How People Use Pond 1: Coordinating Understanding Over Time

there's a method of reading called incremental reading, which involves gathering a set of texts (pdfs, articles, books), splitting them up into small sections, and reading across these rather than going through each text in sequence. as one does this, one highlights and annotates and takes notes on the texts, all of which then forms the basic material for a (standardly software assisted) spaced-repetition practice, during which the reader revisits and engages with (e.g., by completing cloze deletions) flash cards of content at increasingly longer intervals until they've developed a strong familiarity with it.

there are two ideas motivating this method: (a) that 'cross-pollinating' across texts leads to a greater understanding of the material, and (b) that spaced-repetition—encountering and engaging with something at increasing intervals over a period of time—works.

i have no doubt that these ideas are sound, and that incremental reading is an incredibly effective and serious way to learn anything. my problem is that most of the time things are not that serious, and i'm not trying to be optimally effective, but i nevertheless want to benefit from, or at least remember what i'm reading. is this asking for too much? maybe, but to make the vision tractable i'm willing to forego the promise of acquiring proficiency over a short period of time if it allows me the ability to be more divergent and free-ranging with my curiosities.

with that said, here's a method that's been working for me: all the highlights and annotations and notes I take on things I'm reading or watching end up in my pond, through the readwise highlight sync, through dropping marked up pdfs, and through highlights and notes I make from within the app. over time, pond, often with my guidance, but sometimes independently, 'discovers' the lines of thought running through this material and gathers them into a container called a thread.

threads are a useful abstraction: they track an idea, or a hypothesis, or anything more substantial than a passage but less developed or contained than a complete text. threads are the basic structure that make the cross-pollination that is so central to incremental reading possible in pond.

then, pond presents occasions for me to return to a thread and engage with its material at a more natural rhythm than the mechanical ones that tend to accompany standard spaced repetition systems. for example, when i read and create some highlights and annotations about biomechanics, pond might judge that these are relevant to a thread i've been developing over some weeks or months about aviation, and surface the possible connection and its reasoning.

these suggestions aren't always terribly insightful, but they serve as a prompt for me to consider how what i'm presently reading relates to ideas i've been developing, or things i have read about before. it is an opportunity for me to re-engage with my material. the interval at which i do so is certainly not as optimal as FSRS—the schedule or algorithm underlying contemporary spaced-repetition systems—but it nevertheless is a way for me to regularly return to a set of ideas over a period of time. not only this, but the occasions on which i return to these ideas feel more organic: i return because a present interest or topic has drawn me back—not the calendar—and its through the frame of this present interest that i reinterpret the material. the experience feels all the richer.

this method suits the form that my explorations and curiosities take. they play out over months, if not years, and across disciplines and forms of media. sometimes it will mean that some threads slip into a state of dormancy because my reading might concentrate around a certain set of interests, but this is fine really; is it so bad to only keep up with more or less active concerns? there's only so much one can have in mind after all.

to close, let me give a concrete example of pond playing this loose incremental reading scaffold role:

i saved an essay by james duesterberg that i found on substack but hadn't yet read to pond, where a short summary of it was generated (this is optional but useful because it's how a text that i haven't yet highlighted or annotated can appear in pond's suggestions, which then serve as occasions to consider actually reading it).

on the home page, pond used this summary to identify threads that this text might be relevant to/might extend or contribute to:

i'd created this thread ('language machines and the space of reasons') a little while ago to help lay the ground for me to read leif weatherby's reportedly excellent book, Language Machines. in truth, i'd discovered the james duesterberg essay when it was restacked by weatherby, so pond's suggestion was a little redundant here, but you can see that if i'd discovered the essay independently, pond would have gestured at the same connection that weatherby himself had identified.

in any case, i accepted the suggestion, and made a little note in the thread about what came immediately to mind.

all this inspired me to finally pick up the weatherby book; which led to more highlights, and in turn, more thread resonance suggestions.

i won't pick these connections all up now; but it's good to have a sense of the many ways in i could have to the text, and the many ways the text itself enriches some of the things i've been thinking about. pond has made evident the many possible dimensions of this dialogue; it'll take many revisits and periods of reflection and deeper engagement to develop a true understanding of the ideas in play, but i find much comfort in the thought that my environment now offers structure for this kind of activity.

pond's in early alpha at the moment; let me know if you'd like to try it out and help shape it.