-note) links than their original.
Thanks!
This is really helpful and intriguing, Arek. Thank you for sharing your extensive work and process with us.
Marianne
I wonder if a tag or two could help to keep this post from getting lost? I see thereās one for ātemplatesāāmaybe also ācase studyā? (Iām the last person to suggest tags: I have far too many overlapping ones in my own work!)
Hey there, couple
BTW, I looked at a few of your functions. I see that youāve named them an āedict,ā ārule,ā or āaction.ā
I fear this might confuse some people. Functions are action codes that can be used as edicts, rules or actions; they need to be specifically named in this way.
Also, you have several note names with the term ālinked.ā I would expect these actions to perform link operations when in fact, they do not. They are simply provided associates by basing values from one note to another.
I added a couple of tags to the thread, but good ideas can get lost in tags just as with anything else.
I strongly suggest for anyone interested in a given post or thread, that the ābookmarkā feature in Discourse is excellent.
Click the three dots in the tools at the bottom of a post, and then click the bookmark icon
Youāll have several options -ā choose a reminder or choose ānone neededā to make a permanent bookmark for yourself.
When you want to find your bookmarked topic, click your person-icon and choose the bookmark icon again.
OMG!
This is way too much work, but interesting.
I basically batch add bibliographic info into DevonThink for each note from a text, then I just drag the files to TB.
There, I have in a container propositions with adornments which add a statement such as verifies, denies, etc. I drag notes there.
Iām going to have to take a closer look at your work, but for someone like meā¦ Iāll throw my computer out the window!
I also add
Iāve lately changed my workflow substantially. Iām planning to update this topic as soon as I have some spare time 
Thanks. I think we all appreciate the work that goes into documenting (and updating) a process for others to be able ājustā useāeven if some effort is required. Some things do take a bit of effort.
One additional area that could be considered in the design of this model, could be to add a prototype, pAssumption which could represent notes that are assumptions.
Tom
Hi ShiJianhui
Just curious, are you still using / developing this discourse graph? Has anything changed. It has become quite the hit and was even mentioned by Joel Chan in his meetup.
Did you get a chance to watch it?
All the best,
Tom
