Comment in a nutshell: "Practice with very simple tools optimized for swift use in a culture that values logic and evidence are likely to be better than specialized tools with hard coded abstractions."

There's a specialized form of note taking called "flowing" within the policy/CEDA/NDT debate community. Here is a wikipedia article on the subject if anyone wants to hunt in the concept space for keywords and links and such :-)

In this speech community (especially at the higher levels) people tend to speak very swiftly because there is a background theory that "dropping arguments is conceding arguments" which creates an incentive to make many parallel arguments with the same conclusion that reach towards different sets of evidence. It is possible to win simply by making many adequate arguments that your opponent is incapable of handling with enough speed or concision.

(It can be dangerous to spin out many arguments without paying attention to how they interact, because sometimes a response will be something like "I concede argument 2 and 5 which constitute a turn on the general position that works like X. Arguments 1, 3, and 4 were about A, B, and C which are all less important than this, so I win the position.")

In terms of basic mechanics, the note taking process that I learned used strikingly simple tools: a four color pen, a legal pad, and (if you were going to prep for standard positions or responses by taking re-usable notes on what you would say) cover up tape. The process involved taking notes in a very narrow up-down column (about the width of cover up tape) so that iterated response and counter response could move horizontally across the page in rows. Time through each speech flows down on a given page, speech to speech flows horizontally. Arrows sometimes connect points if you underestimate the space requirements for a "clean flow" to help show how the rows are supposed to line up.

An eight minute speech would be flowed on several different pages, usually called "positions". As a matter of helping the judge flow the round (so they can take clean notes on your arguments and will notice when you win) it was wise to start a speech by announcing how many new positions would be introduced and/or the order in which existing positions would be addressed. Then everyone shuffles papers for the first few seconds of the speech. For a debate round on cryonics there might be a positions (each with a separate piece of paper) for "personal friendships and relationships", "health-dollar utility", "dehumanization critique", and so on.

Sometimes a column on a page would be a simple list of numbered points, but sometimes an argument would actually have an implicit tree structure, with mulitple points having sub points of their own. Comparing columns, then, its best to imagine one serialized tree next aligned with another. Structured argument was much more common in the early speeches of a round when significant preparation was possible and evidence was being read whose implications would be argued later on. Also, each piece of paper functioned sort of like its own "major tree branch" for the entire round (if you're trying to map the physical objects to an abstract data structure) and it was rare for cross applications of logic to jump from one page to another, but more common within a page.

The actual note taking would be full of your own jottings for arguments, like I used to use a capital T with a circle around it for "turn" which meant that the argument related to the "sign" (positive or negative) of the general position. An extended example might help...

Imagine that one side argues that the spread of some public perception would have a terrible consequence; then the other side might concede at the consequence level but proceed to argue that the perception already exists in the status quo and is growing and would actually go down on the basis of a proposed course of action... that would be a turn. In notes I might have one side flowed as "2. Perc {right arrow} H" (the 2nd argument was that perception is causally connected to harm) and then the response would be recorded just to the right with something like "T: Perc {up arrow}ing SQ" (perception is increasing in the status quo and somehow the proposed policy would decrease it, changing the sign of the broader argument).

Another important thing for the actual debate was that people would read sourced evidence into the round (at very high speed) so aside from the gist of the content, the author and year of publication could be very helpful because you might be familiar with them and have a retraction (or rebuttal by another author) written two years later in your own debate files. Even if you have no response you can go to the library with the citation and prep for the next time you see that evidence in a round... or you could use it yourself to argue the other side if it's just a great position :-)

One issue is that note taking systems like this work much better when the speakers are trying to be logically intelligible and expect you to flow them. The speakers can "signpost" as they raise a point by saying things like "...and my fourth argument here is that X" and then it can be responded with "their fourth argument about X is spurious because of Z". The numbers in the speech and in the notes help everyone retain the same alignment.

Sometimes I would debate using flowing skills in competitive contexts that were focused on rhetoric rather than reason (less signposting, less note-taking by the judge, speaker points awarded for wardrobe, little tolerance for meta-arguments about debate theory to enforce compliance with social norms, etc) and the note taking system generally still worked.

If the audience has a phobic reaction to stuttered logic and reason you can get hurt relying on the cultural forms of policy debate (in policy debate, T-shirts could reverse-signal to some judges that you relied on "pure logic"), but in the absence of phobias, "flow skills" still appeared to be relatively valuable because they helped you address people's points directly (which audiences usually like) and notice when they tried a logical shenanigan (which academic audiences enjoy seeing called out and given a latin name).

It's been many years since I judged or coached, and even more since I debated myself. My understanding is that in modern times people frequently bring laptops into rounds and flow using spreadsheets. Evidence is stored on hard drives instead of lugging around 100 pounds of paper in a filing system. It sounds altogether more civilized in that respect -- and learning to use the hotkeys for high speed note taking in spreadsheets is probably a skill that would help for years after in contexts like finance and consulting :-)

In any case, these sorts of experiences have adjusted my priors on the subject of rationality prostheses to favor of practice and skill with generic tools (paper and pen, or spreadsheets) as opposed to highly specialized software with a bunch of YAGNI features.

If any readers are in college (or better yet choosing a college), I would recommend googling around to see if you can find an associated intercollegiate policy debate team. Sometimes they are called CEDA (Cross Examination Debate Association) and sometimes NDT (National Debate Tournament) although the formerly distinct communities had already pretty well merged 10 years ago when I was debating. It can cost 20 to 60 hours a week, but I know of no other "game" with as wonderful a mix of adrenaline, logic, and language. In retrospect, I think that policy debating was more educational than many of my actual classes :-)