Restructure Scope Section

[samm82]

From #130 and #137, I should make my "scope discussion chapter [more] focused on the conceptual aspects of what is in/out, and a higher-level reasoned justification for keeping certain things in scope."

• Give a high-level overview of what is in scope and out of scope
• Move detailed discussion on what is out of scope to an appendix; I can then say something like "For a complete analysis of XXX for scope, see Appendix Y for a detailed discussion."
• If remaining scope content is short, say around a page long, fold it into my introduction
  - [ ] Get @JacquesCarette to re-review as appropriate. If the scope gets folded into the intro, this can be done alongside Formalize confusion around "error", "defect", "fault", etc. as a core example in the intro #132. Will be done as Address Dr. Carette's re-review of Introduction Section #141
  - [ ] Figure out why \Cref is referring to \label{app-scope} as "Section 9.4" instead of "Appendix A" Moved to a qtodo in case this isn't worth investigating.
• Ensure changes are correct in paper

[samm82]

After following this advice, I've moved the main Scope subsections (hardware, other V&V, and static testing) to an appendix, which I think makes a lot more sense! Now, the only remaining subsection is Derived Test Approaches, which I think is notable to include, since:

1. I define ways that test approaches can be derived that are in scope, and note this in the abstract. These are attacks and approaches derived from coverage metrics, qualities, and requirements.
2. I define ways to categorize "test approaches" as simply those derived in other ways, resulting in (at least partially) out-of-scope approaches. These are approaches derived from languages and combinations of other approaches.

This seems to tell a different story that what is in scope, instead explaining ways we classify/discuss test approaches. Would this subsection make more sense as a subsection of Terminology? If so, then the Scope section would be reduced to just its intro paragraph, which could be folded into the Introduction (or would it make more sense in our Methodology section?) @JacquesCarette

[JacquesCarette]

A 1-paragraph Scope section belongs in the introduction.

You'll need to remind me of your whole organization. "Derived test approaches" seems like a nice thing to talk about, but where exactly, I'm not so sure!

[samm82]

The main sections I have are Intro, Scope, Methodology, Tools, Discrepancies, and Recommendations.

• Intro and Scope are pretty self-explanatory
• Methodology includes:
  • Our process for finding and grouping sources
  • The terminology we define/observe
  • The procedure we follow
  • The impact of undefined terms, including how we iterate over them and terms that did not naturally arise from the literature
• Tools describes my Python scripts and LaTeX macros I build to support this research (not here)
• Discrepancies (which will be renamed to "Flaws") lists the flaws we find in the literature; these are presented either through automated detection or through automated analysis on manually recorded data (both described in Tools) (not here)
• Recommendations lists the proposal we (I) have made to make the terminology more coherent. Honestly, I think we may end up abandoning this section (although the tools I created to support this could still be included in the Tools section as an "exercise for the reader"), since I don't know how thorough/consistent they will end up being

Re-reading the intro, I think I find a place where I could slot in the scope paragraph into the intro. As for derived test approaches, I think they could find a home in either:

1. The terminology section, since these are groups of test approaches that emerge from (but are not explicitly defined by) the literature, or
2. The procedure section, since they are used as a heuristic for which approaches are in scope and which aren't. This has the added benefit where "Undefined Terms" could then be made a subsection of "Procedure" (since a section shouldn't have a single subsection), which I think is a more logical approach.

[JacquesCarette]

Methodology, most definitely.