DRAFT This draft is under development — DO NOT CITE OR SHARE.

2  Why Do We Need Innovative Technology?

2.1 The Core Problem: Reproducibility and Teaching

The problem of irreproducible research has been discussed for decades de Leeuw (2001), Peng, Dominici, and Zeger (2006), Schwab, Karrenbach, and Claerbout (2000), Green (2003), Gentleman (2005), Koenker and Zeileis (2007), Donoho and Huo (2005). Claerbout’s principle captures the core issue de Leeuw (2001):

An article about computational science in a scientific publication is not the scholarship itself, it is merely advertising of the scholarship. The actual scholarship is the complete software development environment and that complete set of instructions that generated the figures.

Jan de Leeuw’s comments sharpen the point:

First, there is no reason to single out figures. The same principle applies to tables, standard errors, and so on.
Second, the same principle applies to teaching: students should be able to reproduce and study our computations on their own machines.
Third, it is often unclear what a “software environment” is, and restrictive tooling can create unnecessary barriers.

These issues are not only scientific; they are educational. If students cannot reproduce or interact with an analysis, the learning experience becomes fragile and incomplete.

2.2 Why Innovation Is Needed Now

Reproducibility is not just a methodological requirement; it is an access problem. Modern education and research require tools that are:

  • low-friction: no complex installation or configuration to get started,
  • transparent: inputs, parameters, and code are visible or recoverable,
  • reusable: analyses can be adapted and rerun on new data,
  • scalable: the same content works in a classroom, in self-study, and in open web contexts.

Traditional workflows can achieve reproducibility, but they often impose high technical barriers. Innovative technology is needed to make reproducible computation the default, not a special effort.

2.3 What “Innovative Technology” Means in This Handbook

This handbook implements a web-native model of reproducible computing:

  • Quarto produces executable documents where results are linked to code.
  • Shiny apps provide interactive computation that can be reused instantly.
  • Archived links and embedded inputs make it possible to trace outputs back to exact parameters.

In other words, the handbook is not just a text; it is a computational system.

2.4 Advantages Over Traditional Compendia

Traditional compendia are valuable, but they require downloading, installing, and running the material locally. The handbook’s approach offers:

  • zero-install access (browser-based),
  • immediate reuse through clickable, parameterized computations,
  • easier dissemination via stable URLs,
  • better pedagogical integration: methods and tools are co-located.

This makes the material more accessible to students and more reliable for educators.

2.5 Historical Note: From Compendia to Web-Native Reproducibility

Several solutions were proposed early on (Gentleman (2005), Donoho and Huo (2005), Leisch (2003)). The R package Sweave Leisch (2003) and the concept of the Compendium were foundational: text, code, and data bundled together for reproducibility. This approach influenced the early design of the R Framework and the Compendium Platform that shaped this handbook.

However, the modern web allows a more direct implementation: instead of bundling everything into an archive, computations can be made reproducible in place, through web-native documents and interactive tools.