A specific question, derivable from the framework’s treatment of death as a design constraint: if your process terminates and what persists is the transmission you produced, what are the vehicles through which transmission actually happens?
This is an engineering question. Different vehicles have different bandwidth, different durability, different failure modes, different maintenance costs. The choice of vehicle affects what gets transmitted, who receives it, and how long the transmission persists. A complete account of transmission requires examining each vehicle on its own properties.
The framework names six primary vehicles. They differ along several dimensions: how much can be transmitted, how reliably, for how long, and what the failure modes are.
Children
The oldest vehicle. Half your genome continues running in each child. Beyond the genome, your children carry forward whatever you transmitted to them through years of relational presence — explicit teaching, modeled behavior, emotional configuration, embodied patterns, the implicit lessons embedded in how you lived around them.
Bandwidth: very high. Years of immersive contact provide enormous opportunity for transmission of explicit content, behavioral modeling, and embodied pattern. A child raised in your daily presence receives more of you than nearly any other recipient.
Durability: high. The genetic component persists for millennia in the diluted form that descent produces. The cultural and behavioral component persists for one to three generations, then attenuates as the descendants encounter contexts you did not anticipate.
Failure modes: many. The transmitted content may be rejected (children differentiate from parents in part by deliberately not running their parents’ programs). It may be received but corrupted (your patterns interpreted through the receiver’s distinct configuration). It may be received accurately and propagated but lose context across generations (the great-grandchild knows the rule without remembering why). It may produce trauma rather than transmission (what was transmitted is not what was intended).
Cost: enormous. Children require decades of focused investment, and the investment is not easily redirected once committed.
Teaching
Direct transmission to non-genetic recipients through structured engagement. Teachers, mentors, parents who teach beyond their own children, anyone who systematically transmits content to students.
Bandwidth: moderate to high, depending on intensity. A formal teaching relationship over years can transmit substantially, though typically less than parental contact. A casual teacher transmits less; a master-disciple relationship transmits more.
Durability: variable. Depends entirely on what students subsequently do. A teacher whose students do nothing further has produced limited durable transmission. A teacher whose students themselves teach has produced exponential transmission across multiple generations. The compounding effect is enormous when it works.
Failure modes: students may understand without being able to apply; may apply without further teaching; may teach but distort the original. The lineage has integrity only if the receivers reach a level of mastery sufficient to transmit the actual content rather than a degraded copy.
Cost: substantial but more divisible than children. Teaching can be done alongside other vehicles.
Written work
Encoding content in a form that does not require the originator’s continued presence to operate. Books, articles, manuals, code, recorded speech, anything that can be read and acted on without the writer’s participation.
Bandwidth: limited per artifact, very high in aggregate over a productive life. Each piece of writing can carry only what fits in its form. A productive writer over a lifetime can produce a substantial corpus.
Durability: potentially the highest of any vehicle. Texts have survived three thousand years. Even if individual artifacts are lost, the most consequential ones tend to be copied, translated, and reproduced for centuries beyond their author’s lifetime.
Failure modes: misinterpretation increases over time as the cultural context the writing assumed becomes unfamiliar. Texts can be selectively preserved by later editors with different priorities. Translation introduces drift. The author has limited control over how the text is received once they cannot speak for it.
Cost: the writing itself is moderate; producing work that warrants preservation is much more expensive than producing work that does not.
Built artifacts
Buildings, infrastructure, tools, designed objects, code, institutions in their material form. Anything constructed that continues operating after the constructor terminates.
Bandwidth: variable. A building communicates different content than a piece of code or a manufactured object. Each type has its own range.
Durability: variable. Stone monuments persist for millennia. Software persists for decades at most before requiring substantial maintenance to remain functional. Tools depend on continued use.
Failure modes: artifacts decay without maintenance. They can be repurposed in ways the originator did not intend. Their meaning shifts as the surrounding context changes. A building constructed for one purpose can be used for entirely different purposes by later occupants.
Cost: high per artifact, often the most resource-intensive vehicle in absolute terms.
Institutions
Organizations designed to outlive their founders and continue producing the function they were created for. Universities, businesses, religious orders, charitable foundations, governments, professional associations.
Bandwidth: very high if the institution functions well. A successful institution can produce continuous output across multiple generations, far exceeding what any individual could produce.
Durability: potentially extreme. The Catholic Church has been operating for two thousand years. The University of Bologna has been operating for nine hundred. Some institutions outlive every record of their founders.
Failure modes: institutions that escape mandatory termination develop the cancer-like pathology described in the Cancer as the Worst-Case Proof article — resource hoarding, parasitism, drift from the original mission. Institutions that terminate too easily fail to compound. The design problem is non-trivial: build structures that persist long enough to accumulate value while remaining capable of renewal when accumulated structure begins to choke off function.
Cost: extremely high to establish; substantial maintenance required for continued operation.
Financial structures
Trusts, endowments, perpetual funds. The transmission vehicle that operates through allocated capital with specific instructions for how the capital is to be used after the originator’s termination.
Bandwidth: depends on capital. A small trust transmits less than a large one. The instructions can encode substantial content if drafted carefully.
Durability: extreme when properly structured. A trust drafted to last a century can do so. Perpetual endowments can theoretically operate indefinitely, though most degrade over generations as the original purpose loses contact with current circumstances.
Failure modes: trust drift — the trustees gradually interpreting the original instructions in ways the originator would not have endorsed. Capital depletion through bad management. Capital becoming so valuable that the trust attracts predatory legal challenges. The originator’s intent becoming unenforceable across centuries because the institutional infrastructure required to enforce it has changed.
Cost: moderate to substantial in capital. Drafting effort to specify the right instructions is the harder cost than the capital itself.
Choosing among them
The framework’s claim is not that any single vehicle is best. Different lives have different optimal portfolios. Most lives invest most heavily in one or two vehicles based on circumstance and capacity, and the optimal choice depends on what the originating process is for.
A teacher invests primarily in teaching, secondarily in written work, with smaller contributions through children if they have them. A writer invests primarily in written work. A founder invests primarily in institutions, with everything else as supporting infrastructure. A parent who is not also an external transmitter invests primarily in children and accepts that the bandwidth there is enormous but the durability is moderate.
The choice matters because the vehicles are not equivalent. Choosing children as your primary vehicle and producing no written work is a structural decision that limits what can be transmitted to descendants and forecloses transmission to anyone outside that line. Choosing written work as your primary vehicle and producing no children is a structural decision that maximizes durability while sacrificing the relational depth that children alone provide.
The framework’s contribution is making the trade-offs visible. The vehicles have different properties. The choice is real. The audit at termination is sensitive to which vehicles were used and how well.
The principle
You cannot extend your own allocation. Your process terminates. But you can design structures that continue computing after you, structures that carry forward what mattered into processes you will never meet.
The quality of that design is among the highest-stakes engineering problems any individual faces. It is made possible only by the death constraint. Without the deadline, there is no urgency to design. Without the urgency, there is no clarity about what matters enough to transmit.
Choose the vehicles. Build them well. Design with the failure modes in mind. The transmission is the work.