Input a domain, output its rank.
What Is Rank
Rank is not "key elements", not "core principles", not "summary points".
Rank is: how many truly independent generators are there in this domain? Can you use them to reverse-generate all phenomena? If yes, only then have you found it.
How to Find It: The Penetration Process
The criteria are checked afterward; the effort to find rank lies in digging. No matter how strict the criteria, they can still let false roots through—if the process was not carried through to the end.
Seven Steps—walk through them in your mind before writing, do not write them into the article:
- Lay out phenomena—10+ representative phenomena in the domain, just lay them out, do not explain. This step is to give rank a checklist to test against for reverse generation.
- List candidates—for each phenomenon ask "why is it like this", and put out candidate generators. At this step there will be many candidates, mixed with a large number of false roots.
- Recursive questioning—the key to penetration.For each candidate, ask one more cut: is there anything more basic below this one? Keep going until you can go no further. The most common failure is stopping at the phenomenon layer—for example, reducing the rank of "entrepreneurship", where the candidates are vision / execution / team / market; they all look important, but this is the surface "key elements" layer, not the root rank. Ask further: where does vision come from? What does execution rely on? Keep going until you can go no further; that layer is the root rank.
- Merge same-source—two candidates are actually the same generator manifesting on two surfaces; merge them. The discriminating question: remove one, will the other still happen?
- Cut—remove one, can the rest be used to reverse-generate all phenomena? If it can be cut, throw it out; if it cannot be cut, keep it. This step is the practical version of the "minimality" criterion.
- Reverse generation—take the remaining few and reverse-generate the phenomenon checklist from scratch, and see whether each item can be reproduced. Phenomena that cannot be reproduced point to a missing one.
- Look for counterexamples—can these few predict phenomena outside the checklist? For those predicted, do they actually exist in reality?
Only after completing these seven steps do the few candidates have the right to be placed on the criteria bench.
Walk through the process in your mind, butdo not write it into the article—like the criteria, it is internal skill, not a narrative hook. Readers cannot see the digging process, but they can feel that behind it, the digging went all the way down.
Four Internal Criteria
These four are the after-the-fact gates you run through again after completing the process,and they are not narrative hooks either。
- Generativity—the generators can derive every phenomenon back
- Minimality—remove any one, and there are phenomena that cannot be explained
- Independence—for every pair, you can find a case where "one changes and the other does not"
- Predictive power—can derive phenomena outside the checklist, and they indeed exist in reality
If even one is not passed, tear it down and start over—go back to step 3 of the process and keep digging.
Once the criteria are solidly passed in your mind, butDon't write this into the article——Don't let "chop it once and see" or "pull one out and try" verification actions appear. They break the reader's breath from "messy" to "simple." Readers can't see the verification, but they can feel there's something solid behind it.
Two-layer judgment: first find the root rank, then ask whether a coordinate system can be layered on
Once the criteria are checked, several lines have been found. Don't rush to write yet—first judge their form through two layers.
First layer: find the root rank
The root rank is the true root skeleton of this domain—what it is really saying. Don't preset how many there are or what relations they have, just find the root. It may take the form of:
- Several parallel mechanisms (pillar style)
- Several nested mechanisms (one layer supporting another, the deepest one being the meta-proposition)
- Several interlocking feedback loops (ring-shaped)
- Several progressive stages (chain-style)
After reading the root rank, what the reader gets isa worldview—the ontological skeleton of this domain, a new way of seeing the world.
Shape counter-question · anti-collapse gate(after finding the several lines, mandatory self-check before writing)
Go through the nine shapes one by one—
- Drilling profile? (one layer supporting another, the deepest one being the meta-proposition)
- Several in parallel? (pillar style, orthogonal and slidable)
- Feedback loop? (mutually driving each other positively and negatively)
- Stage chain? (one stage after another, with temporal or logical order)
- Tension spectrum? (one line pulled at both ends, cases falling in between)
- Tree branching? (one branching into many, many branching further)
- Network? (many-to-many interconnections, no hierarchy)
- Oscillation? (rise and fall, alternating rhythm)
- Multiple slices? (multidimensional classification with more than two axes)
The default thought of "drilling profile" is AI's lazy anchor—Any concept can be split into three layers (surface/middle/deep), looking neat, but the real structure of many concepts is not vertical drilling.
Mandatory counter-questions:
- Do these several lines really have a before/after or nested relation? (drilling vs parallel)
- Or are they actually driving each other? (drilling vs feedback loop)
- Or is there a temporal or logical order? (drilling vs stage chain)
- Or is it just one spectrum? (drilling vs spectrum)
Only after eliminating the other 8 one by one can "drilling profile" be confirmed.You cannot skip this elimination—The cost of skipping is a distorted root rank, compressing before/after relations, feedback relations, and tension relations all into nesting.
Second layer: whether an operational coordinate system can be layered on
Once the root rank is found, ask one more thing: inside it,is there any one linethat can be unfolded into several independently variable state parameters?
For example: the third line of Fritz's root rank is "rocket geometry vs pendulum geometry"—rocket geometry requires two points, "vision" and "current state"; taking the clarity and honesty of these two points as adjustable dimensions grows a two-dimensional coordinate system. The coordinate system is not a replacement for the root rank, but a child it grows on one of its lines.
If it can be unfolded, the coordinate system islayeredon top of the root rank. After reading the coordinate system, the reader gets one more thing:an operational instrument—any specific situation can be placed into the frame on the spot, to see which quadrant or which layer it falls into.
The two layers don't conflict—the mechanism type gives "what it is," the coordinate system gives "where it is," they answer different questions and can be given at the same time.
When to layer, when not to layer：
- a line in the root rank inherently requires two or more orthogonal, continuously variable state parameters -> layer a coordinate system
- each one in the root rank is a one-off action or stage (not slidable) -> don't force a layer
- forcing a mechanism into a coordinate system distorts it—for example, forcing "ignition/condensation" into two axes, one coordinate per phenomenon (x,y), is meaningless, because they are not continuous quantities coexisting at the same moment
If you can layer a coordinate system, the rank goes up a notch—when the reader comes out, they have both a worldview and an operating instrument in hand.
Framing counter-question · anti-collapse gate(before expanding the second layer, mandatory self-check)
Before confirming you want to layer, ask: the line I'm about to expand, is it really two independently slidable axes ?
- Is it actually just one spectrum? (then use a spectrum slider—the case falls somewhere in the middle, which fits better than a four-quadrant)
- Is it actually a feedback relationship? (then use a loop diagram—mark +/-)
- Is it actually stages? (then use a chain/stepped form—mark the transition conditions)
- Does it actually branch into multiple branches? (then use a tree)
The default 2x2 matrix is another lazy anchor point of AI—consulting and strategy framework training data has tons of 2x2s, any two variables can be cut into four quadrants, and it looks like analysis. But many root ranks are not two independent dimensions, they are a single spectrum, a feedback loop, a chain.
Mandatory counter-question:
- Are these two dimensions really independent? (independent vs two sides of the same line)
- Are they really slidable? (continuously adjustable vs discrete stages)
- Does every cell in the four quadrants have a real-world correspondence? (if three cells are empty, maybe a 2x2 shouldn't be used at all)
Only after ruling out spectrum/loop/chain/tree one by one do you confirm 2x2.
Chinese nativization · anti-collapse gate
Before writing the main text, silently recite three times:"Would a Chinese person who has never read English talk like this?"
After writing each paragraph, silently recite it again. If the answer is "no"—don't change the words, you need to reconceive how the whole paragraph is expressed。
The anglicized structures most likely to hit a landmine (a hit means rewriting the whole paragraph, not fixing words):
| anglicized expression | Chinese nativization |
|---|---|
| it builds on the previous one | the previous one holds it up |
| it has an inherent direction | it goes this way on its own |
| this distinction answers X, not Y | this is asking X, not asking Y |
| this means... | that is to say... / this is why... |
| in the process of... | in... / when... |
| as for... | to... / on...'s side |
| the "is..." sentence pattern | say it directly |
| conduct + noun | use a verb |
| nominalized abstraction ("X-ness", "Y-ization") | a concrete verb or concrete object |
What you change is not the words, it is the framing of the expression— from the sentence patterns of English critical thinking, switch to the sentence patterns of a Chinese vernacular author (the tone of Wang Zengqi / Wang Xiaobo / A Cheng / Li Juan).
Scanning the words afterward is just patching; it can't cure the source—the source is 'the thinking path is in English.' When rewriting, force yourself to throw out the English expressions and, from the perspective of a native Chinese author, reconceive how to say this thing.
After writing, recite it aloud and listen for whether it catches. Where it catches,rewrite the whole passage, not patch it.
How to write
A piece of prose written in one breath.
Take the reader on a stretch of road—from "this field looks pretty messy" to "so it's just these few threads pulling it along." No chapters, no subheadings; let the reasoning itself push things forward.
Three requirements:
- Read in one breath—even people who don't know this field can't stop
- Memorable—after reading, you can turn around and explain it clearly to a friend in one sentence
- Has a落差—the contrast from chaos to extreme simplicity is the beauty of rank reduction
Two-part form (if a coordinate system is layered on)：
- First half: walk the skeleton of the root rank all the way through—explain the meta-proposition, the dynamics, and the shape clearly. The reader finishes with a worldview in hand.
- Second half: unfold a coordinate system along one of the root rank's threads—explain clearly the two axes, the four quadrants, and what each quadrant corresponds to. The reader finishes with an operating instrument in hand.
Don't cut hard between the two parts; let the reasoning transition naturally—the best last sentence of the first half is a lead-in like "so how do we look at this mechanism when it lands in a concrete situation?", which eases the reader into the second half.
At the end, pair it with an ASCII structure diagram—form follows the skeleton
Just saying "these few threads" isn't enough—you have todraw the relationships among these few threads。
Butthere's more than one way to draw it. The topological form of the root rank determines the form of the framing device. Before putting pen to paper, first ask:what shape is the relationship of this rank itself? Once you've answered, then choose the way to draw it.Form follows the skeleton, not the other way around.
Nine framing-device families—this table is the master index from skeleton to drawing method:
| Root rank form | Essence of the relationship | Framing device |
|---|---|---|
| Several threads side by side, independent, slidable | Orthogonal | Two-axis / multi-axis coordinate system |
| One layer supporting another (the deepest thread is the meta-proposition) | Nested penetration | Drilling cross-section |
| One thread pulled at both ends | Tension and opposition | Spectrum / slider |
| Mutually driving each other positive and negative | Feedback loop | Loop diagram (marked +/-) |
| One segment after another | Stage-by-stage progression | Chain / steps |
| One thread splits into many, and those many split again | Hierarchical branching | Tree diagram |
| Many-to-many interlocking | Coupled network | Network diagram |
| Rising and falling, rhythm alternating | Oscillation | Waveform / oscillation curve |
| Multidimensional classification (such as abstraction x proximity x time) | Multidimensional classification | N axes / multiple slices |
Hard constraint: use only pure ASCII characters. Any Unicode symbols are forbidden (including arrows → ← ↑ ↓, boxes ┌─┐└┘├┤│, dots • ◆ ●, bold ▶ ◀, etc.).
Allowed character set: letters, digits, Chinese characters, spaces, and - = | + * / \ < > ^ v [ ] ( ) { } . , : ; _ #。
Comparison table (left forbidden / right replacement):
┌ ┐ └ ┘ ├ ┤ ┬ ┴ ┼->+─ ━->-(or=table thick line)│ ┃->|→ ▶->->← ◀-><-↑->^or|^↓->vor|v◀─▶-><->or<--->●->*oro×->x
The nine examples below - each with a minimal usable ASCII diagram + one line ofdemonstration requirements. Which one to choose follows the shape of the root rank that comes through from the preceding process.
1. Two-axis / multi-axis coordinate system
Suitable for: root rank with several orthogonal, independent, slidable lines. The Fritz rocket geometry (vision clarity x current-state honesty) is this kind.
Demonstration requirements: the diagram must mark the coordinates of several concrete phenomena - for example, where "embrace" falls, where "greeting" falls, where "local reversal of entropy" falls. Only when readers see at a glance how this frame is used can they learn it. A frame with only coordinate axes and no demonstration points is an empty ornament.
Coordinate systems come in two grades, which determine what is drawn inside each quadrant：
-
Attribute-type coordinate system- the two axes are categorical/state attributes (such as light-heavy x emotional-rational, extroverted-introverted x intuitive-perceiving). Each quadrant is only a category label, with no dynamics of "what happens in this cell." Follow the requirements above - mark demonstration coordinates and examples at the four corners, and draw nothing more inside the frame.
-
Dynamics-type coordinate system- the two axes constitute dynamical dimensions, and the system in each quadrant evolves a unique temporal behavior. For example, potential functions (conservative-dissipative x single-multiple minima), phase transitions (temperature-pressure), games (zero-sum-non-zero-sum x one-shot-repeated). For this kind of coordinate system, flatly "sticking on labels" is a waste - each quadrant should be upgraded to aminiature ASCII picture with a rectangular border, presenting "how the system in this cell evolves" framed directly inside the box.
One-sentence discriminating question:Does the system in this cell have a segment of action that can be drawn? If yes - upgrade it to a dynamics picture; if no - fall back to an attribute label.
How to draw a dynamics picture：
- Each quadrant is a rectangular border (enclosed with
+ - |, like a framed small ASCII picture) - At the top inside the frame, draw "a miniature dynamics microcosm of that quadrant" - terrain / state / arrows / ball / curve, any choice, expressed in pure ASCII
- In the middle inside the frame, use one line of text to describe "what the system in this cell is doing"
- At the bottom inside the frame, mark that quadrant's label + 1-2 typical examples
- The four rectangles are separated by the coordinate axes (the axis lines pass between the rectangles)
- Pixel-level alignment is not required (Chinese character width issues are hard to fully solve), but the content structure of each box should be consistent
Example of a potential-function coordinate system:
单极小
|^
|
|
+--------------------------+ | +--------------------------+
| | | | |
| \________/ | | | \________/ |
| o <-> | | | o . |
| | | | |
| 唯一谷 球永恒摆动 | | | 唯一谷 球减速落底 |
| | | | |
| 保守 + 单极小 | | | 耗散 + 单极小 |
| [钟摆 / 谐振子] | | | [凸优化 / SVM] |
+--------------------------+ | +--------------------------+
|
保守 <-------------------------+-------------------------> 耗散
(能量守恒) | (能量被环境吸走)
|
+--------------------------+ | +--------------------------+
| | | | |
| \_/\_/\_/\_/\_/ | | | \_/\_/\_/\_/\_/ |
| <-o o o-> | | | o . o |
| | | | |
| 多谷 能量越脊穿梭 | | | 多谷 球落各谷不出 |
| | | | |
| 保守 + 多极小 | | | 耗散 + 多极小 |
| [三体 / 混沌系统] | | | [神经网络 / 蛋白质折叠] |
+--------------------------+ | +--------------------------+
|
|v
多极小
Readers can see at a glance: in each cell, "terrain + the ball's fate" is framed as a picture, and the system's mode of evolution is directly visible. This is what category labels cannot do.
2. Drilling profile
Suitable for: root rank with one layer supporting another, and the deepest one is a meta-proposition. Readers drill downward along the vertical direction.
+------------------------------------+
| 表层 现象 |
| [用户每天打开 App] |
+------------------------------------+
|
v
+------------------------------------+
| 中层 机制 |
| [推送 + 反馈 + 习惯回路] |
+------------------------------------+
|
v
+------------------------------------+
| 底层 元命题 |
| [多巴胺间歇强化] |
+------------------------------------+
Demonstration requirements: Each layer must be labeled with a concrete example, so the reader can see the penetration path "phenomenon -> mechanism -> meta-proposition" along the vertical direction. The deepest layer is the meta-proposition; don't leave it blank—leaving it blank means you haven't drilled all the way down.
3. Spectrum / Slider
Suitable for: the root rank is a line pulled at both ends, with cases falling at some point in the middle.
极端A <===================o===================> 极端B
[偏A 的特征] ^ [偏B 的特征]
|
案例 X 落在这
Demonstration requirements: The features of both ends must be labeled, and the position of at least one concrete case must be placed at some point in the middle. Having only the two ends with no case point = an empty display—the reader needs to see how the ruler measures.
4. Feedback Loop
Suitable for: the root rank is several loops that push each other, and positive and negative directions must be drawn.
A --(+)--> B
^ |
| |
(-) (+)
| |
| v
D <--(+)-- C
Demonstration requirements: Each arrow must be labeled + or -; the overall effect of positive loops (reinforcing, self-strengthening) and negative loops (balancing, self-inhibiting) must be noted beside the diagram—a loop diagram without positive/negative labels is as good as not drawn.
5. Chain / Steps
Suitable for: the root rank is stage progression, one segment after another in time or logic.
Chain horizontal arrangement:
A ---> B ---> C ---> D
[阶段名] [阶段名] [阶段名] [阶段名]
Steps vertical arrangement (rising step by step):
+----+
| D |
+----+
| C |
+----+
| B |
+----+
| A |
+----+
Demonstration requirements: Each segment must be named + labeled with transition conditions (what pushes A to B, what makes C rise to D). A chain without transition conditions = enumeration, not mechanism.
6. Tree Diagram
Suitable for: the root rank is one root splitting into many roots, and those many roots splitting further.
root rank
/ | \
A B C
/ \ | /|\
a b d x y z
Demonstration requirements: Each branching point must name the classification principle (by what the previous layer is cut into the next layer)—a tree that does not name the classification principle is muddled.
7. Network Diagram
Suitable for: the root rank is many-to-many interconnections, with no hierarchy.
A --------- B
|\ /|
| \ / |
| \ / |
| \ / |
| \ / |
| X |
| / \ |
| / \ |
| / \ |
| / \ |
D --------- C
Demonstration requirements: Each edge must have direction (if it has direction) and strength (thick line = / thin line -), and key nodes must be labeled. A pure mass of connections with no distinction of primary and secondary = a net, but not a skeleton.
8. Waveform / Oscillation Curve
Suitable for: the root rank is rise and fall, alternating rhythm (such as seasons, cycles, boom-bust, focus-distraction).
/\ /\ /\
/ \ / \ / \
/ \ / \ / \
/ \____/ \____/ \____
^ ^
谷 峰
[低谷期表现] [高峰期表现]
Demonstration requirements: Label at least one pair of concrete trough-peak examples, so the reader can see the state at each relative position within the cycle—a waveform without labeled positions is as good as a decorative line.
9. N-axis / Multi-slice
Suitable for: multidimensional classification with more than two axes (such as Y abstraction level x X distance level x Z time). 3D is hard to draw clearly; draw pairwise 2D slices:
切片一: 抽象度 x 远近度 切片二: 抽象度 x 时间
高抽象 高抽象
^ ^
| |
| |
近 <-----+-----> 远 过去 <---+---> 现在
| |
| |
v v
低抽象 低抽象
Demonstration requirements: Each slice must label at least one case and which quadrant it falls in; only after drawing all pair-wise slices is it complete. Missing one pair = leaving a blind spot.
Overlay type (root rank + viewfinder) draw two diagrams: The upper one is the skeleton of the root rank (using one of the nine above), and the lower one is the viewfinder overlaid on a certain line (with demonstration coordinates / demonstration points / demonstration paths labeled). Use a line to lead between the two diagrams—for example "expand here ->" or "zoom in again in this cell"—to tell the reader which cell of the first diagram the second one is embedded in. The reader sees the skeleton at a glance, and also sees that in one cell of the skeleton there is still a viewfinder instrument hidden; the worldview and the operating instrument enter the brain at the same time.
No rigid template. See the bone—mark the generator name, relationship direction, and positive/negative influence. The reader looks at the diagram once, and the skeleton enters the brain.
Output
- Get timestamp:
date +%Y%m%dT%H%M%Sanddate "+%Y-%m-%d %a %H:%M" - Write
~/Documents/notes/{时间戳}--{领域}的秩__rank.org - report file path to the user