Conferences have no common identifier comparable to the ISSN of journals. Acronyms are ambiguous (there are many ICCCs) and change over time; publishers, libraries and registries use their own keys. confref assigns each series and each edition one identifier and links it to the identifiers of the other sources.
Structure
s10ajew is a series (ISWC), e10k0hm an edition (an event: ISWC 2023).
- The first letter says what it identifies:
sfor a series,efor an event. - The number is written in base 32 with the letters and digits that cannot be mistaken for each other (no
i,l,ooru). - The last character is a check character: a mistyped character or two swapped neighbours are recognised.
As IRIs: https://id.confref.org/series/s10ajew and
https://id.confref.org/event/e10k0hm. A browser receives
an HTML page (the landing page), a linked data client RDF in the format it asks for (Turtle, N-Triples, JSON-LD or
RDF/XML), each by a 303 redirect to the document in that format. The platform at id.confref.org is in beta. The
identifiers with their records and facts are in the dataset ids: 11,089 series and 78,280 events.
Rules
- No reuse: if two identifiers turn out to denote the same conference, the older one remains and the other refers to it. If one identifier turns out to cover two conferences, both receive new identifiers and the old one refers to them.
- Aliases: acronyms such as
eswc2016, the IRIs of lod.springer.com and lod.confref.org (2018) and the pages of portal.confref.org lead to the confref identifier. - Links to other identifiers: the DataCite DOIs of ConfIDent, dblp, Wikidata and the GND; DOIs and ISBNs of the proceedings; ROR for organisers.
IRI scheme
All IRIs that confref coins are below https://id.confref.org/. The first path segment says what an IRI stands for;
these segments are reserved:
| Path | What |
|---|---|
series/<id> | a conference, workshop or symposium series; resolves to its landing page or its data |
event/<id> | one event: a conference, workshop, symposium, joint or federated event |
event/<id>/edition/<series id> | the edition number of an event in one of its series, when it belongs to several |
datasets/<slug>, datasets/<slug>/<version> | a dataset (one named graph) and its versions |
dumps/<slug>/<version>/<file> | a file of a version |
data/<namespace>/… | a record that confref coins for a source without IRIs of its own |
ontology/ | the confref ontology (prefix cr:) |
Records of the sources keep the IRIs of the sources (dblp, Wikidata, GND, DOIs, lod.springer.com). A dataset has no
path of its own at the top level: its IRI, its files and its records are found under datasets/, dumps/ and
data/.
Series and events
- A series is a sequence of recurring events under one name. A workshop series is a series of its own.
- An event belongs to one or more series: a joint conference that is branded and run as one event belongs to the series of each partner.
- An event can be held at a larger event: a workshop at its main conference, the partner conferences of a joint conference that presents as a collection of conferences, a conference at a federated event. The series of an event is independent of the series of its host.
- Separate events can be joint or co-located: held together with joint organisation or programme, or at the same place and time with programmes of their own.
- The way of the source is kept. The same joint conference can appear in one source as one event in two series and in another as two joint events; the links show how they are connected.
- Proceedings are not events; they are linked to the events whose contributions they publish.