Zwicky

Last updated: Monday, 23 September 2024

Computer-supported morphological analysis, whatever that looks like. (Let’s find out.)

Zwicky is an exploratory tool for morphological analysis, aimed at mapping and understanding complex problem spaces. Named for Bulgarian-born Swiss astronomer Fritz Zwicky, Zwicky seeks to extend and reimagine morphological analysis through modern computational approaches.

Zwicky seeks to:

  • Create an exploratory (software) tool that supports and enhances the morphological analysis process, encouraging gradual mastery through sustained use
  • Emphasise problem exploration over solution generation, encouraging users to engage with complex issues rather than rushing to a “best” solution
  • Experiment with natural language processing for dimension extraction, helping users identify key parameters from initial brainstorming or relevant texts1
  • Explore nuanced approaches to cross-consistency assessment, moving beyond binary feasibility to gradient-based or multi-criteria evaluations
  • Develop computationally-efficient ways to represent multi-dimensional problem spaces and solution landscapes
  • Experiment with multimodal outputs to cater to different user needs and preferences, potentially including visual maps, textual narratives, and interactive scenarios
  • Explore ways to handle temporality and change over time within the morphological framework

Documentation will be layered and adaptive, supporting users with varying levels of expertise.

Ininitial steps: Conduct a literature review on computational approaches to morphological analysis; develop a proof-of-concept tool for creating and exploring Zwicky boxes.

  • [?] How can we manage the “combinatorial explosion” inherent in complex morphological analyses?
  • [?] What role can machine learning play in enhancing morphological analysis?
  • [?] What are the implications of focusing on problem exploration rather than solution generation? How might this affect the tool’s utility in different contexts?
  • [?] What kind of user interface would be best placed to support both free-form brainstorming and structured analysis within the same tool?
  1. I’m envisaging a two-stage approach for dimension input, with free-form, unrestricted brainstorming, followed by guided refinement (potentially through multi-turn dialogic interaction). ⤴︎

Tags: activity

Backlinks