Explorable Explanations

As always, if any of this inspires you to play around with these concepts, I'd love to see what you come up with.

Ten Brighter Ideas was my early prototype of a reactive document. The reader can play with the premise and assumptions of various claims, and see the consequences update immediately. It's like a spreadsheet without the spreadsheet. Give it a try.

Here is a more simplistic example of the same concept.

California has state parks, including state beaches and historic parks. The current $ million budget is insufficient to maintain these parks, and parks will be shut down at least part-time. Most parks charge per vehicle for admission.

Proposes to charge car owners an extra $18 on their annual registration bill, to go into the state park fund. Cars that pay the charge would have free park admission.

Suppose that an extra was charged to % of vehicle registrationsCalifornia taxpayers. Park admission would be for everyonethose who paid the charge.

This would collect an extralose $ million ($ million from the tax, plusminus $ million additionallost revenue from admission) for a total state park budget of $ million.

This is not sufficient to maintain the parks, and parks would be shut down at least part-time. This is sufficient to maintain the parks in their current state, but not fund a program to bring safety and cleanliness up to acceptable standards. This is sufficient to maintain the parks in their current state, plus fund a program to bring safety and cleanliness up to acceptable standards over the next years. This is sufficient to maintain the parks and bring safety and cleanliness up to acceptable standards, leaving a $ million per year surplus.

Park attendance would risefall by , to million visits each year.

Is this a good proposition? It's hard to evaluate without context. The active reader might wonder, "Why $18? What if the tax were more or less?" Or, "Could park admission be raised instead?" If we were reading this on paper, such questions could only be answered by a phone call or heavy research.

Fortunately, this isn't paper. Some parts of the analysis above are underlined -- try adjusting them and see if you can answer those questions.

Notice how the consequences of your adjustments are reflected in the following paragraph. The reader can explore alternative scenarios, understand the tradeoffs involved, and come to a more confident conclusion about whether the proposition is a good decision.

There's nothing new about scenario modeling. The authors of this proposition surely had an Excel spreadsheet which answered the same questions. But a spreadsheet is not an explanation. It is merely a dataset and model; it cannot be read. An explanation requires an author, to interpret the results of the model, and present them to the reader via language and graphics.

The reactive document integrates spreadsheet-like models into authored text. It can be read at multiple levels, depending on the reader's level of interest. The hurried reader can skim it. The casual reader can read it as-is. The curious reader can adjust the author's scenarios. The engaged reader can explore scenarios of their own devising.

Unlike a spreadsheet, the barrier to exploration here is extremely low -- simply click and drag. This invites casual readers to become engaged and start exploring. It transforms readers from passive to active.

On the author's side, this form encourages a sort of transparency. The author's argument cannot simply be a hodge-podge of soundbites and unsourced data. A reactive document requires the author to disclose the models behind their argument, to open them up for scrutiny. (In Ten Brighter Ideas, the reader can even directly edit the source code of the model, as well as visit the primary sources for all data.)

Dishonest authors will always exist. They might use models based on faulty reasoning or data, but transparency means that the faulty model is available to be examined and refuted. Or they might offer no model at all, but perhaps readers will learn to be skeptical of non-explorable arguments.

Multiple authors could model the same situation, and readers could compare. If you look at the groups for and against the above proposition, you'll see they're basically just hurling around unsourced soundbites, leaving readers little to go on besides emotional appeal. What if both sides were expected to offer reactive documents, and the reader could critically explore their predicted scenarios?

What if readers wanted to explore such scenarios, because it was actually fun?

The following is a typical description of a digital filter, as you might find in a typical textbook.

Below is a simplified digital adaptation of the analog state variable filter.

This topology is particularly useful for embedded audio processing, because Fc (cutoff frequency) and Q (resonance) are controlled by independent coefficients, kf and kq. (With most filters, the coefficients are functions of both parameters, which precludes pre-calculated lookup tables.)

The coefficients and transfer function are: