In your section on “which recommendations are most worthwhile?” you mention using the ITN framework. While this is probably makes for efficient communication since many readers are familiar with it, I have some qualms with applying the ITN framework to actual decisions. Per the framing you described (which I get is meant to be simple and intuitive, but nonetheless), neglectedness seems obsolesced by importance, and tractability would probably also be obsolesced except that it tries taking into account implementation costs (while not addressing other potential disadvantages?).
How efficient at X would I be without using this framework, or how prone to making Y mistake would I be without this framework?
What does my expected usage/implementation of this framework look like (given cognitive or familiarity constraints)?
How efficient would I be at X or prone to making Y mistake while using this framework?
How good is an improvement in X / reduction in Y.
In reality, your description of the ITN framework seems a bit different from normal interpretations, which leads to it looking more like COILS. However, my loose sense is that it’s probably better to formally recognize the limitations of ITN for certain contexts (e.g., specific decisions) and explicitly identify/use alternate frameworks, rather than using “sort-of-ITN.”
In your section on “which recommendations are most worthwhile?” you mention using the ITN framework. While this is probably makes for efficient communication since many readers are familiar with it, I have some qualms with applying the ITN framework to actual decisions. Per the framing you described (which I get is meant to be simple and intuitive, but nonetheless), neglectedness seems obsolesced by importance, and tractability would probably also be obsolesced except that it tries taking into account implementation costs (while not addressing other potential disadvantages?).
Personally, if more people were familiar with it, I would probably recommend an approach more like the COILS framework that I’ve written about:
How efficient at X would I be without using this framework, or how prone to making Y mistake would I be without this framework?
What does my expected usage/implementation of this framework look like (given cognitive or familiarity constraints)?
How efficient would I be at X or prone to making Y mistake while using this framework?
How good is an improvement in X / reduction in Y.
In reality, your description of the ITN framework seems a bit different from normal interpretations, which leads to it looking more like COILS. However, my loose sense is that it’s probably better to formally recognize the limitations of ITN for certain contexts (e.g., specific decisions) and explicitly identify/use alternate frameworks, rather than using “sort-of-ITN.”