Logos52

Skin

4 min read

Skin

Part of Deep Processing

The rough map has to be cleaned before it can reliably support retrieval. Skin is the third step of Bear Hunter System.

The practical goal: consolidate, integrate, simplify, and decide what the final chunk structure should look like.

How To Use Skin

1. Step Back From The Rough Map

Do not start by adding more information.

First ask:

  • What is the main backbone?
  • Which chunks are overloaded?
  • Which chunks are weak?
  • Which arrows matter?
  • Which parts feel like copied source order?
  • Which parts would be hard to reconstruct from memory?

2. Clean The Backbone

The backbone is the main flow of the topic.

It should show the largest chunks and how they relate. If the backbone is unclear, the details will stay hard to retrieve.

Skin the backbone before fixing details.

3. Apply The 2-4 Rule

Use this as a practical chunking check:

  • aim for at least two meaningful branches from a node;
  • avoid more than four branches from a node;
  • split nodes with too many branches;
  • merge nodes that only create single-node chains.

This keeps the map simple enough to hold in working memory.

4. Remove Spiderwebbing

Spiderwebbing happens when arrows become chaotic because the chunks were not formed from relationships.

To repair it:

  • identify the relationships causing the messy arrows;
  • ask whether related concepts should be closer;
  • create a new chunk if several concepts share the same relationship;
  • remove arrows that do not matter enough to test;
  • rebuild from a flat concept list if the source hierarchy is biasing the map.

The goal is cleaner relationships.

5. Rename Chunks

Bad chunk names are labels copied from the source.

Good chunk names explain function or importance.

Ask:

  • What job does this chunk do?
  • Why are these ideas grouped?
  • What relationship holds them together?
  • Would this label help me reconstruct the topic later?

6. Turn The Map Into Retrieval Targets

After Skin, create prompts:

  • Reconstruct the whole map.
  • Explain each major chunk.
  • Explain the relationship between major chunks.
  • Rebuild one overloaded area from memory.
  • Compare two nearby concepts.
  • Explain why the final chunk structure makes sense.

Skin is incomplete until the map can be tested.

When To Skin

Best timing:

  • same day after the main learning event;
  • end of the week for a full week’s material;
  • after several Aim-Shoot cycles;
  • before starting SIR;
  • after retrieval reveals structural weakness.

If you are busy, do a 10-minute Skin pass. Fix the biggest structural problem first.

What Skin Should Feel Like

Good signs:

  • the topic becomes simpler;
  • chunk labels improve;
  • arrows decrease but meaning increases;
  • overloaded nodes become sub-chunks;
  • single-node chains disappear;
  • retrieval prompts become obvious.

Warning signs:

  • you only make the map prettier;
  • you add more detail instead of restructuring;
  • you keep the source order because it feels safe;
  • the map still cannot be reconstructed;
  • you avoid moving chunks because you already spent time drawing them.

Practical Quality Checklist

Before calling Skin done:

  • Is the backbone obvious?
  • Are the major chunks intuitive?
  • Does each major chunk have a reason to exist?
  • Are there overloaded nodes with 5+ branches?
  • Are there single-node chains?
  • Are important relationships visible?
  • Can I brain dump the map closed book?
  • Do I know what to retrieve next?

Common Fixes

ProblemFix
Too many branchesSplit into sub-chunks.
Single-node chainMerge, rename, or move the node.
Chaotic arrowsRe-chunk around the relationships causing the chaos.
Vague labelRename by function or importance.
Source-order biasFlatten the list, then rebuild by relationships.
Pretty but weakBrain dump from memory and repair what collapses.

Output

Skin should produce:

  • a cleaned map;
  • final or near-final chunk structure;
  • clearer labels;
  • reduced clutter;
  • visible backbone;
  • retrieval prompts;
  • a short gap list.

Stopping Rule

Stop Skin when the map is good enough to retrieve from.

Do not keep polishing. If the map can support SIR, move into retrieval. Retrieval will reveal the next repair.

How It Should Feel

Skin should feel like reducing the mental load of the map. The artifact should become easier to reconstruct, not merely neater.

Good signs:

  • vague chunks get renamed;
  • overloaded branches are split;
  • decorative arrows disappear;
  • the main structure becomes obvious;
  • and you can imagine using the result for retrieval.

Warning sign: Skin has become cosmetic when the map looks cleaner but is not easier to explain from memory.

Open Questions

  • What is the user’s stopping rule for a Skin pass?

Graph View

Backlinks