Skip to content

curriculum(APP17): competency-derived semantic HTML/CSS and cumulative evidence - #163

Merged
christophersesugh merged 1 commit into
mainfrom
curriculum/app17-competency-first-reassessment
Nov 18, 2026
Merged

christophersesugh merged 1 commit into
mainfrom
curriculum/app17-competency-first-reassessment

Conversation

@christophersesugh

Copy link
Copy Markdown
Collaborator

M3 / APP17 — competency-first academic reassessment

1. Baseline and scope

2. Independent competency reconstruction and research

A learner entering with APP1's observed network-to-pixels/browser pipeline model must be able to author, inspect, debug and defend a meaningful, accessible, responsive, maintainable multi-page static information site in HTML/CSS. This requires actual semantic relationships, labels/forms, cascade reasoning, intrinsic layout, user/viewport preferences, asset delivery, negative-path investigation, and independent clean reproduction—not JavaScript/TypeScript, React, authentication, databases, public hosting or production infrastructure.

Distinct ownership: APP1 explains HTML/CSS parsing and browser behavior; APP17 authors valid semantic documents and robust styles; APP2 later builds framework-independent dynamic DOM/typed interaction; APP20 specializes in React. APP17 must not silently assume JS or require backend persistence for native forms.

Disciplinary/technical comparison (inform source decisions, not a fixed course/module count):

3. Outcome dispositions

Retain APP17-O1–O12 as published: observable semantic markup, navigation/media, native forms, cascade, visual readability, layout, responsiveness, accessibility, resource-aware assets, maintainable source, failure diagnosis, integrated static-site defense. All remain independently assessable and covered. No artificial outcome consolidation/renumbering or false equivalence to old learner records.

Strengthen an existing assessment gap for O4: its published outcome includes scoping proximity and precedence; the old AS3 evidence specified origin/importance/layers/specificity/inheritance/source order but did not explicitly test a scoped-rule tie. Added a supplied-scenario case so the ability is evaluated without assuming browser @scope support. Clarified normal vs important layer ordering as well.

Strengthen O7/O8 evaluation: 200% zoom alone does not demonstrate the 320-CSS-pixel reflow condition; a separate viewport-based check and legitimate data-table exception are now explicit. Clarify 4.5:1 normal-text, 3:1 large-text and applicable 3:1 non-text indicators without representing these samples as full WCAG conformance.

4. Module-by-module lesson audit and actual decisions

Module Entries Reasoned disposition
M1 HTML source/authoring 6 → 4 Combine isolated orientation/history with first authoring action; turn APP1 parsing recap into a source-vs-DOM validation/repair objective; retain metadata, meaningful hierarchy
M2 semantic navigation/media 7 → 6 Apply rather than reteach URL/fragment foundations; distinguish landmarks, links, image alternatives, data tables, local-media fallback and observed programmatic relationships
M3 native forms 7 → 5 Consolidate controls/submission and privacy fixture requirements; keep separate labels/group help, native constraints, keyboard correction and nontransmission
M4 CSS cascade 7 → 6 Organize mechanism by actual cascade decision order; retain origin, importance, layer, specificity, scoping, inheritance, state and a contradiction diagnostic
M5 visual design/geometry 7 → 7 Retain: typography, contrast, box geometry, flow/logical properties, intrinsic sizing, custom-property tokens and preference stress have distinguishable mechanisms and error signals
M6 Flow/Flex/Grid 7 → 6 Keep real layout models and min-content/positioning faults; move long/localized-content stress to responsive evaluation, where failure evidence is decisive
M7 responsiveness 7 → 6 Unify content-derived breakpoint/fluid sizing; distinguish viewport vs container queries and reflow vs text resize; retain preference and unknown-content failure tests
M8 accessibility evaluation 8 → 5 Change from repeated beginner labels/keyboard/contrast lessons into a genuinely independent scoped audit, negative-case repair and evidence-limits stage
M9 assets/maintenance 6 → 5 Separate semantic alternatives (M2) from actual resource-format, responsive-image, reserved-size, fallback and refactoring/regression work
M10 integrated site 7 → 5 Convert pseudo-independent implementation checklist into one cumulative product, adversarial bounded change, clean reproduction and defense; no replay of five old submissions

Observed structural consequence: 10 coherent modules retained; 69 → 55 focused lesson entries, not because 55 is a target. M5 intentionally remains seven lessons. Nothing requires a standard number per module.

5. Outcome → assessment → evidence audit

Retain all five stable assessment IDs, but clarify their different decisions on the same evolving artifact rather than five independent builds or repetitive evidence packets:

  • APP17-AS1 → O1/O2/O11: establish two linked pages, meaningful document/figure/table semantics, validator + actual DOM observations and two genuine repairs.
  • APP17-AS2 → O3/O8/O11: extend that site with a native form, accurate labels/groups, valid/invalid keyboard recovery, and no required data transmission. Browser validation does not establish backend security or assistive-technology-announced dynamic errors.
  • APP17-AS3 → O4/O5/O6/O10/O11: add CSS/layout, predict and inspect cascade including origin/layer/importance, supplied scoped-rule case, reproduce and repair overflow and confirm style refactor.
  • APP17-AS4 → O2/O6/O7/O8/O9/O11: evaluate that same site for reflow, separate zoom, accessible focus, meaning and contrast, image loading/geometry and supported/fallback CSS, repairing and rechecking actual failures.
  • APP17-AS5 → O1–O12: evaluate one at-least-three-page reproducible static site, previously accepted evidence references, a new bounded unfamiliar content/accessibility/CSS/asset defect with regression verification, and a real technical defense. Passing AS5 does not erase a missing AS1–AS4 competency.

Every required task publishes Work, Criteria and Evidence; all twelve outcome IDs are linked to actual required assessments. A private grading service, public hosting, real personal-data collection, mandatory JS runtime, or inaccessible screen-reader study is not needed. Open issues and limitations must be distinguished from verified behavior; blockers to the public completion criteria cannot pass simply by being documented.

6. Cross-course effects, revision and access

  • APP1's merged foundation remains the sole direct prerequisite; APP2/APP20 inherit authored HTML/CSS capability via the declared graph. APP9 graduation prerequisites are unchanged.
  • HTML/CSS visual semantics are not assumed equal to dynamic DOM application competence; no course responsibility transferred to an elective.
  • No mandatory form submission, user-data collection, public deployment, paid tool or JavaScript/backend implementation introduced.
  • No change to stable outcome or assessment IDs, provisional 10–12-week duration, course identity or draft readiness. Historical completion requirements must remain traceable to their applicable source revisions; the cumulative evidence clarification must not retroactively regrade old learners.
  • No APP catalog change is necessary because metadata/prerequisites remain identical. APP catalog's separate M0 F4 entry-equivalence ambiguity and APP15 Rust preparation remain outstanding for their own review.

7. Findings, severity and resolutions

  • Major (resolved): APP17/M8 duplicated introductory labels, keyboard and contrast in M1–M3/M5/M7 without new assessed depth. Refocused on independent criteria, negative cases, repairs and assessor-limit reasoning.
  • Major (resolved): APP17-AS1–AS5 did not consistently establish one cumulative artifact; five independent site/evidence burdens were a credible reading despite a late permissive sentence. Made reuse operational in every Work/Evidence contract, with separately decidable criteria.
  • Major (resolved): APP17-O4/AS3 mentioned scoped cascade proximity as an outcome but not as an explicit assessment decision. Added a labeled supplied tie case to measure it while keeping browser support assumptions honest.
  • Major/minor (resolved): APP17/M7, AS4 conflated responsive-width and 200%-zoom evidence; added explicit 320 CSS px reflow, qualified exceptions, and applicable non-text contrast/focus criteria.
  • Minor (resolved): isolated history/onboarding and duplicate APP1 parsing/URL explanations. Consolidated into first implementation/diagnostic work without removing required foundations.
  • Retain (justified): ten modules remain due to distinct learning models, with M5 keeping seven lessons. No count-based rule, forced uniformity, or artificial variety.
  • Outstanding: draft study-hour estimates, actual cross-browser/assistive-technology behavior, real learner time/effectiveness and independent teaching efficacy need future genuine evidence, not guessed results.

8. Verification and limits

  • Independently refetched committed branch source; pinned comparison against 23567c8 shows only curriculum/APP/APP17.md changed.
  • Verified ten sequential modules; required Outcomes and Lessons subsections; ordered lesson numbering in every module; 55 actual instructional entries; twelve declared outcome IDs, five stable required assessment IDs, and Work/Criteria/Evidence fields for each. Every required outcome appears in summative assessment definitions; no static inconsistency detected.
  • Rechecked relevant APP1/APP2 course boundary and program catalog. Compared standard-specific WCAG, CSS and HTML assertions with the linked primary sources.
  • Not run: full repository Python validator/regression suite, CircleCI, browser/AT learner experiments, full link tests, actual independent assessor exercises. The local checkout environment cannot resolve github.com. Static checks are not a claim CI passed. Review PR checks and curriculum validity before merging.
  • No learner-tested time estimate, full WCAG conformance, independent certification or upgraded tested/verified status is asserted.

9. Decision and merge protocol

Proposed for maintainer review and CI, not already academically verified. Merge only after review and satisfactory checks. Keep APP17's branch separate from M1 SEF, SYS, active AI integration and temporary audit protocol. Next APP M3 review after merge: APP18 — JavaScript Programming.

Branch: curriculum/app17-competency-first-reassessment. PR base: main. Temporary protocol: unmerged PR #150 stays open until all 42 course reviews and cross-program closure.

…umulative evidence

Signed-off-by: Christopher S. Aondona <christohybrid185@gmail.com>
@christophersesugh
christophersesugh force-pushed the curriculum/app17-competency-first-reassessment branch from 77ab95d to 8d5432f Compare November 18, 2026 12:47
@christophersesugh
christophersesugh merged commit c0a0142 into main Nov 18, 2026
7 checks passed
@christophersesugh
christophersesugh deleted the curriculum/app17-competency-first-reassessment branch November 18, 2026 12:48
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant