Latest / Soft Ratios Research / Mould seams reveal who decides (on Ursula Franklin)

Transcript

Mould seams reveal who decides (on Ursula Franklin) · Soft Ratios Research

• 0:09
  Figure: Soft Ratios Research.
• 0:12
  Ground: The voices are synthetic [pause] no act, just the thinking.
• 0:19
  Figure: Hoyd Breton, a designer at Soft Ratios Studio, teaching himself in the open [pause] which is half a conversation. [pause] A review is the other half.
• 0:29
  Ground: One subject at a time, twice a week [pause] Tuesday and Thursday. [pause] Today, Ursula Franklin and who keeps judgment.
• 0:38
  Figure: A clay model sits at the beginning. Clay is pressed around it, removed in sections, fired, then fitted back together around a core. Molten bronze enters through prepared channels. When it cools, the mould is broken. [pause] The vessel keeps traces of that sequence. Seams. Joins. Small casting flaws. Evidence of decisions made long before the pour.
• 1:02
  Ground: The finished bronze conceals a crowd. [pause] Someone understood the whole vessel before anyone else completed a part. Someone preparing clay had to anticipate what hot metal would do later. Someone pouring metal inherited choices already hardened in the mould. [pause] That isn't sinister. It's coordination made physical. The unease comes from a smaller fact: by the time some hands enter the work, the meaningful choices have already been made. [pause] The object is beautiful because the sequence held. The same sequence can also tell each person, very firmly, where thought is allowed to happen.
• 1:48
  Figure: That was Ursula Franklin's opening. Not machines first. Arrangement first. She looked at ancient bronze production and found a way to ask where knowledge, authority, and responsibility sit inside a process.
• 2:02
  Ground: That shifts the usual question. A tool can be sophisticated or simple, digital or made of clay. None of that reveals who governs the work.
• 2:12
  Figure: Franklin called technology formalized practice. Her plainer version was “the way things are done around here.” Procedures, divisions of labour, credentials, timing, vocabulary, accepted evidence, measures of success. The equipment is only one part.
• 2:29
  Ground: Then a timetable can be technological. So can a grading rule. Or a clinic form that decides which parts of a person become visible.
• 2:39
  Figure: Only if the rule belongs to an organized way of doing something. Franklin's point wasn't that every habit deserves the name. Established practices can shape an activity as deeply as any machine.
• 2:50
  Ground: Deeper, sometimes. A machine can sit unused. A procedure can decide what counts as real before anyone begins.
• 3:00
  Figure: Take academic credit. Teaching doesn't count merely because understanding occurred. Institutions specify approved courses, periods, assessments, records, and authorities. The practice helps define what teaching is permitted to become.
• 3:15
  Ground: That sounds severe. Institutions need shared forms. Without them, trust becomes personal favor, and access can depend on knowing the right person.
• 3:26
  Figure: Franklin wasn't asking for a world without common rules. She acknowledged what disciplined coordination can do: reliable quantities, consistent quality, work beyond one person's capacity, goods that improve material life. [pause] Her distinction was about the location of judgment. In whole-process work, a practitioner understands and controls a meaningful process. Conditions change, experience answers, and the method can change with them. [pause] In prescriptive work, the process is divided into steps. Each step can be performed without command of the whole. The sequence, tolerances, pace, and acceptable result are defined elsewhere.
• 4:04
  Ground: So whole-process work doesn't mean handmade, solitary, or vague.
• 4:11
  Figure: No. Several people can collaborate and still retain real authority over their parts. A shared score can coordinate an orchestra without reducing every musician to a button pressed on cue.
• 4:22
  Ground: But the score remains fixed.
• 4:24
  Figure: And that's where the boundary starts moving. A score can establish enough common ground for interpretation. It can also become an excuse to punish any response the score didn't anticipate.
• 4:34
  Ground: Then the important contrast isn't freedom against rules. It's judgment near the event against judgment removed from it.
• 4:44
  Figure: Franklin's distinction runs through that contrast. She separated work-related technology from control-related technology using the same physical device. A freestanding word processor assists the typist. Connect it so assignments can be divided, interactions observed, and workers timed, and its role changes.
• 5:04
  Ground: The keys don't change. The relationship around them does.
• 5:08
  Figure: At the Ontario Research Foundation, Franklin used X-rays to examine industrial parts that had failed, jammed, or exploded. A radiograph didn't simply announce the cause. Differences in the image had to be read backward toward hidden thickness, voids, inclusions, or errors in fabrication. [pause] She carried that habit into archaeometry. Start with a visible trace. Reconstruct the concealed arrangement that could have produced it. Test the account without destroying the object. [pause] Then she enlarged the question. If a bronze seam records a manufacturing sequence, perhaps a workplace records an institutional sequence. Who prepared the conditions? Which later actions became possible? Which became forbidden?
• 5:52
  Ground: That move is tempting enough to become dangerous. [pause] Metal keeps its structure. People resist, misunderstand, negotiate, conceal, and change the arrangement while living inside it. A mould seam can't testify about consent. [pause] It can show that stages depended on one another. It can't name a supervisor, prove coercion, or reveal what an artisan thought about the work. [pause] The image of a factory hidden inside the bronze becomes too neat if the bronze is asked to supply an entire politics.
• 6:28
  Figure: The later archaeology requires that correction. Shang foundries show planned sequences, distinct clay and metal operations, substantial resources, and coordinated pours. They don't establish one continuous royal assembly line. [pause] Anyang contained several bronze-production zones. No two known Anyang bronzes are identical. Repeated vessel forms and shared methods coexisted with variation, reworking, and skilled decisions by artisans. [pause] Even the Houmuwu ding, weighing more than eight hundred kilograms, proves an enormous feat of preparation and synchronized labour. It doesn't reveal whether authority belonged to a court, a lineage, a workshop, or some mixture. [pause] Franklin's political reading remains an inference from the operation, not a social order extracted intact from a seam.
• 7:23
  Ground: Does that weaken her distinction, or rescue it?
• 7:26
  Figure: It rescues the method by limiting the claim. The bronze shows that technical dependence constrains later choices. That's enough to ask what kind of human arrangement those constraints invite. It isn't enough to declare the answer finished.
• 7:40
  Ground: A warning, then, not a verdict. Once one stage can ruin every stage after it, advance planning gains power. The fact that planning is necessary doesn't settle how that power should be shared.
• 7:55
  Figure: And it prevents a lazy rejection of prescription. The casting works because some choices become fixed. Molten metal above a thousand degrees doesn't pause while every participant renegotiates the vessel.
• 8:07
  Ground: Nor should a surgical team improvise whether the patient has been identified or whether equipment has been counted.
• 8:15
  Figure: Known hazards can justify firm safeguards. Checklists can catch memory slips and help specialized teams coordinate. The difficulty begins when a safety floor becomes a ceiling on attention.
• 8:27
  Ground: A floor says: these few things mustn't be missed. A ceiling says: nothing outside these boxes deserves notice.
• 8:38
  Figure: This goes deeper than a contest between craft and industry. Franklin's pair of labels becomes an examination of who gets to decide.
• 8:46
  Ground: Who names the purpose. Who defines an acceptable result. Who divides the work. Who controls the clock. Who can see the whole. Who may stop when the situation refuses the sequence.
• 9:00
  Figure: And who answers for the consequence. Prescriptive arrangements separate conception, execution, monitoring, and responsibility. Planners define the end and the steps. Operators perform fragments. Supervisors watch for conformity. Harm outside the chosen measure can belong to no one in particular.
• 9:22
  Ground: Or it can be pushed downward. The person nearest the harm is blamed for not saving a plan they weren't allowed to design.
• 9:30
  Figure: That happens in automated work as well. Software can issue the next warehouse order and time its completion. A bathroom break may be formally allowed, yet the resulting decline in the worker's numbers quietly discourages it.
• 9:43
  Ground: Permission on paper, discipline in the clock.
• 9:47
  Figure: Some self-employed delivery riders may technically reject work, unlike salaried riders, while opaque assignment rules can still leave them afraid that refusal will bring worse offers. Drivers may be able to dispute a rating while believing the dispute has little chance of changing anything.
• 10:03
  Ground: Choice survives as a door that appears unlocked but may set off an alarm.
• 10:10
  Figure: That image risks making every measure look coercive.
• 10:13
  Ground: The measure isn't the issue. The cost of contradicting it is. If a person can challenge the number, supply missing context, and alter the decision, the number remains evidence. If it silently governs pay, access, pace, or suspicion, it has become an order.
• 10:33
  Figure: Then Franklin's “social mortgage” enters. Prescriptive technology can produce real material wealth while accumulating a political debt. Repeated external direction makes compliance feel natural. The approved procedure begins to resemble the only imaginable procedure.
• 10:50
  Ground: That debt isn't paid in dramatic rebellion. It's paid in fewer moments of practiced judgment. [pause] A person stops asking whether a task makes sense and asks whether it was completed correctly. An anomaly becomes irritating because it delays the sequence, even when the anomaly contains the most important information. [pause] Eventually the capacity to decide doesn't merely go unused. It thins. The hands remain busy while the authority once carried in them migrates elsewhere.
• 11:25
  Figure: That erosion shows up in practice. When an accounting office discontinued software it had long relied on, gaps in employees' understanding of the underlying process surfaced and disrupted the work. [pause] In repeated flight simulations, trust in highly reliable automation increased while people became worse at noticing when it failed. Algorithmic direction of task order and pace is also strongly associated with reduced worker discretion.
• 11:52
  Ground: Still, independence isn't automatically wiser. Junior trauma clinicians’ departures were mostly errors. Senior-led teams produced more innovations and fewer errors.
• 12:04
  Figure: That forces a revision. The power to deviate isn't enough. Judgment requires knowledge, experience, and responsibility for consequences. Removing all guidance can abandon the least experienced person under the name of autonomy.
• 12:19
  Ground: And rigid guidance can prevent experience from becoming judgment. If every unfamiliar case is treated as disobedience, no one learns how to meet the unfamiliar well.
• 12:31
  Figure: So the dividing question changes again: when reality departs from the plan, where does qualified authority reside, and can the plan learn from the departure?
• 12:41
  Ground: The classroom makes that question uncomfortable because its product can't be poured into a mould. [pause] Learning has pauses that look like failure, speed that conceals confusion, silence that can mean withdrawal or concentration. Two students can produce the same answer by entirely different paths. [pause] A teacher close to the room may connect an answer to an earlier misconception, a strained relationship, a language barrier, a disability, or an explanation that simply failed. [pause] Break design, delivery, assessment, and certification into separate jobs, and each person may perform competently while the thread between them disappears. No one necessarily sees what the learner's response was a response to.
• 13:32
  Figure: Educational unbundling promises flexibility, specialized expertise, and scale. Those aren't imaginary benefits. Course designers can improve structure. Accessible media can remove barriers. Remote participation can reach people excluded by place, work, or disability.
• 13:51
  Ground: The arrangement turns prescriptive when assistance becomes command. A platform can distribute materials, or it can define acceptable teaching through mandatory presence, standardized interaction, scorecards, and surveillance.
• 14:08
  Figure: Learning dashboards expose the same split. They can reveal missed work or participation patterns that would otherwise be difficult to see. When teachers help shape the dashboard and decide what its clues mean, the tool extends perception.
• 14:22
  Ground: But a click isn't curiosity. Time logged in isn't understanding. A video left open says almost nothing about the person who left it open.
• 14:33
  Figure: Digital traces are attractive because the platform already possesses them. Once those traces trigger warnings or labels, the available evidence quietly becomes the official account of learning.
• 14:44
  Ground: Then the dashboard doesn't merely describe the room. It teaches the room what counts.
• 14:50
  Figure: Remote proctoring goes further. Institution administrators can select webcam, microphone, browser, screen recording, room scans, and automated flags. The student adapts body and home to the requirements of machine-readable integrity.
• 15:04
  Ground: And the teacher inherits suspicions produced elsewhere. Attention moves from the quality of thought to whether a face stayed in frame or a body moved in an approved way.
• 15:17
  Figure: That arrangement can intensify anxiety, frustrate disability accommodations, and treat unequal technical conditions as evidence of misconduct. Yet concern about assessment integrity doesn't vanish merely because surveillance is crude.
• 15:32
  Ground: A defensible answer would keep the concern human. Make the rule visible. Limit what's collected. Allow another route. Require explanation and review before harm. Let students and teachers revise the arrangement, not merely appeal after it acts.
• 15:50
  Figure: The difference matters. A complaint box inside a fixed design isn't control. Franklin's idea of reciprocity requires feedback capable of changing the first assumptions, the goal, or the structure of authority.
• 16:03
  Ground: And that's harder than offering settings. Choosing a notification color isn't the same as deciding whether the notification should govern anything.
• 16:12
  Figure: The medical record shows the same pressure in a different form. After electronic records arrived, documentation took a much larger share of both physicians' and nurses' working time. Nurses answered mismatches with paper notes and duplicate entries.
• 16:27
  Ground: Workarounds are usually described as little acts of disorder. Sometimes they're the place where reality refuses a poor abstraction.
• 16:38
  Figure: Sometimes they're dangerous, too. A hidden note or an improvised route can break continuity. The existence of a workaround doesn't prove the worker is correct; it proves the official path isn't carrying the full case.
• 16:51
  Ground: Nor can any one person contain a patient. There's always something another person can see. Specialization exists because bodies exceed individual knowledge.
• 17:04
  Figure: That's why Franklin's whole-process work can't mean one heroic generalist controlling everything. Shared clinical pathways can reduce unnecessary care, support communication, and validate a difficult decision. Specialists need common evidence and reliable handoffs.
• 17:20
  Ground: Yet clinicians also describe tunnel vision, pressure to adhere, and guilt over exceptions. A pathway can coordinate many eyes while narrowing what every eye is allowed to notice.
• 17:34
  Figure: Local changes draw a practical boundary. When clinicians can modify a checklist locally, participation and ownership improve. When production pressure turns the check into a hurried tick-box exercise, recorded compliance can conceal weak practice.
• 17:50
  Ground: So the useful standard doesn't pretend to finish the work. It marks known dangers, gives people a common place to meet, and leaves the case open enough for someone to say: this person doesn't fit.
• 18:03
  Figure: The phrase “professional judgment” can also hide arbitrary power. A practitioner may defend habit, prejudice, or convenience as sensitivity to context.
• 18:14
  Ground: Then judgment must be answerable without becoming remote. Give reasons. Keep records. Invite challenge from colleagues and from the person receiving care. Review exceptions for what they teach, not merely for whether they breached a rule.
• 18:31
  Figure: The middle is demanding. Prescription should be narrow, transparent, revisable, and open to reasoned exception. The people performing and receiving the work need access to the evidence and some power over what follows.
• 18:46
  Ground: There's a burden hidden there. Local discretion can become another way for an institution to transfer risk downward: the rule remains fixed, but a practitioner is told to improvise around its damage and accept blame for the result.
• 19:05
  Figure: Then exception rights aren't enough. If only remote authorities can alter the goal, staffing, timetable, or measure, adaptation becomes a patch supplied by people denied structural power.
• 19:17
  Ground: The whole plan must be contestable. Otherwise judgment is welcomed only as unpaid repair work.
• 19:24
  Figure: Franklin favored small and reversible steps partly for this reason. Infrastructure hardens political choices. By the time a harmful assumption becomes obvious, contracts, training, records, and dependence can make reversal prohibitively expensive.
• 19:40
  Ground: Consultation after the concrete sets is a ceremony around a finished mould.
• 19:45
  Figure: A prescription remains democratic only when affected people can enter before lock-in, inspect the values built into it, trace human responsibility, and change more than its surface operation.
• 19:59
  Ground: This is where a common rule can enlarge freedom instead of shrinking it.
• 20:03
  Figure: The IndieWeb gives us a useful case. A person can control a domain, choose how to publish, and keep the local work whole. But independent sites can't exchange replies, identities, or subscriptions without shared conventions.
• 20:17
  Ground: Total self-sufficiency would leave everyone sovereign and alone.
• 20:22
  Figure: Narrow web standards constrain builders at the point of connection. In return, people can choose among compatible tools without surrendering their address or all their data to one provider.
• 20:33
  Ground: That bargain stays tolerable only while the common rule can be inspected, changed, implemented by more than one party, or refused without exile.
• 20:44
  Figure: Open proposals, documented objections, working examples, replaceable services, and portable data make prescription contestable. They don't guarantee decentralization. Power can gather again in hosting, discovery, identity, funding, or expertise.
• 21:00
  Ground: A decentralized diagram can still contain one narrow doorway everyone must use.
• 21:07
  Figure: That returns the inquiry to structure. [pause] Franklin began with crystal structure: the properties of a whole depend on the arrangement of its ingredients as well as the ingredients themselves. She carried the question into institutions without claiming that people behave like atoms.
• 21:24
  Ground: The continuity is in the attention. Look at the relations between parts. Look at which boundaries permit movement and which become walls. Change an arrangement and the same people may gain or lose capacities without changing who they are.
• 21:39
  Figure: A useful examination can follow one task from conception to consequence. Find who defines success. Reconstruct the handoffs. Mark where judgment is permitted, where it's punished, and who can override whom. [pause] Then compare the gains with the burdens. Who saves time? Who supplies extra hidden work? Who becomes visible to monitoring? Who may act on the information? Who remains answerable when the sequence causes harm? [pause] Then ask Franklin's question: what can no longer be done after this arrangement is installed? A technology reveals itself partly through the alternatives it removes.
• 22:18
  Ground: That examination still feels like an observer's privilege. It names the distribution of power without returning any.
• 22:26
  Figure: Description is only the first move. Information with no route to accountable authority can keep the opposition busy forever. A flawless diagnosis changes nothing if no responsible body must answer.
• 22:39
  Ground: Then the practical work is to attach knowledge to consequence. A visible name behind the decision. An appeal that can alter the outcome. A pilot that can stop. A record that affected people can inspect. A standard that can be retired.
• 22:59
  Figure: And participation with binding force. Municipal budgeting changed public spending when residents actually selected priorities. Worker representatives in German and Norwegian call centres secured limits on algorithmic monitoring. Voice mattered because it reached the rules.
• 23:17
  Ground: Not every workplace can be governed as a public assembly, and not every clinical step can wait for collective debate.
• 23:24
  Figure: No. The scale of participation must fit the decision. A person facing an immediate hazard needs clear authority. A shared protocol needs periodic review by those using and receiving it. A public institution needs open reasons and routes for challenge.
• 23:41
  Ground: The common thread is that no layer gets to declare itself merely technical. Timing, defaults, targets, staffing, and access rules all settle human questions, even when written as neutral settings.
• 23:57
  Figure: By then, Franklin has turned the question around. The visible tool is no longer the main object of judgment. Attention moves to the organized practice around it: its purpose, sequence, permissions, observers, beneficiaries, and avenues of refusal.
• 24:12
  Ground: That also moves the argument away from fear of technology. A caption can open participation. A checklist can prevent a fatal omission. A shared format can let strangers cooperate. A machine can extend perception beyond unaided senses. [pause] The danger isn't that tools possess intentions. It's that human intentions become fixed inside arrangements and then disappear from view, leaving operators to experience a political choice as an unavoidable property of the task.
• 24:44
  Figure: Efficiency deserves the same scrutiny. Faster for whom, toward which end, and with what removed from consideration? A process can become very efficient at producing the wrong account of learning, care, safety, or work.
• 24:59
  Ground: The answer isn't slowness for its own sake. It's keeping enough time and authority near the event for reality to correct the plan.
• 25:08
  Figure: And keeping responsibility near authority. If a vendor defines the sequence, a manager chooses the target, and a frontline worker absorbs the consequence, the arrangement has separated power from answerability.
• 25:21
  Ground: Mould seams are useful because they show that a finished object had a history. Procedures have seams too: handoffs, exceptions, duplicate notes, appeals, delays, quiet acts of repair.
• 25:34
  Figure: Those traces reveal where the formal account and the lived activity fail to meet.
• 25:39
  Ground: A humane arrangement doesn't erase that gap. It gives the people inside it enough standing to speak from the gap and enough authority to change what produced it.
• 25:50
  Figure: The difference isn't whether there's a plan. It's whether the plan still belongs to the people who must answer for what happens.
• 25:57
  Ground: When reality refuses the plan, whose judgment gets to change it?