Revise tutorials/OUTLINE.md as the Priority 2 course design plan - #13
Merged
Merged
Conversation
The outline previously described the delivered ten-notebook v0.1 series and had drifted: it omitted tutorial 11, still marked tutorial 07 as blocked on InSAR covariance support that has since landed, and referenced pre-1.6 names (Fault.centers, top-level geodef.lcurve/model_covariance aliases). Rewrite it as the design plan for the v0.2 learning release (PLAN.md Priority 2): - Textbook-depth register: a chapter anatomy (objectives, prerequisites, time estimates, developed theory, checkpoints, common mistakes, recap, further reading) with 45-90 minute chapters and depth in prose, not compute. - New chapters: 00 preflight (scientific Python for geodesy, closing with the five-minute quickstart preview) and topics 12-16 (triangular faults, interseismic coupling, model misspecification, Bayesian inversion and prior sensitivity, posterior diagnostics). - Per-chapter revision deltas for 01-11: derivations to add (WLS/SVD, filter factors, ABIC, C_m and R, variable projection), migration to the post-1.6 module-path API and named result views, and the atomic top-level alias removal from 2.2. - The 2.1 'start here' layer (quickstart, workflow map, glossary, decision guides) with homes and linking rules; the proposed geodef.synthetic scenario-builder and checkerboard/spike helper module; separately published, CI-executed solution notebooks. - Settled design decisions (stable numbering, explicit-then-builder scenario delivery) and open questions (docs file names, synthetic module naming, backslip-convention timing, CI wall-clock budget, checkpoint answer mechanics, one-vs-two Bayesian chapters, glossary single-sourcing). Record the outline as the phase's design reference under Priority 2 in PLAN.md.
…tled decisions Incorporate the maintainer's resolutions of the outline's seven open questions: - Merge regularization + choosing-lambda into one chapter (04) and design the Bayesian material as a single chapter (14: priors, sampling, and diagnostics), reducing the course to fifteen notebooks (00-14) with a one-time renumbering executed via git mv during the 2.2 rewrite; each merged chapter records a fallback split seam. - Combine the workflow map and decision guides into one docs/workflow.md; docs layer is three files (quickstart, glossary, workflow). - Approve geodef.synthetic; recovered-versus-input comparison becomes geodef.invert.compare with no synthetic alias (one name, one home). - Settle the backslip/coupling convention: per-station block correction into a declared (normally overriding-plate-fixed) frame using the same Euler poles as the per-patch plate rates; backslip anti-parallel to plate motion with non-negative amplitude bounded by the plate rate; coupling in [0, 1]. Record the negative-dip-slip consequence for raw components and the 4.4 boundary. - Adopt the CI budget (~5 min base path, ~10 min gated, solutions in the same job behind a pytest marker) plus a standing goal to shorten the existing JAX/Bayesian test modules. - Adopt collapsed <details> blocks for checkpoint answers. - Make the glossary the source of truth for notation once created, and add a lifecycle note: the outline is transient and retires to git history after the v0.2 course ships. Update the Priority 2 pointer in PLAN.md to match.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The outline previously described the delivered ten-notebook v0.1 series and
had drifted: it omitted tutorial 11, still marked tutorial 07 as blocked on
InSAR covariance support that has since landed, and referenced pre-1.6 names
(Fault.centers, top-level geodef.lcurve/model_covariance aliases).
Rewrite it as the design plan for the v0.2 learning release (PLAN.md
Priority 2):
time estimates, developed theory, checkpoints, common mistakes, recap,
further reading) with 45-90 minute chapters and depth in prose, not
compute.
the five-minute quickstart preview) and topics 12-16 (triangular faults,
interseismic coupling, model misspecification, Bayesian inversion and
prior sensitivity, posterior diagnostics).
filter factors, ABIC, C_m and R, variable projection), migration to the
post-1.6 module-path API and named result views, and the atomic top-level
alias removal from 2.2.
guides) with homes and linking rules; the proposed geodef.synthetic
scenario-builder and checkerboard/spike helper module; separately
published, CI-executed solution notebooks.
scenario delivery) and open questions (docs file names, synthetic module
naming, backslip-convention timing, CI wall-clock budget, checkpoint
answer mechanics, one-vs-two Bayesian chapters, glossary single-sourcing).
Record the outline as the phase's design reference under Priority 2 in
PLAN.md.