About
    • AMMDI is an open-notebook hypertext writing experiment, authored by Mike Travers aka mtraven. It's a work in progress and some parts are more polished than others. Comments welcome! More.
Search
Full

VGR-Liveness

07 Sep 2026 - 08 Sep 2026
Open in Logseq
    • VGR has a recent important theme of Liveness, but I don't quite get what he is pointing at, so this is an attempt to figure it out.
    • He would not object to starting with an LLM-generated text:
      • Ask Claude
        • Rao's liveness theory runs through a Contraptions essay series that began with "In Search of Liveness" (May 2025) and is still developing. The core move. Liveness is a condition of processes, not a property of entities. He wants a machinic, non-vitalist account — a mechanical clock is a better exemplar than an organism, and he explicitly discards Bergson's élan vital as naive vitalism with extra steps. It's also distinct from intelligence and consciousness; those are treated as orthogonal dimensions layered on liveness, with liveness having ontological priority. The definition. His bumper-sticker version: a process condition in which time tells memory how to grow, and memory tells time which way to point. More precisely, a process is live when it sits at the boundary between order and chaos in a way that lets it recognize and respond to itself in its own memories — memory that is orientable, exhibiting a "grain" that resists disruption in memory-preserving ways. The orientation is indexical (borrowed partly from how context works in LLMs) and creates a proto-identity without requiring anything it is "like" to be that process. Consequences he draws.
        • Liveness can jump across entity boundaries — Moore's Law migrating from Intel to TSMC, or the "Steve Jobs process" still running at Apple after his death. This is why he now rejects Samo Burja's "live players" framing as entity-idolatry.
        • A test: a live process sustains some legible entity-hood within itself, but not too much. When the entity captures the liveness, the liveness dies.
        • His paradigm case is liveness in distributed computing — Ethereum proof-of-stake, where the inactivity leak routes the validation process around dead nodes. The process has liveness; the nodes don't.
        • Time is derived rather than assumed. The "past" is the subset of memory in which the process can recognize itself; the "future" is where that recognition is unstable or unraveling; the "present" is order-resistant concurrent state partitioning the two. Chronological time is a projective fiction emerging from this. Framing. He treats liveness as the common denominator of contemporary anxiety — AI, the meaning crisis, institutional decay, climate — all reducible to "liveness insecurity," and something that only became visible as a problem around 2015. It pairs with two sibling themes: bug-fixing (its dual) and haunting (disembodied liveness). Since then it's grown into scaffolding for other work: the "Liveness Machine" is now the newest of his three world machines, just entering its Dawn stage as the Modernity Machine enters Dusk and the Divergence Machine reaches Day. Recent branches include "Writing Liveness" (April 2026), on literally live text that reshapes itself without a live writer behind it, and an open problem he flagged in June 2026: distinguishing organic balance (homeostasis, autopoiesis) from machinic balance as two varieties of liveness.
    • For me, it butts up against a few other theories of Liveness
      • The general Marxist/PKD/Matrix attitude that capitalism is a vast machine for sucking out human life and replacing it with something alienated, commodified, sold back to us, life replaced with pseudo-life.
      • And some of my own ideas, eg WorkPlay and of course my long term interest in animacy, agency, personhood.
    • I think he's trying to avoid the common framing of AI vs "Team Human". That says, more or less, that humans have liveness and AIs are fakes, possibly dangerous fakes. For VGR this is an absolutely bad idea that he is fighting by integraing his own work with LLM writing as much as possible, erasing the hard boundary (I'm not as advanced, I refuse to let my voice get mixed with LLMs, instead relying on devices like the AskClaude block above).
    • dogwalk god my brain is noisy. irritation at preschool and others in the neighborhood, not really justified, I don't own the outside, just like to feel I am alone there.
    • Is it really alone if you keep on yammering at yourself? Seems like a vain attempt to be two people.