Framing through autoethnographic elaboration of the three principles

This chapter applies the supplementary autoethnographic posture (established in Chapter B) to the three curriculum-design principles the main document develops: modularity, learner choice, and continuous feedback. The chapter does not introduce three competing findings. It elaborates on each of the three principles by drawing on the instructor's reflexive position to surface what the DBR analysis of the main document does not. The autoethnographic supplement deepens; it does not replace.

D.1.1 The relationship between the three principles and the autoethnographic elaborations

The main document presents the three curriculum-design principles as the iterative-refinement work's primary outputs. Each principle is supported by evidence across the iterations; each is actionable for adopters; each is empirical in the DBR sense. What the principles do not by themselves convey is what the instructor's reflexive position on her own practice surfaces when she reads the same iterations autoethnographically.

This chapter provides that reflexive layer for each principle in turn. The three principles and their autoethnographic elaborations are:

Each elaboration carries its own sub-claim (named in the right column). The sub-claims are not separate findings competing with the principles; they are concrete patterns that the autoethnographic analysis surfaces within each principle.

D.1.2 Anderson's theoretical analysis criterion as the lens

Anderson's (2006) fifth criterion for analytic autoethnography is commitment to theoretical analysis. The autoethnographer's insider position is not used to produce evocative description; it is used to generate, refine, or extend theoretical understanding of broader social phenomena. This chapter applies the criterion to each of the three principles, drawing theoretical depth from the practitioner's reflexive vantage that the DBR analysis of the main document does not surface.

What the criterion governs is how each elaboration counts as theoretical analysis. An elaboration must name a generalizable pattern that the insider position made possible to see, support that pattern against multiple independent sources, and have implications beyond the four iterations for how scholarship treats the phenomenon.

The three sub-claims meet these tests. Compression-as-curriculum-maturation is a pattern about how the modular architecture matured by distillation rather than by accretion; the corpus supports it across the four iterations. Multi-channel teaching practice is a pattern about how learner choice operated across eight contexts of varying duration, audience, and depth; the corpus documents it across more than two and a half years. Hallucination-as-pedagogy is a pattern about how moment-to-moment continuous feedback with the tools became the instructor's pedagogical practice; the corpus supports it across instructor reflection, public-facing writing, and live workshop delivery.

The label "hallucination-as-pedagogy" is my analytic coinage in this dissertation. The corpus contains the phenomenon — me reframing the tool's hallucinations as teachable moments across multiple registers from January 2024 through September 2025 — but the label itself emerges here, in the analytic work, not in the iterations' own materials. The same is true of the other two sub-claim names: "compression-as-curriculum-maturation" and "multi-channel teaching practice" are coined in this dissertation as analytic instruments rather than recovered from earlier sources.

D.1.3 Cross-validation as the rigor move

Each elaboration is anchored across multiple artifact types from multiple points in the timeline, with convergence treated honestly. The compression-as-maturation sub-claim anchors across the four iteration artifact sets; the multi-channel sub-claim anchors across eight documented contexts (whose heterogeneity is acknowledged in §D.3); the hallucination-as-pedagogy sub-claim anchors across four sources of which one (KT-THEMES-C5) is near-independent and three are Larissa's own surfacings across registers (§D.4).

The convergence across artifact types and across time is what distinguishes a defensible autoethnographic elaboration from an opportunistic observation. The dissertation does not claim more than the convergence supports.

D.1.4 What this chapter does not claim

I name three things this chapter does not claim, so that the scope of the autoethnographic elaborations is clear.

First, the elaborations do not replace the three principles. The DBR analysis of the main document is unaffected by what this chapter does; what this chapter does is sit alongside it and extend the analytic reach within each principle.

Second, the elaborations do not claim to be exhaustive. Analytic autoethnography of a multi-year multi-channel pioneer practice could surface other patterns within each principle that this dissertation does not develop. The three I elaborate are the strongest cross-validated patterns and the most useful for engineering education and HCI as scholarly fields.

Third, the elaborations do not claim that the patterns transfer automatically to other pioneer instructors of other emerging-technology curricula. Each pattern is a claim about my pioneer practice at the generative-AI site between 2023 and 2025. Whether the patterns travel to other sites is an empirical question for subsequent scholarship.

Sections D.2 through 4.4 develop each elaboration in turn.