The classes and properties minted for this export, all in the LADO
namespace http://archaeology.link/ontology#. Everything below is read from the code at generation
time; the definitions are the same strings that appear as English
rdfs:comment in rdf/lado_dating_extension.ttl.
Nothing here duplicates a term that CIDOC CRM or OWL-Time already provides. Where a standard term exists it is used directly — see the crosswalk.
lado:LocationLocation — subclass of crm:E53_Place
An existing LADO class, anchored here beneath crm:E53_Place so that consumers who reason over CIDOC CRM can reach it without knowing LADO.
lado:DiscoverySiteDiscovery site — subclass of lado:Location
An existing LADO class from the published loc_discoverysite dataset. This export references such nodes; it never re-types or redefines them.
lado:FindspotFindspot — subclass of lado:Location
A named context within a discovery site, such as a pit, a ditch or a destruction layer. It falls within its site via crm:P89_falls_within. A site may carry several findspots, dated separately and sometimes far apart — Bregenz is the standing example — which is why the dating hangs on the findspot and not on the site.
lado:DatedTimeSpanDated time-span — subclass of crm:E52_Time-Span, time:ProperInterval
A time-span inferred from dated material rather than recorded directly. Sits beneath both crm:E52_Time-Span and time:ProperInterval so that CIDOC-based and OWL-Time-based queries each find it by their own route.
lado:FindspotDatingFindspot dating — subclass of lado:DatedTimeSpan
The dating of one findspot derived from samian potters’ stamps. The interval is a ‘virtual fuzzy year’ of the form m ± k·σ and is explicitly NOT a confidence interval: it carries no probabilistic claim about where the true date lies.
lado:DatingInstantDating instant — subclass of time:Instant, crm:E52_Time-Span
One boundary of a dated interval. A local class rather than a bare time:Instant so that it can be anchored in CIDOC CRM without making any claim about time:Instant in general: crm:E52_Time-Span is the CRM notion of a temporal extent, and this one has zero duration. Every instance of an application class in this export reaches CIDOC CRM, and the boundaries are no exception — without the anchor a CRM-only consumer would see a time-span whose ends did not exist.
lado:DatingTimePositionDating time position — subclass of time:TimePosition, crm:E54_Dimension
The numeric value of a boundary together with the scale it is read on. Structurally a crm:E54_Dimension: time:numericPosition corresponds to P90 has value and time:hasTRS to P91 has unit. Kept as a local subclass so the alignment is stated here rather than asserted about OWL-Time itself.
lado:YearScaleYear scale — subclass of time:TRS, crm:E73_Information_Object
The temporal reference system the numeric positions are read on. A documented convention, and therefore a crm:E73_Information_Object. Its definition matters because the positions are signed years on a continuous line, not calendar labels — see eraConvention.
lado:DatingModelDating model — subclass of prov:Plan, crm:E29_Design_or_Procedure
The parameterisation from which the intervals were computed. Holds k_min, k_max, τ, w and the fuzziness divisor, so a reader can recompute any interval from the exported figures alone. It sits beneath prov:Plan and beneath crm:E29_Design_or_Procedure, the CIDOC CRM class for a documented procedure; without the second parent this would be the one local class that never reaches CIDOC CRM, and a CRM-only consumer could see the datings but not the method behind them.
lado:DatingActivityDating activity — subclass of prov:Activity, crmdig:D10_Software_Execution
The computation that produced one dating. It sits beneath crmdig:D10_Software_Execution, whose scope note describes precisely this: a run completely determined by its digital input, the software and the generic properties of the device. Using CRMdig rather than reaching for crm:E7_Activity directly means the alignment has been done by the extension’s authors — D10 leads through D7 and E11/E65 to E7_Activity — instead of this vocabulary deciding on its own whether a script run counts as an action intentionally carried out by an actor.
lado:FigureFigure — subclass of crm:E36_Visual_Item
A published figure. Carries the constants that belong to the drawing rather than to the archaeology — padding, row height, colour ramp, sort order — so that they are stated once instead of being repeated on every row.
lado:PlotRowPlot row — subclass of crm:E36_Visual_Item
The presentation layer of a single dating. It deliberately carries the uncertainty whiskers, which the method documentation describes as visual only, keeping them off the time-span where they would read as an archaeological claim.
lado:IndependentTerminusIndependent terminus — subclass of crm:E5_Event
An event whose date is known from something other than samian — a coin series, a dendrochronological sequence, a volcano, the historical record — and which therefore bounds the assemblage found with it. The class names the event’s ROLE in the dating procedure rather than its kind: an eruption and the abandonment of a fort are both events, and both become termini here because their dates come from outside the material being dated. Publishing them closes a gap: the graph used to say that the model was calibrated against five findspots without ever saying against what.
lado:ColourAxisColour axis — subclass of crm:E36_Visual_Item
A complete specification of how one measured quantity becomes a colour: the ramp stops, the interpolation between them, the domain the values are stretched onto and whether that domain is a fixed convention or was read off this corpus. Colour is presentation rather than evidence, so it is published here on the same terms as the other presentation-only quantities: as something a consumer can check and recompute rather than has to trust.
| Property | Domain | Range | Definition |
|---|---|---|---|
lado:renders |
lado:PlotRow |
lado:DatedTimeSpan |
Links a plot row to the dating it depicts. Following this property in reverse tells you whether a given dating appears in any published figure. |
lado:hasRow |
lado:Figure |
lado:PlotRow |
Links a figure to the plot rows it contains. |
lado:undefinedMeasure |
lado:DatedTimeSpan |
<http://www.w3.org/1999/02/22-rdf-syntax-ns#Property> |
Names a measure that could not be computed for this subject. Under the open-world assumption a missing triple is merely unknown; this property distinguishes ‘we computed it and it does not exist’ from ‘nobody has looked yet’. |
lado:constrains |
lado:IndependentTerminus |
lado:Findspot |
The assemblage a terminus sets a bound on. Following it leads from the fixed point to the dating it constrains, and back. |
lado:hasColourAxis |
lado:Figure |
lado:ColourAxis |
Relates a figure to a colour axis its rows can be shaded by. A figure carries several, because sharpness, chronology, dispersion and sample size answer different questions and choosing between them belongs to whoever is drawing the map. |
lado:colourSource |
lado:ColourAxis |
<http://www.w3.org/1999/02/22-rdf-syntax-ns#Property> |
The measured property an axis takes its values from. Following it leads from the colour back to the evidence. |
lado:normalisedProperty |
lado:ColourAxis |
<http://www.w3.org/1999/02/22-rdf-syntax-ns#Property> |
The property under which this axis’s normalised value appears on a plot row. Named in the graph so that a client can discover the axes rather than having the property names compiled into it. |
lado:hexProperty |
lado:ColourAxis |
<http://www.w3.org/1999/02/22-rdf-syntax-ns#Property> |
The property under which this axis’s colour appears on a plot row. See normalisedProperty for why it is named rather than assumed. |
lado:possiblyBefore |
lado:DatedTimeSpan |
lado:DatedTimeSpan |
Holds on the modelled interval: this time-span ends before the other begins, with a gap. Every pair of datings stands in exactly one relation of this family, so the set is complete and can be counted. It does not claim that the relation survives a more generous reading of the bounds — where it does, time:intervalBefore is asserted as well, and that property is declared a subproperty of this one. |
lado:possiblyAfter |
lado:DatedTimeSpan |
lado:DatedTimeSpan |
Holds on the modelled interval: this time-span begins after the other ends, with a gap. Every pair of datings stands in exactly one relation of this family, so the set is complete and can be counted. It does not claim that the relation survives a more generous reading of the bounds — where it does, time:intervalAfter is asserted as well, and that property is declared a subproperty of this one. |
lado:possiblyMeets |
lado:DatedTimeSpan |
lado:DatedTimeSpan |
Holds on the modelled interval: this time-span ends exactly where the other begins. Every pair of datings stands in exactly one relation of this family, so the set is complete and can be counted. It does not claim that the relation survives a more generous reading of the bounds — where it does, time:intervalMeets is asserted as well, and that property is declared a subproperty of this one. |
lado:possiblyMetBy |
lado:DatedTimeSpan |
lado:DatedTimeSpan |
Holds on the modelled interval: this time-span begins exactly where the other ends. Every pair of datings stands in exactly one relation of this family, so the set is complete and can be counted. It does not claim that the relation survives a more generous reading of the bounds — where it does, time:intervalMetBy is asserted as well, and that property is declared a subproperty of this one. |
lado:possiblyOverlaps |
lado:DatedTimeSpan |
lado:DatedTimeSpan |
Holds on the modelled interval: this time-span begins first, ends inside the other. Every pair of datings stands in exactly one relation of this family, so the set is complete and can be counted. It does not claim that the relation survives a more generous reading of the bounds — where it does, time:intervalOverlaps is asserted as well, and that property is declared a subproperty of this one. |
lado:possiblyOverlappedBy |
lado:DatedTimeSpan |
lado:DatedTimeSpan |
Holds on the modelled interval: this time-span begins inside the other and outlasts it. Every pair of datings stands in exactly one relation of this family, so the set is complete and can be counted. It does not claim that the relation survives a more generous reading of the bounds — where it does, time:intervalOverlappedBy is asserted as well, and that property is declared a subproperty of this one. |
lado:possiblyStarts |
lado:DatedTimeSpan |
lado:DatedTimeSpan |
Holds on the modelled interval: this time-span begins together with the other and ends first. Every pair of datings stands in exactly one relation of this family, so the set is complete and can be counted. It does not claim that the relation survives a more generous reading of the bounds — where it does, time:intervalStarts is asserted as well, and that property is declared a subproperty of this one. |
lado:possiblyStartedBy |
lado:DatedTimeSpan |
lado:DatedTimeSpan |
Holds on the modelled interval: this time-span begins together with the other and outlasts it. Every pair of datings stands in exactly one relation of this family, so the set is complete and can be counted. It does not claim that the relation survives a more generous reading of the bounds — where it does, time:intervalStartedBy is asserted as well, and that property is declared a subproperty of this one. |
lado:possiblyDuring |
lado:DatedTimeSpan |
lado:DatedTimeSpan |
Holds on the modelled interval: this time-span falls strictly inside the other. Every pair of datings stands in exactly one relation of this family, so the set is complete and can be counted. It does not claim that the relation survives a more generous reading of the bounds — where it does, time:intervalDuring is asserted as well, and that property is declared a subproperty of this one. |
lado:possiblyContains |
lado:DatedTimeSpan |
lado:DatedTimeSpan |
Holds on the modelled interval: this time-span strictly encloses the other. Every pair of datings stands in exactly one relation of this family, so the set is complete and can be counted. It does not claim that the relation survives a more generous reading of the bounds — where it does, time:intervalContains is asserted as well, and that property is declared a subproperty of this one. |
lado:possiblyFinishes |
lado:DatedTimeSpan |
lado:DatedTimeSpan |
Holds on the modelled interval: this time-span begins later and ends together with the other. Every pair of datings stands in exactly one relation of this family, so the set is complete and can be counted. It does not claim that the relation survives a more generous reading of the bounds — where it does, time:intervalFinishes is asserted as well, and that property is declared a subproperty of this one. |
lado:possiblyFinishedBy |
lado:DatedTimeSpan |
lado:DatedTimeSpan |
Holds on the modelled interval: this time-span begins earlier and ends together with the other. Every pair of datings stands in exactly one relation of this family, so the set is complete and can be counted. It does not claim that the relation survives a more generous reading of the bounds — where it does, time:intervalFinishedBy is asserted as well, and that property is declared a subproperty of this one. |
lado:possiblyEquals |
lado:DatedTimeSpan |
lado:DatedTimeSpan |
Holds on the modelled interval: this time-span has the same bounds as the other. Every pair of datings stands in exactly one relation of this family, so the set is complete and can be counted. It does not claim that the relation survives a more generous reading of the bounds — where it does, time:intervalEquals is asserted as well, and that property is declared a subproperty of this one. |
dcat:Dataset| Property | Range | Definition |
|---|---|---|
lado:identifierScheme |
xsd:string |
The recipe by which findspot URI fragments are built. Any further implementation must reproduce it character for character, including the Unicode normalisation step, or it will mint different URIs for the same findspot. |
lado:ColourAxis| Property | Range | Definition |
|---|---|---|
lado:rampName |
xsd:string |
The name of the colour scale. Deliberately distinct from colourRamp, which holds the name of a D3 function that this export does not reproduce. A ramp named here is fully specified by its stops and its interpolation, and the resulting colours are reproducible from the graph without any library. |
lado:rampStops |
xsd:string |
The anchor colours of the scale, as hex values separated by spaces, running from the low end of the domain to the high end. |
lado:rampInterpolation |
xsd:string |
How colours between two anchors are obtained. ‘linear-rgb’ means each channel is interpolated linearly. This is not what D3’s spline interpolators do; the two agree at the anchors and differ by a few units between them. |
lado:domainMin |
xsd:decimal |
The value mapped to the low end of the ramp. Values below it are clamped rather than dropped. |
lado:domainMax |
xsd:decimal |
The value mapped to the high end of the ramp. |
lado:domainBasis |
xsd:string |
Whether the domain is a fixed convention or was observed in this corpus. The distinction matters for anyone comparing two snapshots: an observed domain moves when findspots are added, so an unchanged findspot can change colour without its dating having changed at all. |
lado:domainScale |
xsd:string |
Whether values are stretched linearly or logarithmically. Logarithmic where the quantity spans orders of magnitude, so that the ramp is not exhausted by a handful of large sites. |
lado:DatedTimeSpan| Property | Range | Definition |
|---|---|---|
lado:nStamps |
xsd:integer |
The number of stamps recorded at the findspot. Note that this is not the n used by the k formula — see nStampsWithDie. |
lado:nStampsWithDie |
xsd:integer |
The number of stamps that carry a die attribution. This is the n in the k formula. It is smaller than nStamps because the underlying query counts only rows where a die is recorded. Substituting nStamps here yields a different k from the one actually used, which is why both are exported. |
lado:nDies |
xsd:integer |
The number of distinct dies attested at the findspot, counted over the pair (potter, die). |
lado:dieRepetition |
xsd:decimal |
nStampsWithDie divided by nDies. High values indicate that the same dies recur, which is characteristic of a hoard or a closed deposit rather than of accumulated site refuse. |
lado:qRepetition |
xsd:decimal |
1 − 1/repetition, the second quality axis. Deliberately not combined with qInterval: a findspot that is not a deposit is not thereby badly dated, and averaging the two would assert exactly that. |
lado:qInterval |
xsd:decimal |
The first quality axis, describing how tightly the contributing potter datings agree. It divides by the interval width, which is a difference and therefore carries no calendar origin. |
lado:qStart |
xsd:decimal |
Sharpness of the start date, exp(-sigma / referenceLength). Earlier versions divided by the mean calendar value, which penalised material dated near the era boundary for its position in the calendar rather than for any property of the evidence; since v27c the denominator is a fixed length and the measure is independent of epoch. The earlier form is kept in the source query as q_start_legacy, for reconstructing published figures. |
lado:qEnd |
xsd:decimal |
Sharpness of the end date, on the same scale as qStart. |
lado:sigmaYears |
xsd:decimal |
The dispersion entering the interval width, obtained by variance decomposition: the internal fuzziness of each potter’s range plus the scatter of the range midpoints. Together with kFactor it determines the width of the box, which is 2·k·σ. |
lado:kFactor |
xsd:decimal |
The volume-based multiplier applied to sigmaYears. It falls from k_max towards k_min as evidence accumulates, so a findspot with many stamps receives a narrower interval than one with few. |
lado:kNoDieRecord |
xsd:boolean |
True where no die attribution exists at all for the findspot. Up to revision 30a this condition set k to k_max and widened the interval for a reason unconnected with the material; it now has no effect on the dating and records a gap in the die record instead. |
lado:midpointYear |
xsd:decimal |
The centre of the averaged interval, the m in m ± k·σ. |
lado:avgDatemin |
xsd:integer |
Mean of the contributing potters’ start dates. |
lado:avgDatemax |
xsd:integer |
Mean of the contributing potters’ end dates. |
lado:minDatemin |
xsd:integer |
Earliest start date among the contributing potters. Drawn as the left extreme stub in both figures. |
lado:maxDatemin |
xsd:integer |
Latest start date among the contributing potters. |
lado:minDatemax |
xsd:integer |
Earliest end date among the contributing potters. |
lado:maxDatemax |
xsd:integer |
Latest end date among the contributing potters. Drawn as the right extreme stub in both figures. |
lado:intervalLabel |
xsd:string |
The interval rendered as text by the source query, retained for display and for comparison against the published web output. |
lado:DatingModel| Property | Range | Definition |
|---|---|---|
lado:kMin |
xsd:decimal |
Lower bound of the k factor, approached as evidence accumulates. |
lado:kMax |
xsd:decimal |
Upper bound of the k factor, applied when evidence is thin or absent. |
lado:tau |
xsd:decimal |
The saturation constant of the k curve. At n = τ roughly 63 per cent of the available narrowing has been achieved. |
lado:volumeWeight |
xsd:decimal |
Set to 1.0, recording that k depends on volume of evidence alone and on no other weighting. |
lado:calibrationBasis |
xsd:string |
How the coverage factor was arrived at, in words. The value of tau was not chosen for elegance: it is the smallest value at which every reference findspot’s terminus ante quem still falls inside its interval. k_min and k_max remain set conventions, because five reference ensembles cannot separate three parameters — attempting it produced 2 424 parameter combinations that fit the data equally well. |
lado:calibratedAgainst |
lado:Findspot |
A findspot whose date rests on dendrochronology, coins or a historical event rather than on samian ware. Naming them in the graph is the point: a calibration whose reference set is not published cannot be checked, and one calibrated against samian-dated ensembles would be circular. Two candidates were dropped for exactly that reason. |
lado:referenceLength |
xsd:decimal |
The common time scale on which qStart and qEnd are read, called t0 in the source query. Because it is a fixed length rather than a calendar value, the two edge measures no longer depend on the epoch of the material. Its value of 20 years is anchored on expert thresholds: a dispersion of 5 years counts as a sharp dating, one of 25 years as unusable, giving q of about 0.78 and 0.29 respectively. Until the calibration of tau it happened to carry the same value, which invited the two to be read as one constant; they are unrelated, and now they differ. |
lado:fuzzinessDivisor |
xsd:integer |
The value 12, being the variance of a uniform distribution over a unit range. This is where the per-stamp distributional assumption enters the model, and stating it makes that assumption inspectable rather than implicit. |
lado:excludedDatemax |
xsd:integer |
A datemax value removed by the source query’s filter. The reason for the exclusion is not currently documented. The filter is not neutral: it removes the affected potters at every findspot, so it is recorded here rather than left silent. |
lado:eraConvention |
xsd:string |
How negative years in the source are to be read. Under ‘historical’, −40 means 40 BC, and the derived xsd:gYear label is shifted by one because XSD counts astronomically. Only the calendar label is shifted; the numeric position is left untouched. |
lado:Figure| Property | Range | Definition |
|---|---|---|
lado:padYears |
xsd:integer |
Years of empty space added at each end of the time axis. |
lado:extremeStubYears |
xsd:integer |
Length of the stub drawn at each extreme value. The full range line was replaced by these stubs in the web application, and both figures follow that decision. |
lado:rowHeight |
xsd:integer |
Height of one row, in pixels. |
lado:svgWidth |
xsd:integer |
Width of the figure canvas, in pixels. |
lado:marginLeft |
xsd:integer |
Left margin, in pixels; it accommodates the site and findspot labels. |
lado:marginRight |
xsd:integer |
Right margin, in pixels. |
lado:marginTop |
xsd:integer |
Top margin, in pixels. |
lado:marginBottom |
xsd:integer |
Bottom margin, in pixels. |
lado:bandPadding |
xsd:decimal |
Fraction of each row left empty between bars, following the d3.scaleBand convention. |
lado:colourRamp |
xsd:string |
Name of the colour ramp mapping quality to colour, recorded so that a reader can reproduce the fill colours rather than eyeballing them. |
lado:rowOrder |
xsd:string |
The rule by which rows are sorted, allowing the order of the published figure to be reproduced from the graph alone. |
lado:IndependentTerminus| Property | Range | Definition |
|---|---|---|
lado:terminusKind |
xsd:string |
What kind of bound this is. A terminus ante quem says the material was in place by that date; it is emphatically not a production date. Pompeii AD 79 means the ware was there when Vesuvius erupted, not that it was made in 79. Blurring the two is the standard way of over-reading a fixed point, and the distinction is why this property exists rather than a bare year. |
lado:independentEvidence |
xsd:string |
Where the date comes from. It has to be independent of samian, or the calibration would be circular — the model would be tuned against dates the model itself produced. |
lado:withheldFromCriterion |
xsd:boolean |
True where a terminus appears in the figures but was kept out of the calibration criterion. A contested date should not set a published parameter; dropping it altogether would hide that it was ever a candidate and why. The accompanying reason says which is the case here. |
lado:withheldBecause |
xsd:string |
Why a terminus was withheld from the criterion, in plain words, so that the judgement can be disagreed with rather than merely obeyed. |
lado:PlotRow| Property | Range | Definition |
|---|---|---|
lado:uncStartYears |
xsd:integer |
Sample standard deviation of the contributing start dates. Presentation only: it measures the scatter of the interval edges, whereas the box is built from the scatter of the midpoints, so the whisker does not continue the box in any statistical sense. |
lado:uncEndYears |
xsd:integer |
Sample standard deviation of the contributing end dates. Presentation only, as for uncStartYears. |
lado:uncIntervalYears |
xsd:integer |
Combined edge dispersion, sqrt(var(datemin) + var(datemax)). Presentation only. |
lado:normQInterval |
xsd:decimal |
Presentation only. The value of lado:qInterval stretched onto the unit interval by the axis samian:axis_qInterval. Dating sharpness on its own 0-1 scale, so the domain is a fixed convention rather than the range this corpus happens to show. Green is sharp. Matches the legend of the box plot. |
lado:hexQInterval |
xsd:string |
Presentation only. The colour corresponding to lado:normQInterval on the axis samian:axis_qInterval. Recomputable from that axis’s stops and interpolation; carried in the graph so that two maps of the same data cannot disagree about the scale. |
lado:normMidpointYear |
xsd:decimal |
Presentation only. The value of lado:midpointYear stretched onto the unit interval by the axis samian:axis_midpointYear. Chronological position. A sequential ramp on purpose: red-green would read as a judgement, and late material is not worse material. |
lado:hexMidpointYear |
xsd:string |
Presentation only. The colour corresponding to lado:normMidpointYear on the axis samian:axis_midpointYear. Recomputable from that axis’s stops and interpolation; carried in the graph so that two maps of the same data cannot disagree about the scale. |
lado:normSigmaYears |
xsd:decimal |
Presentation only. The value of lado:sigmaYears stretched onto the unit interval by the axis samian:axis_sigmaYears. The dispersion the interval is built from. Reversed, so that small sigma is green and agrees with the qInterval axis. |
lado:hexSigmaYears |
xsd:string |
Presentation only. The colour corresponding to lado:normSigmaYears on the axis samian:axis_sigmaYears. Recomputable from that axis’s stops and interpolation; carried in the graph so that two maps of the same data cannot disagree about the scale. |
lado:normNStamps |
xsd:decimal |
Presentation only. The value of lado:nStamps stretched onto the unit interval by the axis samian:axis_nStamps. Sample size. Logarithmic, because the counts run from 2 to 170 and a linear domain would put four fifths of the findspots in the darkest eighth of the ramp. |
lado:hexNStamps |
xsd:string |
Presentation only. The colour corresponding to lado:normNStamps on the axis samian:axis_nStamps. Recomputable from that axis’s stops and interpolation; carried in the graph so that two maps of the same data cannot disagree about the scale. |