Foreword — Status of This Volume

Note on the translation — This translation re-fixes the source text without displacing its substance. A few terms of the corpus have no obvious equivalent in English; the choices made, and how they are to be read, are set out in the section “Translation choices” at the end of the volume. This translation is not frozen: it will evolve with readers’ feedback and with the improvement of the means of translation. The French version is authoritative.

What This Volume Is

This volume is a matrix. It formalizes, for a very small structure — artisan, sole trader, workshop —, the information indispensable to the conduct of its activity, and it does so in a form to be filled in: sheets, cadence sheets, indicators, grids, which a given structure fills in to equip itself with its own quality and activity management system. It is therefore not the portrait of a particular business, nor of a particular trade; it is the structure common to every activity that inscribes, circulates and rereads data in order to conduct itself — and the mould in which each structure casts its own.

Its principle comes down to a shift. One does not describe the activity by its products, but by the movement of the datum that accompanies it. At each stage, the work on the material is accompanied by an inscription; it is the succession of these inscriptions — reread — that makes the activity governable. Documenting this movement is the object of this volume.

The Twin Volume, and the Two Questions

This volume does not work alone. It forms a pair with Technical Data, and the two, together, equip the data artisan (a craft yet to be created) to answer two questions that every assignment raises:

• Which technical solutions should one rely on to dig the canal at the client’s premises? — this is the business of Technical Data, the workbench stocked with the tools of the moment.
• How should one document the movement that this canal produces? — this is the business of the present volume, the management matrix.

The one supplies the tools, the other the way of holding what they produce. Technical Data says with what and how one digs; this volume says how what flows from the canal, once dug, is governed.

A Dated Volume, with Low Volatility

Like Technical Data, this volume is dated: it claims to be usable on the ground of entrepreneurship in 2026. But its dating is not of the same nature. Technical Data goes stale fast, because its tools change fast — languages, models, versions. This volume, for its part, engages a ground that lasts.

For what a management matrix documents are relations: to the client, to the supplier, to the partner, to the staff. These relations do not change in their nature. One orders, one gets supplies, one commits, one works with others — today as yesterday, and probably tomorrow. What changes belongs to form: tooling is modernized, like any activity that equips itself; and regulatory obligations, changing by nature, are held here for what they are — parameters of the system, whose impact is anticipated rather than undergone. Hence the low volatility of this volume: it documents a movement whose poles are stable. It will be revised, but slowly.

Its Place in the Corpus

This volume does not belong to the trilogy Tried by the Flow. It is an extension of it, the sixth continuity volume, alongside The Canal and the Workshops, the Complementary Notes, The Data Artisan, The TRACE Programme and Technical Data. The trilogy keeps to substance: it says what is true of the datum independently of the techniques that produce it, and for that reason forbids itself to prescribe form. This volume makes the opposite wager, a measured one: it prescribes a form — an apparatus, processes, inscriptions, indicators —, but a form general enough to be applied without going stale fast.

This apparatus has a history: in practice, it preceded the ontology that today formulates its substance. What this volume handled without naming it — the datum as what is noted, put away, circulates between media, is reread to decide —, the trilogy raised to the rank of an ontology. One is therefore not plastering a theory onto a tool here: one recognizes in the tool the ground where the theory germinated. This substantive rereading is the object of a separate note on the re-edition; the present volume is its dated implementation.

The Two Functions: Quality and Activity

A management matrix must, at the very least, render two services to the entrepreneur, which give the volume its two sides.

Quality — presenting the activity and controlling what circulates. Presenting the activity is establishing the data that identify and situate it (section II). Controlling what circulates is keeping the discipline of processes (sections III and IV) and of traceability (section V) — knowing, at every moment, what comes in, what is transformed, what goes out, and through which hands.

Activity — steering and sharing. Constituting flows of observation is equipping the activity with indicators read in time, in order to conduct it (section VI). But these same flows, once qualified, are worth something beyond the workshop: well drawn, well dated, well traced, they are high-value data — a potential value, which only reading realizes —, and their sharing opens the way to the proposal that the corpus carries (section VI also). Project 00 of 2007 stopped at local steering; this volume extends its logic towards pooling.

And, beneath these two gestures, a third holds them: tending the agreed — the shared language in which the activity draws and through which its datum becomes legible to another. This agreed is a tended stable source (see the complementary notes); keeping it attuned to the real is what makes the datum both controllable within and shareable without.

The Vocabulary Is That of the Corpus

This volume adds no concept to the core of the corpus. It uses its vocabulary, fixed by the trilogy, and puts it to work: flow of the real, drawing, act of inscription, families of constitution, flow of observation, canal, cadence, freshness, lapse, validity for a use, trace, situated reader, read indicator, constituted datum, actualized datum, observational optimum, stable sources. The notions specific to it — high-value datum, craft matrix, seam point — and the agreed, which it shares with the complementary notes, appear in the canonical lexicon as satellite entries; canal-keeper, finally, is a term of the continuity volumes, not of the trilogy. Where a notion comes from a precise tome, this is indicated, so that the reader can go back to the source.

It rests in particular on the grid of eleven questions of The Canal and the Workshops, which groups by phase — making, circulation, reception — the questions to be asked in front of any datum. The apparatus described here is, in many respects, the putting into practice of this grid in the workshop: its cadence sheets answer the questions of making, its circuits the questions of circulation, its indicators the questions of reception.

The Craft This Volume Calls For

The trilogy calls for a craft that does not yet exist: the data artisan (canal-keeper, in the continuity volumes) — the one who would draw, circulate and reread the datum with the probity of a skilled practitioner. This volume shows, in its own way, why this craft is missing. The gestures it presupposes — someone notes, puts away, rereads, decides — are already present in it, but scattered among functions that go by other names and pursue other ends. No one, in the workshop, holds these gestures together as a craft of its own. This volume makes them visible as such; it nowhere claims that they would already form an established profession. The craft remains to be created; this volume documents the need for it — and, through the matrix it proposes, equips its hand in advance.

A Matrix for the Public Proposal

Because it is a matrix, this volume is a starting point. Versions will be derived from it for a given trade; it will be used to set up a concrete management system, to automate certain stages of inscription or circulation, to supervise the activity from an overall view. It is also from it that the tools of the proposal carried by the corpus’s site — artisanat-de-la-donnee.fr — will be built: equipping data artisans in the service of very small structures, as close as possible to the territory. All these applications remain downstream: they presuppose the matrix set out here, and add to it the form that their situation calls for.

I. The Overall Apparatus

Objective and Method

The apparatus aims at a balance, not a maximum. It is a matter of stabilizing what can be stabilized — the operating cost, the fixed part of supply, the rhythm of deliveries —, of delimiting what varies, and of installing a regulation that keeps the activity within a range of tolerance, before the gaps become critical. Neither under-steering, which lets the activity drift blindly; nor over-steering, which drowns the workshop under checks. This balance is, in the vocabulary of the corpus, an observational optimum: the point where the arbitration between the benefit of steering and its cost is most satisfactory for the intended use. It depends on the context; it is not a fixed value, but a question to be asked again.

The method is relational. Managing the relation to the client according to the quality aimed at; managing the relation to the supplier with the objective of a constant flow, through order cadence sheets that stabilize delivery frequencies; managing the relation to contributors through clear commitments that plan what is foreseeable. In every case, the same gesture: inscribing what counts, circulating it, rereading it to decide.

The Three Operations on the Flow of Observation

The apparatus is organized around three nested operations, which all work on the same material — the datum of the activity — and support one another.

Detailed caption. This diagram presents the static architecture of the apparatus: three nested operations, leading to one purpose.

The frame. The large light rectangle, entitled the activity, contains the three operations. They are not three separate departments but three ways of working one and the same material — the flow of observation that the workshop builds on its own activity, through the cadenced accumulation of drawings.

Level 1 — the quality system (ochre band, number 1). This is the moment when the canal is dug: one collects the inscriptions and chooses the improvements. The ochre colour, throughout this document, marks the datum and its inscription. The quality system formalizes the input, the storing and the follow-up of the data that circulate in the activity; it feeds the workshop’s corpus.

Level 2 — production management (blue band, number 2). This is the moment when the inscriptions become indicators: one exploits the data and follows the state of production. Blue marks exploitation and reading. Production management plans and manages the need on the basis of what the indicators reveal.

Level 3 — the management of internal flows (green band, number 3). This is the moment of situated reading and of action: one optimizes production and controls performance. Green marks the action that stabilizes. It is here that the reading of the indicators is converted into corrective or preventive decisions.

The purpose (navy-blue bar, at the bottom). An arrow goes down from the frame to the bar control of performance — assessment, planning, result. The three operations are worth something only through this end: keeping the activity around its potential, to allow it to absorb fluctuations.

Overall reading. The staggering of the three bands, like steps, states their logical order: one steers (2) only what one has first inscribed (1), and one acts (3) only on what one has known how to read. None of the three is sufficient on its own.

Here one finds again the two functions announced in the foreword. The first operation — the quality system, which collects the inscriptions and controls what circulates — is the quality side. The two following ones — production management and flow management, which read the indicators and act — form the activity side. Presenting and controlling, on one side; steering and, tomorrow, sharing, on the other: the whole volume lies in this articulation, and each section finds its place in it.

The Seven Phases of Setting Up

Installing the apparatus follows an order. One begins by framing the requirements specific to the activity. One then determines the elements that make up each process, then one installs the processes, their records and their checks. Then comes the training of the contributors in the apparatus — not so that they carry it out passively, but so that they become its readers and its improvers. One then installs a simple production management — supply, planning, upkeep — which one optimizes progressively, until one is able to manage the internal flows. Seven phases, then, from the first requirement framed to the management of internal flows; each presupposes the previous one, none is skipped.

These seven phases are those of the deployment of the apparatus. They must not be confused with another sequence, of a different register, which section II will detail: the five phases of evolution, which do not say how one installs the system, but how the structure that will carry it is built — its plan, its team, its market, its attributions, its investment. The first equip; the second build up.

The advantage of this setting up is less control than clarity: clarification of the roles, of the functions and of circulation — of staff, of the product, and of information — in the activity. A structure thus clarified becomes modular: one can open a new production sector in it, or graft a supervision onto it, without undoing what exists.

II. Presenting and Constituting the Activity

Presentation of the Activity

The presentation gathers the data that identify the activity and situate it: its legal form and its company name, its activity code and its establishment identifier, its legal representative and the capacity of the signatory, its approvals, its dates of creation and of last update, and its useful contact details (telephone, email, website).

These fields are not a mere administrative sheet: they form a corpus in the proper sense — a set of data that makes sense together, and in which each datum sheds light on the others. An approval has reach only with regard to an activity; an update date is valid only with regard to what it dates. That is why one keeps the fields that hold together and sets aside those that a tool of a given period imposed without their making sense any more — a means of contact that has fallen into disuse, for example, does not have to appear. The presentation is kept up to date: it carries its date, and ceases to be valid for presenting the activity if the real it describes has changed; the earlier version remains valid for the history.

One point deserves to be noted. No field, in this presentation, names the one who would hold the discipline of the datum for the activity — who would answer for the fidelity of the drawings, the cleanliness of the circuits, the accuracy of the readings. This absence is not an oversight: it is the imprint, in the sheet itself, of a craft that does not yet exist.

The Relational Model and the Lasting Ground

Presenting the activity through its administrative fields does not yet say what it is. An activity is better grasped as a field of relations: it stands at the centre of four links, each of which is an edge through which the datum comes in and goes out. The original apparatus already knew it — it closes on this very diagram, the business at the heart, surrounded by its partners.

Detailed caption. This diagram places the activity at the centre of the four relations that constitute it, and carries the volume’s thesis of status.

The centre (navy blue). The activity itself — management and team — at the heart, coordinating. It is the point from which the four links hold.

The four poles. To the north, the client: downstream, the destination of the product, a relation managed according to the quality aimed at. To the east, the supplier: upstream, the provenance, a relation managed for a constant flow, through the cadence sheets. To the south, the partner — contributor, service provider —, a relation managed through commitment and planned intervention. To the west, the staff: the hands of the workshop, where roles and attributions are distributed, and where the gesture that inscribes is accomplished.

The links (double arrows). Each link is bidirectional: the datum circulates along it in both directions. Its gloss names what passes through — order, need, commitment, attribution. Each of these links is a place of inscription: it is there, in contact with a partner, that the activity produces most of its data.

The gloss (at the bottom). It states the thesis. These four relations form the ground of the activity, and their nature does not change. What changes — products, tooling, regulatory obligations — belongs to form: the tool is modernized, the rule is parameterized, the relation remains. This is what grounds the low volatility of the volume: it documents a movement whose poles are stable.

It is also this diagram that gives meaning to the relational method set out in section I. Managing one’s activity, for a very small structure, is first managing these four relations — and doing so is inscribing, circulating and rereading the datum that is exchanged at each edge.

The Lasting Ground Rests on a Tended Agreed

Saying that the four relations form a ground whose “nature does not change” is right, but does not say everything. This stability is not only that of a fact one observes: it rests on an agreed — the shared language through which one designates a client, an order, a supplier, an address. This agreed is the armature of the activity (see the complementary note Armature: the Canal Dug in Stable Sources): the names, the types, the formats in which the drawings come to be arranged.

This agreed is a stable source, but of a particular kind: a tended stable source (see the note The Envelope: Quality as the Tending of the Agreed). Its stability is not undergone, like that of the height of a mountain, which drifts slowly of its own accord; it is intended — one maintains only what one decides to maintain. And here is the decisive point: its variations are intended too. When a rule changes, when a format evolves, when a technology imposes another form, the agreed is revised, deliberately, so that the agreement holds and follows the fragment of the real that the activity, through its very action, has helped to shift. Where the drift of an undergone source is a danger because it goes unnoticed, the drift of a tended source is the very object of attention: it is anticipated, governed. Tending the agreed is making sure that its imprint still says what it contains.

This is the deepest role of quality. Beyond presenting the activity and controlling what circulates, quality tends the agreed: it keeps the shared armature attuned to the real. The volume already has the gesture — it says that a process is “a canal that one tends and digs again as the activity evolves”. The agreed is the object of a tending of the same nature, but more fundamental still, for it is the common bed on which all the relations build. These conventions are, for an activity as for a society, what foundation ground is for a building: firm enough, at the scale of use, for lasting edifices to be raised on them — even though one knows that the ground will end up moving (Tome 1, The Two Times of the Datum). Identifying them, protecting them, transmitting them: this is part of what quality does.

The Functional Organization

The activity is organized around functions: a management; a management assistance and an accounting function; a production direction (supply, making); a quality function; a distribution direction (shipping, sales, delivery). These functions describe roles and the circulation between them — who receives, who transforms, who ships, who collects payment — and not persons: one and the same person can hold several of them, and the structure is modulated according to size.

It is in this organization that the finding of the corpus is verified. The gestures of the datum are everywhere in it, and nowhere brought together. The reception function inscribes: it assigns the batch number, it opens traceability. The production function circulates and rereads: it carries the batch from one manufacturing order to the next, it reads the indicators and decides. The distribution function closes: it links the arrival inscription to the dispatch note. Each accomplishes a part of the data artisan’s craft — without this craft being named, held, or transmitted as such. Bringing it together would be the work of the canal-keeper; here, it remains distributed.

Scope and Location

The activity is inscribed in a territory: a place of establishment, a zone of clientele served, circuits of supply and distribution. Representing this scope belongs to a generic location map — an anchor point and its service area — and not to a named place: the matrix binds itself to no particular geography. An application for a situated trade will, for its part, fill in the real map; that will be dated form.

The Departments and Their Processes

The apparatus divides the activity into three departments, each carrying its processes.

The accounting department holds the purchasing process (materials, supplies, equipment), the payments process (regular, occasional, irregular) and the collections process (in the shop, on delivery, client invoicing).

The production department holds the three processes of realization — supply, making, distribution — which form the heart of the activity and the heart of this document.

The quality department holds the follow-up process (training, maintenance, traceability, management of non-conformities, improvement proposals, regulatory monitoring), the control process (upkeep, performance, safety) and the improvement process. It is the department that watches over the probity of the inscriptions and the accuracy of the readings — the most visible part of the craft of the datum, although still scattered.

The Five Phases of Evolution

Presenting the activity is fixing its state; constituting it is following the order through which a structure is built and grows. The original apparatus distinguishes, alongside the seven phases of setting up the system (section I), five phases of evolution which do not say how one installs the tool, but how the structure that will carry it is built up. They are read in the light of the relational model: each works one pole of the lasting ground.

First, writing the business plan — structure, functioning, means: the presentation document that serves as a basis for setting up the system, and that places the activity at the centre. Next, training the management and its team — roles, objectives, methods: the internal methodology that allows the coordination of the actors, that is, that equips the staff pole. Then establishing the market plan — products, articles, markets: the choice of partners, of products sold, of clients, which delimits the domain of production and engages the supplier and client poles. Then distributing the attributions — roles, functions, responsibilities: the definition of staffing needs and the distribution of tasks, which consolidates the team and specifies who holds each gesture of the datum. Finally, projecting the investment plan — anticipation, forecasting, planning: the projection of production, of gains and of investments, which defines the evolution of the structure.

These five phases go stale no more than the relations they order. An investment tool will change; the fact of projecting before committing will not change. Like all the rest of the volume, they belong to movement, not to instruments — and that is why they have their place in a matrix with low volatility.

III. The Process

What a Process Is

A process transforms a product into a product with added value, to serve a client or to make the activity function. This product can be goods, a service, a document or a piece of information; the added value, a handling, an intervention, a recording, a storage.

The trilogy makes it possible to read in it what remained implicit. A process is not only a chain that makes a product move forward: it is a place where the activity, which flows like all the rest of the real, becomes datum. At each stage, the work on the material is accompanied by a second gesture, discreet and decisive: one arrests a value — a designation, a quantity, a number, a time — and fixes it on a medium. This gesture is the act of inscription. Depending on what it inscribes, it belongs to one of the three families of constitution: a quantity, a time are drawn from what is happening (observational route); a batch number is instituted under the workshop’s numbering rule; the indicator calculated from these inscriptions is a derived datum. When it draws, it does not pick the datum like a fruit already ripe: it digs a canal in the flow of the activity and draws a datum from it, the elementary grain of a flow of observation that the workshop builds, drawing after drawing, at the cadence of its production.

Hence the shift that governs the whole apparatus: the process describes two twin flowings. That of the product — the circulation of the product. And that of the datum — the inscription that is born at each post, travels from medium to medium, and waits to be reread. The first flowing makes the work; the second makes it governable.

Detailed caption. This diagram re-founds the diagram for describing a process. It keeps the usual structure — incoming product, transformation with added value, outgoing product — and makes the concepts of the trilogy visible in it.

General composition. The diagram is read from left to right, on three horizontal registers. At the top, a light-blue band, the flow of the real: the activity of the workshop flowing with no break of its own, and always exceeding the cuts one makes in it — the product passing by. The two curves that cross it represent this flowing, in which no caesura comes of itself. In the centre, the chain of the product in three blocks. At the bottom, a summary band, the three phases of the inscribed datum.

Stage 1 — the incoming product (left block). The product arrives from its origin; it is received, stored. In ochre italics, the conceptual annotation: on arrival, the real becomes legible — it is a trace, a piece of the real carrying information, but which no one has yet measured. A trace is not a datum: it waits for the act that will draw it.

Stage 2 — the transformation (central block, strong frame). This is the heart of the process, where value is added. Two elements are superimposed in it. The curved shape carries the word canal: the gesture of work does not merely modify the material, it digs a canal in the flow and constitutes a flow of observation — the ordered succession of the workshop’s drawings. Below it, the ochre disc, the act of inscription: the precise moment when a value is arrested and fixed. At the foot of the block, the resources mobilized: workstation, contributors, means.

Stage 3 — the outgoing product (right block). The product comes out with its added value and continues its circulation towards its destination.

The product of inscription — the datum (ochre frame, bottom right). An ochre arrow, starting from the inscription disc, leads to this frame: it represents the drawing. The inscription comes away from the product and from now on exists as a datum of its own. In it one reads what every workshop inscription retains: designation, quantity, batch number, date and time. The mention cadenced accumulation recalls that an isolated datum says nothing: it is their succession, drawn at the rhythm of production, that forms the usable flow of observation.

Bottom band — the three phases of the inscribed datum. It states the guiding thread of the whole document, referring to the three tomes. (1) Making: the act of inscription fixes the value on a medium — this is the record. (2) Circulation: the datum passes from medium to medium — from the binder to the cadence sheet, from the cadence sheet to the slip, from the slip to the screen. (3) Reception: a situated reader reads the indicator it feeds and decides the action. The arrows between the three boxes mark that these phases follow one another: a datum made but that does not circulate, or that circulates but that no one rereads, remains constituted without being actualized, and acts on nothing.

Overall reading. The diagram says in one image the thesis of the document: the process does two things at once — transforming the product and producing the datum. The apparatus only makes explicit the second side, long held to be a mere administrative by-product, and raises it to the rank of an object of its own, with its three phases and its requirements.

The Constituents of a Process

Describing a process is naming its elements. The procedures describe the stages; they are detailed in order to reach the objective systematically. The instructions specify the stages that call for a particular explanation. The documents transmit the information and the production data: they are the media on which the inscriptions are laid down and circulate. The users are the members of staff who implement the procedure — the agents of the gestures of the datum. The storage places designate the places where the procedures, documents and instructions are stored or filed — where the datum rests between two uses, and from where it is taken up to be reread.

These methods are defined by quality and can be improved through improvement proposals. The apparatus is therefore not frozen: it is a canal that one tends and digs again as the activity changes.

The Craft Matrix

Describing the constituents of a process is holding its map. There remains a step that the artisan expects of their work: moving from the map to the helm. This volume has named two matrices. The management matrix — this very volume — documents the relations of an activity: from whom one orders, from whom one gets supplies, with whom one works. The support matrix designates the automatable part, common to every activity — its processes exist, but they are as it were frozen by their generic character: one orders, one collects payment, one follows up everywhere in the same way. There remains what is frozen nowhere: the work proper. The craft matrix is its modelling — not the relations, not the common support, but the conditioned sequence of the singular work: in what order the states of the workpiece follow one another, and under what condition one passes from one to the next.

One clarification keeps it from any misunderstanding. The principle remains: one never touches the heart of the craft. The craft matrix does not model the gesture — designing, assembling, cutting, caring are not dictated —; it models the itinerary of the workpiece: the states it goes through, their sequences, their conditions. It borders the heart, it does not dictate it. That is even its reason for being: giving the gesture proper a legible outline, so that everything that surrounds it — follow-up, the datum, steering — stops weighing on it.

Modelling one’s activity as a craft matrix is declaring in it the stages that convey the flows — material, products, data —, setting in it the criteria that link them, adjusting them to the observational optimum (neither drawing too much nor too little, at each node of the itinerary), and steering the activity along the trajectory laid out. The map said what is; the craft matrix says what must follow, and on what condition.

The shift is slight in appearance, decisive in fact. A described process is read after the fact; a craft matrix is read in advance — it is written, legible, one can follow it, go back up it, suspend it, dig it again. It has all the features of the canal, applied this time not to a single flow of observation, but to the sequence of the canals of an activity. It is neither rigid nor arbitrary: it is the result of the state of an entity which, passed through the prism of the matrix, designates the continuation of the process. The state speaks, the matrix settles, and the way chosen can be read. Its object is control of the activity: choices conditioned by the objectives the artisan has set — never a blind rule, never a throw of the dice.

The Seam Point

Every activity shares with the others a part of its management — the common language of contacts, orders, accounting: the agreed, this tended stable source through which the datum of one activity becomes legible to another. And every activity holds, at its centre, a singular gesture that no other makes the same way, the one that is not automated: designing, assembling, cutting, caring. The first is common and is prescribed; the second is proper and is not dictated.

Between the two runs a junction, and it is there that the highest-value datum is drawn. An accepted order is a drawing from the common register: the agreed expresses it as a quote, an amount, a client. But carrying out what the order lays down is the business of the gesture proper, and the agreed says nothing about it. The seam point is the place where this generic drawing becomes the head of a craft process: the order enters the field of the craft matrix, goes through states in it that recount its carrying out, and comes out of it, once done, handed back to the common register to be invoiced. Generic at the entrance, craft in the middle, generic at the exit.

Each state gone through is a dated act of inscription in the canal of the craft — and these acts, put end to end, form the flow of observation most specific to the activity: no longer what it buys or what it sells, but how it makes. It is the datum that no common register carries, the one that tells the lead times, the reworks, the bottlenecks; the one on which the craft matrix feeds in order to steer, and which, well drawn, well dated, well traced, is a high-value datum in the strictest sense. How far the craft lets itself be modelled in this way, the seam does not impose: a single state — done, not done — is enough for whoever wants only a reference point; a fine matrix serves whoever wants to steer as closely as possible. The artisan chooses their threshold, as they set a lapse: a workshop gesture.

The List of Processes

In all, the apparatus distinguishes: for accounting, purchasing, payments, collections; for production, supply, making, distribution; for quality, follow-up, control, improvement. The three production processes — the object of the next section — are those in which the double flowing, of the product and of the datum, shows itself most clearly.

IV. The Three Processes of the Activity

The three processes follow the product; but it is the datum they inscribe that concerns us. For each, one states the objective — what it does to the product — then the datum it produces and the way that datum lives its three phases.

The Supply Process

Objective: securing the stock. The process links three methods: defining the stock need, placing the order with the supplier, receiving the deliveries.

The definition of the need assesses the state of the stock with regard to the manufacturing schedule, and inscribes a list of needs: for each product expected to run out, a designation, a quantity, a desired date of receipt. It is a drawing from the present state of the stock — dated, and valid for the order it prepares: a list of needs ages as the stock moves.

The planning of orders takes up this list, sorts it by supplier, sets a delivery date, places the order, and inscribes the order placed on the receipt cadence sheet. The cadence sheet is the register in which one keeps, at regular intervals, the cadence of deliveries: a repeated drawing which, accumulated, outlines the supply flow and allows stabilized planning.

The receipt of orders checks the delivery note, inspects the goods, assigns an internal batch number to each incoming product, and completes the cadence sheet. It is here that the trace — the arrival of the product, legible but not yet measured — becomes a datum through the act of inscription. If the check finds a gap, the non-conformity is inscribed in its turn: a gap noted is a gap that can be reread and corrected; a gap not noted is lost.

The Making Process

Objective: transforming the stock. Three methods: recording the client’s order, planning the making, carrying out the making.

The recording of the order creates, for each product ordered, a manufacturing order on the schedule: designation, quantity, date of making, client. The manufacturing order is an inscription of intention — a datum that looks forward, and that will be confronted with the real when the making takes place.

The planning distributes the orders by post and by time slot, and prints the making lists. It is a sorting — the internal circulation of the datum, which passes from the overall schedule to the lists for each post.

The carrying out takes the product from stock, checks its designation and its appearance, copies the internal batch number of the product worked onto the manufacturing order, notes the quantity used, and closes the order. Copying over the batch is the central link of traceability: it is what connects the transformed product to the products that made it up. Closing the order turns an intention into an accomplished fact — a verisimilar datum (“we will make”) into a fresh datum (“we have made, with these batches, this quantity”). The distinction is not purism: an intention guaranteed by a schedule is not a fact guaranteed by the real.

The Distribution Process

Objective: selling off the stock. Three methods: shipping the orders, supplying the shop, selling directly in the shop.

Shipping prints the shipping notes, organizes the delivery round, delivers, and has the note signed by the client. The signature is an inscription of reception: it closes the order and attests, on the recipient’s side, that the product has indeed arrived. It is the last link of the batch thread, and the point where the flowing of the product and the flowing of the datum meet.

Supplying the shop stocks the display as goods flow out, estimates the need for the next order, and adjusts the cadence sheet. The adjustment is a situated reading: the shop manager reads the real outflow, compares it with the planned order, and corrects — a short loop between the datum and the real.

Direct sale serves the client and records the sale: designation, quantity, batch number, time. Each sale is a drawing from the shop’s outflow; their succession feeds the indicator of demand.

V. Traceability

The Principle

Each product entering and leaving the activity must be accompanied by a traceability. The traceability of a product gathers all the information linked to it; it must make it possible to establish the link between the origin and the destination of the product, between the transformed product and the products used to transform it, and between the transformed product and the method employed.

In the vocabulary of the corpus, traceability is the discipline that converts the passage of the product — which, left to itself, would leave only a trace, a piece of the real that is legible but not measured — into a chain of intended inscriptions. The trace is caused: no one decides it. The datum is intended: it is an act, which the procedure provides for and which an agent accomplishes — the intention of the apparatus and the intention of the occurrence that The Trace and the Reader (Tome 1) distinguishes. Traceability is the art of leaving nothing to the chance of the trace, and of bringing everything to inscription.

The Internal Batch Number

The internal batch number is the canonical form of the act of inscription in the workshop. Assigned to each product from reception onwards, it is the mark that tips the trace over into a datum.

Detailed caption. This diagram breaks down the internal batch number and makes its reach explicit.

The token (at the top). The number appears as a succession of segments separated by full stops: three date segments — day, month, year — then a rank segment, a three-digit number between 000 and 999. The date segments are in blue (the stable part, the timestamp); the rank segment is in ochre (the part proper to the inscription). The format is given as a generic template — DD, MM, YY, NNN — and not by a dated value.

The left brace — day · month · year. It groups the three date segments under a gloss: the date of the drawing — the reference freshness. This is the decisive point: the date is not a datum beside the datum, it is inscribed in the number itself. The batch knows when it was taken.

The right brace — 000 → 999. It comments on the rank segment: the rank in the day’s cadence. The rank orders the drawings of one and the same day; it situates the inscription in the cadenced succession of arrivals.

The three properties (ochre bullets). The number is assigned on arrival, to each product: it is the act of inscription, the threshold where the trace becomes datum. It is carried by the product and copied onto each medium it goes through: it is the thread of traceability. It is unaltered, never rewritten: a copy is not a new datum, it is the same datum carried elsewhere — in keeping with the rule of the corpus according to which the inscription is duplicated without the datum splitting in two.

The key idea (ochre band, at the bottom). The whole diagram fits in one sentence: the batch inscribes the date inside the datum; each record thus carries what is needed to judge its validity for a use. This is what distinguishes a well-made datum: not that it is valid for every use and at every time, but that it says when it was taken. Even stale for deciding in the present, it remains exact for the history of the batch.

The Chain of Traceability

The batch number is worth something only if it is taken up at each link. Traceability is this taking up — the thread that runs from one end of the itinerary to the other.

Detailed caption. This diagram follows the batch thread along the itinerary of the product, and shows how each stage re-inscribes the datum.

The upper arc — origin ⟷ destination. A dotted arc connects the first link to the last: it represents the link that traceability has the mission of guaranteeing — being able, from any point, to go back up to the origin and come down again to the destination.

The row of stages. Four stages, plus the recipient, follow one another from left to right, connected by dark arrows — the direction in which the product moves forward. Reception: inspection and unpacking, assignment of the batch number — the ochre gloss underlines it, it is here that the trace becomes datum. Transformation: roughing and finishing, the number is copied onto the manufacturing order, with the quantity used. Preparation: packaging, the number is carried onto the order and the item is labelled. Shipping: loading and delivery round, delivery note, closure and signature. Client: the destination, where the itinerary ends.

The batch thread (ochre bar, at the bottom). A bar crosses the four stages; an ochre dotted line, marked by a point under each stage, materializes the same number, re-inscribed at each link. It is not a new number each time: it is the same one, taken up — proof that the datum has travelled without breaking.

The gloss. It states the principle: without inscription, the passage of the product would leave only a trace; traceability converts this passage into a chain of intended inscriptions, and links the origin and the destination, the transformed product and what transformed it, and the method employed. Final warning: the thread holds only if each link re-inscribes the batch — a stage that does not inscribe breaks the chain, and nothing of what follows connects to the origin any more.

Basic Traceability and Complete Traceability

Two depths are distinguished. Basic traceability gathers the information on the circulation of the product, from primary production to use: each contributor who works on the product completes its traceability by indicating the designation, the supplier, the quantity worked and the recipient. It is enough to inform the inspection services and production management. Complete traceability adds the descriptive information that details the original product — class, place of production, category — and makes it possible to guarantee the quality of the product to the client.

In both cases, the gesture is the same: drawing, dating, fixing, linking. The difference is not of nature but of resolution — how much detail one draws. And this resolution is, it too, an arbitration of the observational optimum: too little, and one is blind to the defect; too much, and one drowns. The right level depends on the use and can be reopened each time.

VI. Steering and Sharing

This section carries the activity side of the volume. It unfolds in two phases, each of which extends the other. First steering: constituting, for the workshop, the flows of observation that conduct it — this was already the object of Project 00 of 2007, and the notions below take it up. Then sharing: recognizing that these same flows, once qualified, are worth something beyond the workshop, and open onto the public proposal of the corpus. The first phase is inherited; the second is the contribution of this re-edition.

Basic Notions

Flow management is the tool through which the activity determines its yield, situates its effective capacity, and assesses its projects. It follows from the quality system and from production management: it is what determines the actions to be applied to stabilize productivity. Its objective is twofold — optimizing productivity and controlling performance — and its method comes down to three words: stabilization, anticipation, flexibility. Stabilizing around a sufficient threshold of yield; anticipating gaps before they become critical; keeping enough flexibility to absorb fluctuations.

The Three Flows

The activity evolves around three flows. Distinguishing them is understanding that the same product is read in three ways.

Detailed caption. This diagram presents, on three parallel lanes, the three flows through which one and the same product can be followed across the stages of the activity.

The stages (columns). In the background, four column markers — reception, transformation, preparation, shipping — mark out the itinerary, common to the three lanes.

Lane 1 — the product flow (blue, at the top). Established by the designations given to the products sold, it corresponds to the circulation of the product: it says what. The blue line follows it from one end to the other.

Lane 2 — the material flow (slate, in the centre). Established by the quantities distributed, it corresponds to the distribution of the product: it says how much. The slate line runs parallel to the first.

Lane 3 — the data flow (ochre, at the bottom). Established by the internal batch number filled in at each stage, it corresponds to the evolution of the batch. The ochre line is marked by a point at each stage — the successive inscriptions of the same batch. It is the lane that the other two presuppose.

The right brace. It connects the product lane and the data lane: it represents the fact that the first two flows can be read only through the third. Without the batch number filled in everywhere, neither the designations nor the quantities would connect from one stage to the next.

The gloss. The product is single; one follows it from three points of view. The product flow says what, the material flow says how much — but these two become legible and connected only through the data flow, the batch number filled in everywhere. It is this third flow — the datum itself — that processual ontology takes as its object; the other two depend on it.

The Breakdown of the Data Flow

The data flow has its own cycle, which replays, at the scale of the workshop, the three moments of the trilogy.

Detailed caption. This diagram breaks down the cycle of the datum in the workshop into five stages, distributed under the three phases of the trilogy, and leading to three outcomes.

The band of phases (at the top). Three coloured segments cover the stages: making (ochre) above recording; circulation (green) above collection and sorting; reception (blue) above reading and action. The workshop’s cycle is thus set against the three phases of the datum.

The five stages (numbered row). Connected by dark arrows: (1) Recording — the inscription is fixed. (2) Collection of the data. (3) Sorting of the data. (4) Reading the indicators — the situated reader interprets. (5) Resulting action — the decision is taken. Stages 1 to 3 prepare the material; the meaning of the cycle is played out in 4 and 5, at reception.

The fork towards the three outcomes. From the action (5) a fork leads to three possible outcomes, according to what the reading reveals. Continuity (green) — productivity as expected: the datum follows the real, nothing to correct. Adjustment (amber) — preventive action, triggered by a lapse that is announcing itself, such as the end of a batch or a late order: one acts before the gap. Correction (terracotta) — corrective action, triggered by an ascertained gap, such as a stock-out or an order cancellation: one repairs after the fact.

The return loop — the period (on the left). A dotted ochre arrow goes back from the outcomes to recording, closing the cycle. At its foot, a stylized clock carries the word period: the cadence of the flow. The cycle does not take place once but at each period, at the rhythm of the workshop — it is this cadence that makes the apparatus a living flow of observation, and not a collection of archives.

The gloss. At each period, the workshop draws (1), circulates (2–3) and rereads (4–5) its datum; reading decides: maintaining, anticipating an announced lapse, or correcting an ascertained gap. Warning: a datum recorded but not reread does not act — the cycle closes only at reception. This is the lesson of the third tome applied to the workshop: a datum that no one reads does not cease to be a datum — it remains constituted, but not actualized.

The Indicator / Action Loop

Reading, at stage 4, is not done on the raw datum but on an indicator: a derived datum, calculated from the inscriptions and shaped so as to make the gap visible. It is the point of contact between the datum and the decision.

Detailed caption. This diagram shows how an indicator is read and triggers an action, on the model of a control chart.

The graph (on the left). The horizontal axis is time — the successive drawings, at the cadence of the workshop. The vertical axis carries the value followed. Three zones are superimposed: in the centre, a green band, the range of tolerance; on either side, amber bands of transition; beyond, terracotta bands. Two red dotted lines mark the upper and the lower tolerance threshold; a thin central green line marks the target — designated as the observational optimum. A curve connects the successive drawings: its points are green as long as the value stays within tolerance, amber when it approaches it, terracotta when it crosses a threshold. The curve represents a drift: the value rises steadily, crosses the alert zone, then leaves the tolerance.

The panel of readings (on the right). Three cases, each pairing a state of the indicator with a response. Continuity (green): value within tolerance, the real follows the expected datum — nothing to correct. Adjustment — preventive action (amber): the value approaches the threshold, a lapse is announcing itself (end of batch, late order). Correction — corrective action (terracotta): the threshold is crossed, the real has moved away from the datum (stock-out, cancellation).

The gloss. It holds the conceptual thread. The indicator is not production: it is a succession of drawings that someone must read for it to act — without a reader, the indicator does nothing. The green band is the workshop’s observational optimum — the gap tolerated between the datum and the real for the intended use. Crossing a threshold does not say that the datum is false: it says that the real has moved too far from it — it is lapse that speaks. Last warning: setting the threshold too tight drowns the workshop in alerts, too loose makes it blind; the right setting is a question, not a value.

Origin and Reading of the Flows

The three flows are built by recording at the phases of the product’s flowing: the internal batch assigned on arrival contains the information on distribution, use and evolution; combined with pre-recorded data, it feeds the indicators. Recording can be manual or automatic according to the intensity of the flowing — this is, moreover, one of the points where a tooled application will take over: automating inscription where the cadence requires it, without ever dispensing with reading nor with the arbitration that follows it, for which a situated reader answers. The indicator defines an interval of tolerance; its visualization — control chart, figure, alert, warning light, curve — establishes the state of production. The form of the visualization is contingent; what it serves — making the gap legible in time — is not.

One last case of gap deserves its place here, for the tooled application will make it ordinary: the activity can keep several workbenches — the workshop’s, the worksite’s — and inscribe on both sides while they are separated. When they join up again, it happens that the same thing has been inscribed twice, differently. This divergence is neither a fault nor a false datum: two dated findings follow one another, and the more recent carries the current reading — the other remains, with its date and its workbench. But it is a signal for steering, in the same way as a crossed threshold: the gap is made legible, with its two inscriptions, and it is the situated reader who judges — confirming, correcting, or leaving it. The machine ascertains; it never settles in silence. Arbitrating a divergence is a moment of the craft, not a plumbing problem.

The Sharing of High-Value Data

Local steering does not exhaust the value of the flows of observation. Project 00 of 2007 stopped there: drawing, inscribing, rereading in order to conduct one’s own workshop. But once these flows are qualified, they are worth something beyond the activity that produced them. It is the contribution of this re-edition to extend the same logic one notch further: from steering to sharing.

What is a high-value datum? Not a rare or secret datum, but a datum whose making is controlled: well drawn — one knows its intention, its criteria, its fidelity; well dated — it carries its freshness, and one can judge for what use it is still valid; well traced — a chain of inscriptions connects it without a break to its source. Such a datum lets itself be interpreted by someone other than the one who made it, because it says where it comes from and what it is worth. This value remains potential: it depends on the making, and is realized only in reading, in a use. One recognizes here, exactly, what the quality side of this volume produces: presenting, controlling processes, keeping traceability. Quality is therefore not a prerequisite separate from sharing — it is its condition. One shares usefully only what one has first made well.

To this condition is added another, of the same origin. For a datum to let itself be read by another, it is not enough that it says where it comes from: it must also be said in a shared agreed — an address formed as the other expects it to be, an order with fields recognized on both sides. This is standardization: the agreement on the A, without which two workshops, each showing probity in its own corner, would remain inaudible to each other. This agreed is a commons — no one owns it, and it holds only through the tending of those who use it (here one finds again the governance of the commons and the tending of referents of the third tome). Tending it is therefore, for the data artisan, a gesture of sharing before the fact: it is preparing the ground where data will be able to meet.

Not all data are to be shared, and discernment is the same here as elsewhere: an observational optimum. Sharing everything would be drowning oneself and drowning others; sharing nothing would be depriving oneself of a return. The question is asked again each time: this datum, for whom, does it gain by circulating? The artisan chooses what has value for a common use, and leaves the rest to the workshop.

Sharing obeys a simple principle, which belongs to substance and does not go stale: the datum stays as close as possible to its source, and one makes travel only what must travel. It is not the raw data of an activity that are pooled — they stay with whoever holds them, under their control — but what can be transposed from them: lessons, reference points, models. The very small structure thus keeps control of what concerns it; it does not cede its canal, it links it. The technical form of this linking — how, with which tools, the datum teaches without moving — belongs to the twin volume, Technical Data; here, its principle and its governance are set out.

From this sharing a circle is born. High-value data, pooled by several activities of one and the same territory, feed a collective learning; this learning gives a better model — more accurate, and more sober, because it rests on drawings made with probity; and this model comes back to the client, who finds it again in tools better adjusted to their trade than any generic tool would be. The contributor is not dispossessed: they are the first beneficiary of what they helped to constitute. It is this virtuous circle — high-value data, learning, a more accurate model, return to the client — that the public proposal of the corpus carries.

Three safeguards keep it straight, and they are of the same spirit as the whole volume. Sharing is voluntary and reversible: at any moment, the structure can take back control without losing anything of what is its own — its sovereignty over its own datum is not negotiable. Arbitration remains imputable: the model proposes, the situated reader decides and answers for it; no pooled return dispenses the artisan from rereading what concerns them. And one rewards the right gesture, not the raw figure: pooling must not push towards producing data for the sake of data, but towards drawing better what counts.

There remains, here again, the craft. Holding the local canal and linking it to the common canal — drawing with probity, qualifying, choosing what has value, guaranteeing that nothing circulates without intention or without return —, is one and the same gesture, which calls for one and the same hand. This hand, the corpus names it: the data artisan, the canal-keeper. At the scale of a territory, federating these artisans in the service of very small structures is precisely the object of the proposal presented on artisanat-de-la-donnee.fr — a proposal that this volume, through its matrix, puts in a position to be equipped. The craft remains to be created; but one sees here, more clearly still, the work that awaits it.

VII. Production Data

The Workshop’s Corpus

Beyond the data that circulate in the course of production, the activity rests on a body of data that describe and situate it: its productive organization, its staff, its partners, its means, its circuits, its product categories, its critical points. This whole is not a pile of sheets: it is a corpus in the sense of the corpus — a set of data that makes sense together, and whose value depends on coherence. A datum is worth something through the corpus that surrounds it; isolated, it says little. That is why these data are kept together, and updated together.

This corpus divides into two regimes of freshness. One part belongs to the stable sources: the plan of the premises, the list of means, the staff, the partners, the product categories evolve slowly, and one can rely on them for a time without re-drawing them — relatively to use, never absolutely, for there is no outside-the-flow. The other part flows fast: production curves, the day’s quantities, non-conformities. Keeping the corpus is knowing which of its data sleep and which stay awake — and re-drawing the latter at the cadence their real imposes.

The Sheets of the Corpus

The sheets that follow are generic templates, to be filled in for a given activity. Their structure is given here — the columns, and therefore the criteria of inscription — without a dated value. Filling in a sheet is digging a precise canal in the real of the activity.

Production follow-up. For each period, the quantity of product sold, compared with the previous production, and the curve of evolution. It is the basic drawing on which the indicators feed.

Period Quantity sold Reference (previous period) Gap ——– ————— —————————– —–

Staff. The contributors and their post — the map of the agents of the gestures of the datum.

Name Post Functions in the flow of the datum —— —— ————————————

Partners. Suppliers, contributors, service providers — the other hands through which the datum and the product pass.

Name Function Contact —— ———- ———

Means. The equipment, its characteristics, its maintenance — the tools that extend the hand and carry a part of the know-how.

Category Means Characteristics Supplier Maintenance Use ———- ——- —————– ———- ————- —–

Supply and distribution circuits. Upstream, the provenance of the materials and the share of each; downstream, the destination of the products. These are the two ends of the thread — the origin and the destination that traceability links.

Product categories. For each category, the type of order, the place of storage, the use-by date — so many criteria that fix the freshness specific to each product and therefore the cadence of its restocking.

Category Type of order Place of storage Use-by date ———- ————— —————— ————-

The Productive Organization and the Disposals

The productive organization links the phases — reception, first to fourth transformation, distribution — between which the product passes through work in progress, and outside which the waste leaves. Each phase is a process, and is described by the diagram already given: an incoming product, a transformation with added value, an outgoing product, and the inscription that accompanies them. The production diagram and the disposals diagram are, from this point of view, only sequences of the process diagram — each post carries in them its operating time and its specific datum, that is, its own drawing.

The Analysis of Critical Points

The activity finally identifies the points where a gap would be serious, and installs there in advance a check, a tolerance, and the actions to be kept ready. It is the grid in which the workshop sets, before the incident, where to draw and what tolerance to accept — its codified observational optimum.

This grid deserves to be read in the light of the corpus. The check is a drawing: one observes the real at a chosen point. The tolerance is the band of the observational optimum: the gap admitted between the expected datum and the real. The preventive action responds to a lapse that is announcing itself; the corrective action, to an ascertained gap. To the critical points is attached a list of good practices of work: not abstract rules, but the proven gestures that keep the real within tolerance — the part of the know-how that, in an established craft, would be transmitted from a master to an apprentice.

Opening

This volume implements, on the ground of entrepreneurship in 2026, a quality and activity management system reread in the light of the processual ontology of the datum. It has plastered nothing on: it has recognized, in an apparatus that described the activity by the itinerary of its information, the same intuition that the trilogy carried to the concept — the datum as a dated drawing from the flow of the real (on its observational route), whose validity is judged for a use, which realizes its value only in reading, and which exists only once inscribed.

From this rereading, three established results remain, which belong to substance and do not go stale. First, that a process always does two things: transforming a product, and producing a datum — and that the second is not an administrative by-product, but an object of its own, with its three phases. Next, that traceability is the discipline that converts the passage of the product into a chain of intended inscriptions, and that the batch number is its canonical form — a datum that carries its own date, from which its validity for each use is judged. Finally, that steering an activity is reading indicators in time, within a band of tolerance that is an observational optimum, and acting before or after the gap according to what the reading reveals.

All the rest — a named craft, a tool, a figure, a situated map — is form, and will come downstream. But this volume is not, for all that, a timeless base: it is a dated matrix, resting on a stable ground. Its form will be revised, but slowly, at the rhythm at which the relations it documents change — that is, little. Versions by trade will be derived from it, which will fill in the sheets left as templates and illustrate the substance with dated examples; it will be used to equip an activity, to automate inscriptions where the cadence requires it, and to supervise the whole from a coherent view.

This work is not done alone. For the technical part — with what and how to dig the canal —, this volume refers to its twin, Technical Data; together, they answer the two questions of every assignment. And for the collective part — once the movement is well documented —, it opens onto the sharing of high-value data and, through it, onto the proposal that the corpus carries: federating data artisans in the service of very small structures, as close as possible to the territory.

There remains the craft. Throughout this volume, one has seen the gestures of the datum at work — inscribing, circulating, rereading, deciding — distributed among functions that pursue other ends and do not hold them together. Bringing them together, naming them, transmitting them in degrees: that would be the work of the data artisan — the canal-keeper of the continuity volumes —, whose craft the trilogy calls for, and which no one, for the time being, practises. This volume does not create this craft. It sets out its workshop plan, and shows, sheet after sheet, the scale of the work that awaits the hand that will be willing to take it up.

Translation choices

These notes say how to read the English words that render the terms of the corpus; they are fixed for the whole English translation by the Translation Charter of the Corpus.

• datum / data. The corpus treats the datum as an individual: constituted by an act, dated, carried by media. The translation therefore says datum in the singular and data as its plural.
• drawing / to draw (prélèvement, prélever): the gesture of drawing from the flow, as one draws water; re-draw renders reprélever. Input renders saisie, the gesture of keying a datum in. The apparatus (dispositif) is the whole set-up this volume describes; device keeps its everyday sense.
• the agreed; armature; tended stable source (le convenu; l’armature; source stable entretenue): the shared language in which an activity draws, the structure drawn from it, and the kind of source whose stability is intended rather than undergone. Tending and to tend render entretien and entretenir; upkeep renders the entretien of equipment, and maintenance the French maintenance.
• management matrix; support matrix; craft matrix (matrice de gestion; matrice de soutien; matrice de métier): the relations of an activity, its automatable common part, and the conditioned sequence of its own work. The heart of the craft (cœur de métier) is the client’s gesture, which the craft matrix borders and never dictates. The seam point (point de couture) is where a drawing from the common register (registre commun) becomes the head of a craft process.
• high-value datum; potential value (donnée à haute valeur; valeur potentielle): well drawn, well dated, well traced, as the canonical lexicon defines it; its value is potential until a reading realizes it.
• cadence sheet (cadencier): the register that keeps the cadence of orders or deliveries. Batch and batch number render lot and numéro de lot.
• lapse; stale (péremption; périmée): a relation, not a decay; a datum goes stale for a use, and validity for a use is the measure of it. Gap renders écart; finding and ascertain render constat and constater.
• trace (trace): the mark that the real leaves of itself, legible but not yet measured (relevé); it becomes a datum only through an act of inscription. Traceability renders traçabilité.
• path; itinerary; route (chemin; parcours; voie): path is kept for the three paths of non-reading; itinerary renders parcours (the itinerary of the product, of the workpiece); route renders the observational voie, way the ordinary chemin. The lanes of the diagram of the three flows render its voies.
• The data artisan (l’artisan de la donnée): the referent whose domain would be the datum itself, a craft yet to be created; the canal-keeper (canalier) is its name in the continuity volumes.
• Diagrams. The labels and captions of the diagrams are translated; their layout is unchanged. The template of the batch number becomes DD.MM.YY.NNN.
• Titles. Tried by the Flow renders À l’épreuve du flux; this volume is the SGQA — Quality and Activity Management System; its twin is Technical Data. Project 00 renders Projet 00.