|
1 | 1 | # Mechanics |
2 | 2 |
|
3 | | -Mechanics are the **structural reinforcements** |
4 | | -that turn organizational capabilities into lived reality. |
| 3 | +This directory contains the **canonical mechanics of Protopipe**. |
5 | 4 |
|
6 | | -They are neither abstract principles |
7 | | -nor concrete tool implementations. |
| 5 | +Mechanics are the structural core of Protopipe. |
| 6 | +They encode **non-negotiable consequences** that shorten feedback, |
| 7 | +reduce systemic risk, and make certain forms of work unavoidable |
| 8 | +while making others impossible or costly. |
8 | 9 |
|
9 | | -Mechanics exist to make |
10 | | -culture, feedback loops, and responsibility |
11 | | -**operational and unavoidable**. |
| 10 | +Mechanics are **not solutions**, **not patterns**, and **not best practices**. |
| 11 | +They are enforced trade-offs. |
12 | 12 |
|
13 | 13 | --- |
14 | 14 |
|
15 | | -## What Mechanics Are |
| 15 | +## What a Mechanic Is |
16 | 16 |
|
17 | | -Mechanics are repeatable, structural means |
18 | | -that enforce desired behavior in everyday work. |
| 17 | +A mechanic exists to enforce a structural consequence. |
19 | 18 |
|
20 | | -They answer the question: |
| 19 | +A mechanic: |
21 | 20 |
|
22 | | -> **How do we make the right behavior |
23 | | -> the easiest behavior — by design?** |
| 21 | +- shortens feedback loops |
| 22 | +- removes options instead of adding features |
| 23 | +- encodes a deliberate trade-off |
| 24 | +- forces behavior through structure, not discipline |
| 25 | +- hurts a little — by design |
24 | 26 |
|
25 | | -Mechanics operate at the intersection of: |
26 | | -- architecture |
27 | | -- process |
28 | | -- feedback |
29 | | -- organizational design |
| 27 | +If a rule can be ignored without consequence, |
| 28 | +it is not a mechanic. |
30 | 29 |
|
31 | | -They are simple to explain, |
32 | | -but difficult to implement holistically. |
| 30 | +--- |
| 31 | + |
| 32 | +## The Core Principle |
| 33 | + |
| 34 | +> **Mechanics shorten feedback by enforcing consequences.** |
| 35 | +> They do not describe solutions. |
| 36 | +> They remove options. |
| 37 | +
|
| 38 | +This principle is the foundation for all mechanics in Protopipe. |
33 | 39 |
|
34 | 40 | --- |
35 | 41 |
|
36 | | -## What Mechanics Are Not |
| 42 | +## Dependency Direction (Non-Negotiable) |
| 43 | + |
| 44 | +Mechanics sit in a strict dependency chain: |
37 | 45 |
|
38 | | -Mechanics are **not**: |
| 46 | +``` |
| 47 | +Problem |
| 48 | +↓ |
| 49 | +Use-Case |
| 50 | +↓ |
| 51 | +Mechanic |
| 52 | +↓ |
| 53 | +Implementation |
| 54 | +``` |
39 | 55 |
|
40 | | -- features |
41 | | -- tools |
42 | | -- platforms |
43 | | -- frameworks |
44 | | -- methodologies |
45 | 56 |
|
46 | | -Tools may implement mechanics. |
47 | | -Frameworks may describe them. |
48 | | -But neither defines them. |
| 57 | +- Problems describe tensions and pain. |
| 58 | +- Use-Cases describe desired, observable outcomes. |
| 59 | +- Mechanics enforce structural consequences. |
| 60 | +- Implementations are replaceable realizations. |
49 | 61 |
|
50 | | -A mechanic exists |
51 | | -even if the underlying tooling changes. |
| 62 | +**Noesis ends at the Mechanic.** |
| 63 | +Implementations must never flow back. |
52 | 64 |
|
53 | 65 | --- |
54 | 66 |
|
55 | | -## Why Mechanics Matter |
| 67 | +## Naming Convention (Mandatory) |
56 | 68 |
|
57 | | -Most organizational change fails |
58 | | -not because of missing intent, |
59 | | -but because behavior is optional. |
| 69 | +The **name of a mechanic is part of the mechanic**. |
| 70 | +It expresses the enforced trade-off explicitly. |
60 | 71 |
|
61 | | -Culture is declared. |
62 | | -Feedback is discussed. |
63 | | -Responsibility is assigned. |
| 72 | +All mechanics **must** use **one of the following three naming forms**. |
| 73 | +No other forms are allowed. |
64 | 74 |
|
65 | | -But nothing in the system |
66 | | -*forces* these concepts to matter. |
| 75 | +### 1. `X over Y` — Explicit Trade-off |
67 | 76 |
|
68 | | -Mechanics close this gap. |
| 77 | +Use this form when a mechanic enforces a clear preference |
| 78 | +between two competing approaches. |
69 | 79 |
|
70 | | -They embed: |
71 | | -- feedback into delivery |
72 | | -- responsibility into boundaries |
73 | | -- learning into execution |
74 | | -- steering into daily work |
| 80 | +**Meaning** |
| 81 | +X is structurally preferred. |
| 82 | +Y is still possible, but discouraged or costly. |
75 | 83 |
|
76 | | -Without relying on discipline or heroics. |
| 84 | +**Examples** |
77 | 85 |
|
78 | | ---- |
| 86 | +- *Composability over Convenience* |
| 87 | +- *Observation over Orchestration* |
| 88 | +- *Eventual Consistency over Global Process Engines* |
79 | 89 |
|
80 | | -## Mechanics and Culture |
| 90 | +--- |
81 | 91 |
|
82 | | -Culture does not scale through values. |
83 | | -It scales through constraints and feedback. |
| 92 | +### 2. `X by Default, Y by Exception` — Default with Justification |
84 | 93 |
|
85 | | -Mechanics are the **technical expression of culture**. |
| 94 | +Use this form when a behavior is allowed, |
| 95 | +but only with explicit justification. |
86 | 96 |
|
87 | | -They determine: |
88 | | -- whether problems surface early or late |
89 | | -- whether learning is optional or mandatory |
90 | | -- whether responsibility is visible or diffused |
91 | | -- whether decisions are revisited based on evidence |
| 97 | +**Meaning** |
| 98 | +X is the assumed baseline. |
| 99 | +Y requires conscious, explicit reasoning. |
92 | 100 |
|
93 | | -If a mechanic contradicts the culture, |
94 | | -the mechanic wins. |
| 101 | +**Examples** |
95 | 102 |
|
96 | | -Always. |
| 103 | +- *Eventual Consistency by Default, Atomicity by Exception* |
| 104 | +- *Open Source by Default, Closed Source by Justification* |
97 | 105 |
|
98 | 106 | --- |
99 | 107 |
|
100 | | -## Mechanics and Feedback Cycles |
| 108 | +### 3. `No X without Y` — Hard Constraint |
101 | 109 |
|
102 | | -The strongest impact of mechanics |
103 | | -is on feedback speed and quality. |
| 110 | +Use this form when a behavior must never occur |
| 111 | +unless a prerequisite is fulfilled. |
104 | 112 |
|
105 | | -Good mechanics: |
106 | | -- shorten feedback loops |
107 | | -- localize feedback to the point of change |
108 | | -- make cause and effect observable |
109 | | -- reduce the need for coordination and escalation |
| 113 | +**Meaning** |
| 114 | +X is forbidden unless Y is present. |
110 | 115 |
|
111 | | -Poor or missing mechanics result in: |
112 | | -- late feedback |
113 | | -- global impact |
114 | | -- defensive behavior |
115 | | -- political decision-making |
| 116 | +**Examples** |
116 | 117 |
|
117 | | -Feedback is not a reporting problem. |
118 | | -It is a structural one. |
| 118 | +- *No Process Step without Testable Outcome* |
| 119 | +- *No Cross-Context Read without Event-Carried State* |
| 120 | +- *No Release without Observable Feedback* |
119 | 121 |
|
120 | 122 | --- |
121 | 123 |
|
122 | | -## Position in the Protopipe Layer Model |
| 124 | +### Naming Guidance (Important) |
| 125 | + |
| 126 | +- If the name does **not** clearly express loss or constraint, |
| 127 | + it is not a valid mechanic name. |
| 128 | +- If the name sounds comfortable or neutral, |
| 129 | + it is likely a pattern or guideline — not a mechanic. |
| 130 | +- The name should make the trade-off obvious |
| 131 | + even without reading the document. |
123 | 132 |
|
124 | | -Mechanics sit **below the narrative layer** |
125 | | -and **above concrete implementations**. |
| 133 | +If the name does not “hurt” a little, |
| 134 | +rework it. |
| 135 | + |
| 136 | +--- |
126 | 137 |
|
127 | | -The overall layering is: |
| 138 | +## Mechanics vs. Patterns vs. Features |
128 | 139 |
|
129 | | -1. **Problems** |
130 | | - Systemic blockers that prevent learning and delivery. |
| 140 | +To avoid dilution, the following distinctions apply: |
131 | 141 |
|
132 | | -2. **Use-Cases** |
133 | | - Observable outcomes once problems are addressed. |
| 142 | +- **Mechanic** |
| 143 | + Enforces a consequence. Removes options. |
134 | 144 |
|
135 | | -3. **Capabilities (APDP)** |
136 | | - Organizational abilities required to sustain these outcomes. |
| 145 | +- **Pattern** |
| 146 | + Suggests a recurring solution. Optional. |
137 | 147 |
|
138 | | -4. **Mechanics** |
139 | | - Structural reinforcements that make these capabilities real. |
| 148 | +- **Feature** |
| 149 | + Describes functionality or capability. |
140 | 150 |
|
141 | | -5. **Implementations** |
142 | | - Concrete technical realizations (tools, platforms, code). |
| 151 | +Only mechanics belong in this directory. |
143 | 152 |
|
144 | | -Mechanics translate **capabilities into system behavior**. |
| 153 | +Patterns, features, tools, and implementations |
| 154 | +must live elsewhere (e.g. `dist/`, reference implementations). |
145 | 155 |
|
146 | 156 | --- |
147 | 157 |
|
148 | | -## Why Mechanics Are Hard |
149 | 158 |
|
150 | | -Mechanics are difficult because they: |
| 159 | +## Reference Implementations (Dogfooding) |
| 160 | + |
| 161 | +Every mechanic **may** provide a reference implementation. |
| 162 | + |
| 163 | +Reference implementations are: |
| 164 | + |
| 165 | +- explicitly **non-canonical** |
| 166 | +- intentionally minimal |
| 167 | +- used to **experience the enforced constraints in practice** |
| 168 | +- a means to apply the “eat your own dogfood” principle |
| 169 | + |
| 170 | +Their purpose is **not** to prove technical correctness, |
| 171 | +completeness, or scalability. |
| 172 | + |
| 173 | +Their sole purpose is to validate the **consequence** of a mechanic: |
| 174 | + |
| 175 | +- Are options actually removed? |
| 176 | +- Is the constraint felt in daily work? |
| 177 | +- Does feedback become shorter or clearer? |
| 178 | +- Is the trade-off tolerable? |
| 179 | + |
| 180 | +A mechanic candidate **must not** be promoted to approved status |
| 181 | +unless its constraints have been experienced |
| 182 | +through a reference implementation. |
| 183 | + |
| 184 | +Dogfooding is **necessary, but not sufficient**. |
| 185 | + |
| 186 | +If a reference implementation fails, |
| 187 | +this does **not** automatically invalidate the mechanic. |
| 188 | +However, the mechanic **must be questioned**. |
151 | 189 |
|
152 | | -- cut across team and system boundaries |
153 | | -- expose hidden assumptions |
154 | | -- change power and responsibility distribution |
155 | | -- require consistent application |
156 | | -- cannot be “half implemented” |
| 190 | +Implementation must never justify the mechanic. |
| 191 | +The mechanic must justify the implementation. |
157 | 192 |
|
158 | | -They often fail |
159 | | -not due to technical complexity, |
160 | | -but due to organizational resistance. |
161 | 193 |
|
162 | 194 | --- |
163 | 195 |
|
164 | | -## How to Read This Section |
| 196 | +## Candidate vs. Approved Mechanics |
165 | 197 |
|
166 | | -Each mechanic described here: |
| 198 | +Not every good idea is immediately canonical. |
167 | 199 |
|
168 | | -- links back to one or more APDP capabilities |
169 | | -- explains the behavioral effect it enforces |
170 | | -- avoids prescribing specific tools |
171 | | -- highlights trade-offs and failure modes |
| 200 | +Mechanics may exist in two states: |
172 | 201 |
|
173 | | -This section is **product knowledge**, not narrative. |
| 202 | +- **Candidates** |
| 203 | + Under evaluation, consolidation, or refactoring |
174 | 204 |
|
175 | | -It is intended for: |
176 | | -- architects |
177 | | -- senior engineers |
178 | | -- technical leadership |
179 | | -- presales and solution design |
| 205 | +- **Approved** |
| 206 | + Stable, invariant, and accepted as canonical |
180 | 207 |
|
181 | | -Not for pitching. |
| 208 | +Candidates are expected to change. |
| 209 | +Approved mechanics are expected to endure. |
182 | 210 |
|
183 | 211 | --- |
184 | 212 |
|
185 | | -## Summary |
| 213 | +## Final Rule |
186 | 214 |
|
187 | | -Mechanics are the **leverage point** of Protopipe. |
| 215 | +If a mechanic does not: |
188 | 216 |
|
189 | | -They make culture executable. |
190 | | -They make feedback unavoidable. |
191 | | -They make responsibility structural. |
| 217 | +- shorten feedback |
| 218 | +- enforce a trade-off |
| 219 | +- remove options |
| 220 | +- create visible consequences |
192 | 221 |
|
193 | | -Without mechanics, |
194 | | -capabilities remain aspirational. |
| 222 | +then it does not belong here. |
195 | 223 |
|
196 | | -With mechanics, |
197 | | -behavior changes — even under pressure. |
| 224 | +Be strict. |
| 225 | +Protopipe gains strength through constraint. |
198 | 226 |
|
0 commit comments