1.1.3.5. Routes, Itineraries, Networks, Groups and Periods¶
This page describes how WHG models things that are made of other places: a route (a way through places in order), an itinerary (a journey along a way, dated stop by stop) and a network (places and the connections between them, in no single order). It also says where two things that earlier drafts modelled in the same way now sit instead: a gazetteer group, which is a group of Gazetteers, and a period, which is an Authority. Neither of those is a SpatialEntity.
The model is PLATO’s, as released in PLATO 0.7.1 (doi:10.5281/zenodo.23060220; all versions: doi:10.5281/zenodo.21688313). A route, an itinerary or a network is a SpatialEntity, typed as such, and everything about it (its members, their order, its connections and the figures a source gives for them) is an attestation with its own source, date and certainty.
The excerpts on this page come from PLATO’s worked examples of routes, journeys and networks: the Antonine Itinerary (a route), King John in 1215 (an itinerary), the lower River Idle (a physical network) and the Datini letters (a relational network).
1.1.3.5.1. The building blocks¶
1.1.3.5.1.1. Typing: routes, itineraries, networks and segments¶
What kind of entity a SpatialEntity is, is a Type attestation (attests_type). A Type node stands
for one dataset’s use of a concept: its @id, if it has one, is the dataset’s own address, never
the concept’s IRI, which goes in its identifier; a Type may also say its vocabulary (scheme) and
that vocabulary’s version (schemeVersion). PLATO declares four kinds of entity that a platform’s
behaviour depends on, as concepts in the scheme plato:EntityKindScheme, so that any consumer can
tell these entities from places:
Kind |
What it is (PLATO’s definition) |
|---|---|
|
An ordered set of places making a way: a road, a sea lane, a pilgrimage route. Its timespan, if any, is when it was in use, not when anyone travelled it. |
|
A journey made through places in order, each stop with its own timespan: a traveller’s tour, a royal progress, a ship’s voyage. |
|
A set of places and the connections between them, in no single order: a river system, a canal network, a correspondence network. |
|
A physical link with an existence of its own: a leg of a route between two stations, a reach of a river between confluences. |
A Type attestation names one of these concepts by its IRI in the Type’s identifier. A route may
also carry other Type attestations (an AAT term for a Roman road, say); the PLATO kind is what tells
WHG to treat it as a route.
Note
plato:TypeItinerary is not plato:Itinerary. The second is a geometry role: a line tracing a
route or journey, given on a Geometry. An itinerary SpatialEntity may have a geometry with that role,
but the role does not make anything an itinerary.
1.1.3.5.1.2. Membership and sequence: MemberOf¶
Each member of a route, itinerary or network (a station, a stop, a port, a segment) has an
attestation about the member that relates_to the whole with the relation type
plato:MemberOf (label member_of, inverse has_member).
SpatialEntity(member) ←[attests_about]─ Attestation(sequence: n) ─[has_relation_type]→ RelationType(MemberOf)
└─[relates_to]→ SpatialEntity(route | itinerary | network)
Where the source gives an order, the member’s position is plato:sequence on that same
attestation:
The number is scoped by the
relates_totarget. It orders the member within that one route, itinerary or network, so one place can hold different positions on different routes.Lower numbers come first. Gaps do not matter.
Equal numbers are alternatives at that point: two branches of a road.
No number means the source gives no order. It does not mean “last”, and it is not an error.
The order is one source’s. Two sources that list a road’s stations differently are recorded as their own attestations, each with its own sequence; where an editor judges between them, that judgement is a meta-attestation.
MemberOf is not containment: a town on a road is not within it, and a network’s members need not
lie inside anything. Containment is plato:ContainedIn.
1.1.3.5.1.3. Connections and direction: ConnectedTo and LeadsTo¶
A connection between two members is an attestation about one that relates_to the other:
Relation type |
Label / inverse |
Direction |
|---|---|---|
|
|
None. “A is connected to B” says the same as “B is connected to A”. |
|
|
From the subject ( |
Direction is carried by the relation type and nowhere else. There is no qualifier that turns an undirected connection into a directed one, because a consumer that ignored the qualifier would silently misread a one-way connection as two-way.
A connection known in both directions from sources that each give one direction is better recorded
as two LeadsTo attestations, each with its own source, than as one ConnectedTo.
1.1.3.5.1.3.1. Project relation types: plato:broader_relation¶
A project may declare its own, narrower relation type (“flows into”, “carries post to”) as a
RelationType Authority with its own relation_label and inverse_label, and point it at the
starter type it narrows with plato:broader_relation. A narrower type keeps the direction of its
broader one, so a consumer that knows only LeadsTo still follows “flows into” the right way.
A project type is declared once, in the PLATO JSON document’s relationTypes array (beside
dataSets), and any relation then uses its address as its relationType:
"relationTypes": [
{
"@id": "https://example.org/vocab/flows-into",
"label": "flows into",
"inverseLabel": "receives",
"broaderRelation": "https://w3id.org/plato#LeadsTo"
}
]
(From PLATO’s JSON guide.) PLATO’s
spreadsheet tables take only PLATO’s own relation types, so a project type cannot be declared there;
from spreadsheets, a project uses LeadsTo itself, as the River Idle example does. See
plato:broader_relation and
plato:declares_relation_type.
1.1.3.5.1.4. Segments: BeginsAt, EndsAt and HasEnd¶
A physical link with its own existence (a leg of a road, a reach of a river) is a segment: a
SpatialEntity with a Type attestation naming plato:TypeSegment. It is a member of its route or network
(MemberOf, with a sequence where the source orders it), and it is joined to its end places with
relation types of its own:
Relation type |
Label / inverse |
Use |
|---|---|---|
|
|
A directed segment begins at the target: a leg at the station it leaves, a reach at its upstream confluence. |
|
|
A directed segment ends at the target: a leg at the station it reaches, a reach at its downstream confluence. |
|
|
An undirected segment has the target as one of its ends. It has two |
Segment SpatialEntity (TypeSegment)
←[attests_about]─ Attestation ─[has_relation_type]→ RelationType(MemberOf)
└─[relates_to]→ Route or Network
←[attests_about]─ Attestation ─[has_relation_type]→ RelationType(BeginsAt)
└─[relates_to]→ Place A
←[attests_about]─ Attestation ─[has_relation_type]→ RelationType(EndsAt)
└─[relates_to]→ Place B
←[attests_about]─ Attestation ─[attests_property]→ PropertyValue(length, …)
Segments never use ConnectedTo or LeadsTo for their ends. Those always join one place to
another, so “the places connected to Londinium” never returns a segment, and a walk across a network
does not take two steps for every one.
A segment’s geometry is optional: a leg known only by its ends needs none. Because a segment is not a place in the everyday sense (“Road from Londinium to Durobrivae” would be a pseudo-place), WHG may leave segments out of lists and searches of places.
1.1.3.5.1.5. Figures on links: PropertyValues¶
Figures about a link (a distance, a journey time, a toll, a cargo, a count of letters) are
PropertyValues (attests_property), each on an attestation with its own source and date. Where
they go depends on what kind of link it is:
Physical link (a segment): the figures are PropertyValues on attestations about the segment.
Relational link (two cities that exchanged letters): the link is the
ConnectedToorLeadsToattestation, and its figures are PropertyValues on that attestation itself.
Note
Earlier drafts of this model had a connection_metadata field, an opaque string for such figures.
It does not exist: PLATO removed it in 0.6.0, and WHG never implemented it.
1.1.3.5.1.6. Computed values: plato:computed¶
Some values can be worked out again from other statements in the same data: an itinerary’s span
from the timespans of its stops, a route’s line from its stations, a network’s hull. WHG marks such
a value computed (plato:computed, JSON computed) when it shows or exports it. A computed value
is not evidence: a consumer may show it, but must not import it as an attestation of its own. It
should compute the value again from the members, or leave it out and say so. See
Computed extents below.
computed is not for every figure that someone worked out. PLATO’s definition says:
It is for a value derived from other statements in the same data, which any consumer holding them could derive again. A figure a project works out from its own source material and publishes as a finding (letters counted from a register, a median taken over them) is not computed in this sense: it is attested, citing the work that produced it as a Source or Dataset derived_from the material it was worked out from.
The letter counts in the Datini example (see Relational networks) are findings of that kind.
1.1.3.5.2. Route¶
A route is a SpatialEntity typed plato:TypeRoute: an ordered set of places making a way. Its
timespan, if it has one, is when it was in use, not when anyone travelled it.
Graph structure:
Route SpatialEntity
←[attests_about]─ Attestation ─[attests_name]→ Name(…)
←[attests_about]─ Attestation ─[attests_type]→ Type(identifier: plato:TypeRoute)
←[attests_about]─ Attestation ─[attests_timespan]→ Timespan(when in use) (optional)
Stations, in the source's order:
SpatialEntity(station 1) ←[attests_about]─ Attestation(sequence: 1) ─[has_relation_type]→ RelationType(MemberOf)
└─[relates_to]→ Route
SpatialEntity(station 2) ←[attests_about]─ Attestation(sequence: 2) ─[has_relation_type]→ RelationType(MemberOf)
└─[relates_to]→ Route
Legs, where modelled as segments:
Segment(1→2) ←[attests_about]─ Attestation ─[has_relation_type]→ RelationType(MemberOf)
└─[relates_to]→ Route
←[attests_about]─ Attestation ─[has_relation_type]→ RelationType(BeginsAt)
└─[relates_to]→ station 1
←[attests_about]─ Attestation ─[has_relation_type]→ RelationType(EndsAt)
└─[relates_to]→ station 2
←[attests_about]─ Attestation ─[attests_property]→ PropertyValue(distance)
graph LR
S1[station 1] -->|MemberOf, sequence 1| R((Route))
S2[station 2] -->|MemberOf, sequence 2| R
S3[station 3] -->|MemberOf, sequence 3| R
L1[segment 1→2<br/>TypeSegment] -->|MemberOf| R
L1 -->|BeginsAt| S1
L1 -->|EndsAt| S2
Legs need not be modelled at all: a route whose source lists only stations is its stations in order. Model a leg as a segment when the source says something about it (a distance, a stage of the road) or when it has a geometry of its own.
In PLATO’s Antonine Itinerary example, the roads between stations are segments and members of the route like the stations, and the two share one sequence: Londinium 1, the road to Durobrivae 2, Durobrivae 3, and so on. The first road, in PLATO JSON, with its membership of Iter III and its length as Iter III gives it (its Type, its membership of Iter IV, its ends and its length as Iter IV gives it are cut):
{
"@id": "https://whgazetteer.org/example/antonine/place/leg-londinium-durobrivae",
"label": "Road from Londinium to Durobrivae",
"entityIdentifier": "leg-londinium-durobrivae",
"attestations": [
// …
{
// …
"relations": [
{
"relatesTo": "https://whgazetteer.org/example/antonine/place/iter-iii",
"relationType": "https://w3id.org/plato#MemberOf"
}
],
"sequence": 2
},
// …
{
// …
"notes": "Written \"mpm XXVII\". As Iter III gives it.",
"properties": [
{
"property": "https://www.wikidata.org/wiki/Property:P2043",
"label": "length (Roman miles)",
"value": 27,
"unit": "http://www.wikidata.org/entity/Q2118176"
}
]
},
// …
],
// …
}
From PLATO’s worked example the Antonine Itinerary.
1.1.3.5.2.1. Routes within routes¶
A route may be a member of another route (one stage of a pilgrimage, one iter of a larger
itinerary), with MemberOf and a sequence like any other member. A route may not be a member of
itself through any chain of memberships, and plato-tools reports a cycle as an error. See
plato:MemberOf.
1.1.3.5.2.2. Branches and alternatives¶
Two members with the same sequence number are alternatives at that point: two branches of a
road, two stations a source offers for one stage. Neither is preferred by the data; if an editor
prefers one, that is a meta-attestation. See plato:sequence.
1.1.3.5.2.3. Unknown order¶
A member with no sequence is in an order the source does not give. It is still a member; it simply has no position.
1.1.3.5.2.4. Two sources, two orders¶
Each source’s order stays on that source’s attestations. A second source that orders the same
stations differently is a second set of MemberOf attestations, not a correction of the first.
1.1.3.5.3. Itinerary¶
An itinerary is a SpatialEntity typed plato:TypeItinerary: a journey made through places in
order, each stop with its own timespan. The distinction from a route is WHG’s: a route is a way,
and an itinerary is a journey made along one, dated stop by stop.
Each stop is a member (
MemberOf), with asequencewhere the source orders the stops.A stop’s timespan is its stay, from arrival to departure, given on the member’s
MemberOfattestation (attests_timespan).The order of a journey known only from its dates may be left to those timespans, with no
sequence.A leg between stops that the source describes (a distance, a day’s ride) is a segment, as for a route.
Where the source states how long a stay lasted, the timespan carries
plato:duration, anxsd:duration(P4Dfor four days; weeks are written as days,P42D), beside its dates or, if the source gives none, alone.The itinerary’s own span is computed from its stops’ timespans unless a source gives it (see Computed extents).
Graph structure:
Itinerary SpatialEntity
←[attests_about]─ Attestation ─[attests_name]→ Name(…)
←[attests_about]─ Attestation ─[attests_type]→ Type(identifier: plato:TypeItinerary)
Stops, each with its stay:
SpatialEntity(stop 1) ←[attests_about]─ Attestation(sequence: 1)
├─[has_relation_type]→ RelationType(MemberOf)
├─[relates_to]→ Itinerary
└─[attests_timespan]→ Timespan(arrival – departure)
SpatialEntity(stop 2) ←[attests_about]─ Attestation(sequence: 2)
├─[has_relation_type]→ RelationType(MemberOf)
├─[relates_to]→ Itinerary
└─[attests_timespan]→ Timespan(arrival – departure)
Span shown by WHG (not stored as evidence):
Itinerary ── Timespan(earliest arrival – latest departure, computed: true)
In PLATO’s King John example, Marlborough is a member of the journey twice, as stop 12 and stop 17,
each with its own stay on its own MemberOf attestation; the first gives a length as well as dates.
Its two MemberOf attestations, in PLATO JSON:
[
{
"timespans": [
{
"sourceLabel": "the first four days of July",
"startEarliest": "1215-07-01",
"endLatest": "1215-07-04",
"duration": "P4D"
}
],
// …
"relations": [
{
"relatesTo": "https://whgazetteer.org/example/king-john/place/itinerary-1215",
"relationType": "https://w3id.org/plato#MemberOf"
}
],
"sequence": 12
},
{
"timespans": [
{
"sourceLabel": "the following day",
"startEarliest": "1215-07-08",
"endLatest": "1215-07-08"
}
],
// …
"sequence": 17
}
]
From PLATO’s worked example King John in 1215.
Note
A journey is not a Linked Traces relation between a place and a traveller. PLATO is explicit:
a journey’s waypoints, stops, origin and destination are members of an itinerary, with MemberOf,
ordered by sequence, their stays given by timespans. The relation types for a place’s part in a
person’s life (BirthplaceOf, ResidenceOf and the rest) are for other things; see
Associations.
1.1.3.5.4. Network¶
A network is a SpatialEntity typed plato:TypeNetwork: a set of places and the connections
between them, in no single order. Its members are MemberOf the network, usually with no
sequence. What joins them depends on whether the links are physical or relational.
Physical network |
Relational network |
|
|---|---|---|
A link is |
a segment (a SpatialEntity, |
a |
Its ends |
|
the attestation’s subject and target |
Its figures |
PropertyValues on attestations about the segment |
PropertyValues on the connection attestation itself |
Its geometry |
optional, on the segment |
none of its own |
Examples (PLATO) |
a river system, a canal network |
a correspondence network |
1.1.3.5.4.1. Physical networks: segments¶
A river system or a canal network is made of links that exist in their own right. Each reach or
cut is a segment: MemberOf the network, with BeginsAt and EndsAt its end places (or HasEnd
twice if it has no direction), and its length and other figures as PropertyValues.
Rivers that flow into one another are places, and are joined directly with LeadsTo, or with a
project type such as “flows into” whose broader_relation is LeadsTo. The reaches never are.
PLATO’s comment on LeadsTo gives “one river flowing into another” as its first example, and says:
It joins two places; the reaches and legs of a network are segments, whose direction is given by plato:BeginsAt and plato:EndsAt.
Where the source orders the reaches (downstream, say), their MemberOf attestations carry a
sequence, as for a route.
Network SpatialEntity (TypeNetwork)
Segment(reach) ─MemberOf→ Network (sequence, where the source orders the reaches)
Segment(reach) ─BeginsAt→ junction (upstream)
Segment(reach) ─EndsAt→ junction (downstream)
river (tributary) ─LeadsTo / "flows into"→ river (optional)
graph LR
T[tributary river] -->|LeadsTo<br/>or a narrower type| N((river network))
R1[reach A–B<br/>TypeSegment] -->|BeginsAt| C1[junction A]
R1 -->|EndsAt| C2[junction B]
R1 -->|MemberOf| N
From PLATO’s River Idle example, in Turtle: a tributary joined to the river with LeadsTo, and one
reach’s membership (numbered downstream) and its upstream end. Its downstream end (EndsAt) and
its length, a PropertyValue as for a road’s, follow in the full example.
# Rivers are places, and LeadsTo joins places: the Ryton flows into the Idle,
# the Idle into the Trent.
whgx:river-idle\/attestation\/ryton-into-idle
a plato:Attestation ;
plato:attests_about whgx:river-idle\/place\/river-ryton ;
plato:relates_to whgx:river-idle\/place\/river-idle ;
plato:has_relation_type plato:LeadsTo ;
plato:has_citation [ a plato:Citation ;
plato:cites whgx:river-idle\/source\/rewt-v020 ;
plato:locator "link 53DEF62E, to_node 69D120BE" ] .
# …
whgx:river-idle\/attestation\/reach-04-member
a plato:Attestation ;
plato:attests_about whgx:river-idle\/place\/reach-04 ;
plato:relates_to whgx:river-idle\/place\/river-idle ;
plato:has_relation_type plato:MemberOf ;
plato:sequence 4 ;
plato:has_citation whgx:river-idle\/citation\/reach-04 .
whgx:river-idle\/attestation\/reach-04-begins
a plato:Attestation ;
plato:attests_about whgx:river-idle\/place\/reach-04 ;
plato:relates_to whgx:river-idle\/place\/node-03 ;
plato:has_relation_type plato:BeginsAt ;
plato:has_citation whgx:river-idle\/citation\/reach-04 .
# …
From PLATO’s worked example the lower River Idle.
1.1.3.5.4.2. Relational networks: connections with figures¶
A correspondence network, or any network whose links are relationships rather than things, has no
segments. Each link is a ConnectedTo (either way) or LeadsTo (one way) attestation between two
places, and each figure a source gives about that link (letters sent, a delivery time) is a
PropertyValue on that same attestation, with the attestation’s own source and date.
SpatialEntity(city A) ←[attests_about]─ Attestation
├─[has_relation_type]→ RelationType(LeadsTo)
├─[relates_to]→ SpatialEntity(city B)
├─[attests_property]→ PropertyValue(<figure>)
├─[attests_timespan]→ Timespan(<when>)
└─[sourced_by]→ Source
Several figures about one link from one source may sit on one attestation; figures from different sources, or for different dates, are different attestations. Connections appear and disappear over time in the same way: as attestations with different timespans.
In PLATO’s Datini example, the letters from Florence to Pisa are a LeadsTo attestation about
Florence, carrying the count as a PropertyValue; the median delivery time is a second attestation
of the same kind, and Pisa to Florence is two more. The counts were worked out from a dataset of
letters, so they cite that work as a Source of its own, derivedFrom the dataset: they are
attested findings, not computed values (see Computed values).
{
"timespans": [
{
"sourceLabel": "1355-1402",
"startEarliest": "1355",
"endLatest": "1402"
}
],
"citations": [
{
"source": {
"@id": "https://whgazetteer.org/example/datini/source/letters-by-pair",
"title": "Letters and delivery times by pair of cities",
// …
"derivedFrom": "https://whgazetteer.org/example/datini/source/datini-metadata",
// …
},
"locator": "FIRENZE to PISA",
"citationFunction": "http://purl.org/spar/cito/citesAsDataSource"
}
],
// …
"relations": [
{
"relatesTo": "https://whgazetteer.org/example/datini/place/pisa",
"relationType": "https://w3id.org/plato#LeadsTo"
}
],
"properties": [
{
"property": "https://www.wikidata.org/wiki/Property:P1114",
"label": "letters sent",
"value": 16261
}
]
}
From PLATO’s worked example the Datini letters.
1.1.3.5.5. Gazetteer groups¶
A gazetteer group is a thematic or organisational grouping of Gazetteers, not of places:
PLATO’s example is “EMEW — Gazetteer of Early Modern England and Wales”. It is the class
plato:GazetteerGroup, and a Gazetteer belongs to one with plato:member_of_group.
Gazetteer ─[member_of_group]→ GazetteerGroup
Gazetteer ─[member_of_group]→ GazetteerGroup
It is not a SpatialEntity. It has no name, type, geometry or timespan attestations, and it does not appear among places.
Membership is not
MemberOf.plato:member_of_grouplinks a Gazetteer to its group directly; it is not an attestation and carries no sequence.Anything a group appears to “cover” in space or time is worked out from its Gazetteers’ contents and, if shown, is computed.
1.1.3.5.6. Periods¶
A period is an Authority (plato:Period), alongside Sources, Datasets, RelationTypes and
CertaintyLevels: a named historical period (“Byzantine”, “Early Modern”), optionally linked to
PeriodO or defined locally.
Property |
Use |
|---|---|
|
The period’s name. |
|
Its temporal extent, as a Timespan, for a locally defined period. |
|
The PeriodO definition it aligns with. |
|
The region to which it applies, as free text or a URI. |
It is not a SpatialEntity, and it has no members. Places are not
MemberOfa period.An attestation’s date may be given relative to a period: a Timespan with
plato:relative_topointing at the Period (“during the reign of Justinian”).A Timespan may also carry a
plato:periodo_uriof its own, where it corresponds to a PeriodO period.PeriodO periods come into WHG as Period authorities, not as SpatialEntities (this follows from
plato:Periodbeing an Authority; the import itself is not yet specified).
1.1.3.5.7. Computed extents¶
WHG works out some values from the members of a route, itinerary or network. They are useful to show and to export, and harmful to import: read back as attestations, they would become evidence that no source gave, and they would outlive any correction to the members they were computed from.
Timespan of an itinerary. Unless a source gives one, the span is computed from the members’
MemberOf timespans:
start_earliest= the minimum of the members’start_earlieststart_latest= the minimum of the members’start_latestend_earliest= the maximum of the members’end_earliestend_latest= the maximum of the members’end_latest
Geometry of a route or network. Unless a Geometry is attested for the entity itself, WHG may derive one: a route’s line through its stations in sequence order (skipping unordered members), or a network’s hull of its members’ geometries.
Attested beats computed. A timespan or geometry that a source gives for the route, itinerary or network is an ordinary attestation and is shown as such. A computed value is shown only where there is no attested one, and is always labelled as computed.
Worked out is not the same as computed. Only a value that can be derived again from the other
statements in the same data is computed. A figure a project has worked out from its own source
material (the Datini letter counts) is attested, citing that work as a Source derived_from the
material, and is imported like any other attestation.
On export, a computed value carries computed: true (on the attestation, or on the facet in its
qualification). On import, WHG does not take in anything marked computed as an attestation;
it computes the value again from the members.
For King John’s itinerary of 1215, WHG would compute the span from the stops’ own dates (the first
arrival, 1 June, to the last departure, 17 July) and export it as one more attestation on the
itinerary, marked computed:
{
"timespans": [
{
"startEarliest": "1215-06-01",
"endLatest": "1215-07-17"
}
],
"citations": [
{
"source": {
"@id": "https://whgazetteer.org/",
"title": "World Historical Gazetteer",
"authorityType": "dataset"
}
}
],
"computed": true,
"notes": "Computed by WHG from the stops' dates: earliest arrival to latest departure."
}
computed says how the value arose; the citation says who says so (WHG itself, as a dataset). A
consumer that sees both knows to recompute the span from the stops rather than import it, since it
is not evidence. (This attestation is not part of
PLATO’s King John example;
added to a copy of it, the document still passes plato-tools.)
Several geometries: PLATO does not accept a GeoJSON GeometryCollection; each geometry is its
own Geometry node. When one source gives several geometries together (a point and a polygon for the
same thing, say), they may sit in one attestation, each with a role where they depict different
things. A geometry from another source, or for another date, goes in its own attestation, because
provenance belongs to the attestation.
1.1.3.5.8. On place pages¶
A station, a stop or a port is an ordinary place, and its page shows the routes, itineraries and networks it belongs to:
Memberships: each
MemberOfattestation, with the whole it relates to and, where given, its position (sequence) and its neighbours in that source’s order.Connections: its
ConnectedToandLeadsToattestations, showing direction (“leads to” / “reached from”) and any figures.Segments are not listed as places. A segment’s own page, if it has one, shows its ends and figures.
1.1.3.5.8.1. Associations: places in the histories of people, objects and events¶
Linked Traces annotated places with their part in the lives of people, the history of objects and
the course of events. In PLATO each such statement is an attestation about the place that
relates_to the person, object or event by its IRI, with a plato:related_label to name it, and one
of these relation types:
Relation type |
Label / inverse |
Target |
|---|---|---|
|
|
person |
|
|
person |
|
|
person or group |
|
|
object |
|
|
event |
|
|
person or group |
|
|
image, map or drawing |
|
|
record, document or publication |
The last two, new in PLATO 0.7.0, link a place to images and records held elsewhere: a photograph
or plan that shows it (DepictedIn), or an archival file, site record or publication about it
(SubjectOf). Each is a statement of its own, usually sourced from the catalogue that identified
it, not a citation of evidence for another statement (which is has_citation).
The target is usually not a SpatialEntity, and PLATO deliberately gives relates_to no range so
that it is not inferred to be one. PLATO’s comment on relates_to says:
Platforms should show relations to entities that are not spatial apart from spatial ones, and leave them out of containment, clustering and matching.
So WHG shows associations in their own section of a place page, labelled with related_label, and
does not use them in containment, clustering or matching. Thematic keywords (“coastal trade”,
“politics”) are not relations: they belong in notes or, where they classify, in a Type.