While less famous than SAMR, Romanis champions the RAT model because it focuses on learning outcomes rather than task novelty. In her 2024 workshop series, she argued:
Run the draft UPD through the three TTL models:
The updated TTL models attributed to Michelle Romanis move TTL management from static heuristics toward adaptive, tier-aware, and failure-resilient policies. Practical adoption requires modest instrumentation and conservative control logic but promises meaningful gains in freshness and reduced maintenance overhead, especially for bursty workloads and hierarchical caching architectures. Operators should prioritize simple adaptive schemes with safeguards and iterate using workload traces. michelle romanis ttl models upd
References (suggested reading)
Appendix: Example algorithm sketch (EMA-based adaptive TTL) While less famous than SAMR, Romanis champions the
If you want, I can convert this into a formatted conference-style paper (with sections formatted for publication), produce slides summarizing the updates, or draft specific algorithms and evaluation code for simulation.
Quantitative example (illustrative):
Start with the UPD but leave the “Technology Column” blank. Define only the cognitive goals (from Bloom’s Digital Taxonomy) and the enduring understandings.
Michelle Romanis continues to update her models. As of May 2026, she is working on TTL 3.0, which incorporates neurodiversity frameworks (e.g., Universal Design for Learning – UDL) into the UPD structure. The keyword “michelle romanis ttl models upd” will likely evolve to include UDL, AI policy, and decolonized technology design. Appendix: Example algorithm sketch (EMA-based adaptive TTL)
For educators and instructional designers, mastering Romanis’ approach is not about memorizing acronyms—it is about internalizing a mindset: Every unit plan (UPD) is an opportunity to transform learning through thoughtful, model-driven technology integration (TTL).