StorageGraphBfgTripleProvenance

Data providers shall use this to specify access requirement.

For KE internal use only. Data providers shall not set this. At ingress, LG Record service will read access_required and properly translate it to access_required_int. KE horizontal systems shall use this field instead of access_required; so that binaries would not depend on the release of a newly introduced AccessRequirement enum. (go/easy-ar-onboarding)

Freebase: the freebase user id in the form '/user/userid' KG: the Google LDAP of the developer or MDB group that set up the triplification and import pipeline for this data source.

Freebase & KG: the dataset the assertion was loaded from

Below are deprecated Provenance fields. They are not indexed or served in KE infrastructure (they are stripped at ingress in Livegraph).

Indicates that the corresponding data is supporting evidence for reconciliation only, and is not an assertion that should be visible to other systems or to external users. Note that this also means that no provenances indicating supporting data will be visible in the composed graph. Please see go/supporting-kg-triples-design-doc for additional details and background. If a triple is sent to Livegraph with multiple provenances each of them must have is_supporting_data bit set for it to be considered valid supporting evidence triple.

Internal metadata used by Livegraph and possibly other horizontal KG infra systems. This is not part of the logical triple or its provenance, and contents may not be visible downstream of LG.

Metadata specifying data governance policies. This information will be processed and enforced in KE systems. For more context, see go/ke-triple-dg-policy-and-metadata. WARNING: This field is WIP and please do not populate it without consulting ke-data-governance@.

Used to uniquely identify data sources. Freebase: the OAuth application KG: the name of the source directory the triple was loaded from (eg, "amg", "tms ", "collections"). KV: identifier of an extraction system, e.g., SAFT or Tractzor.

'rankingtoken' (which must be accompanied by the 'process' field above or will be ignored) is used to distinguish subsets of data within a single process, solely for the purposes of composition in Livegraph. This field is useful when partitioning of data is needed, but using a separate process is intractable due to organizational or infrastructure limitations. For example, if a process such as "geo" wishes to distinguish some of their /type/object/name triples as higher priority in Composer's conflict resolution rules, then they can add a config entry with a dedicated ranking_token and tag the relevant triples with that ranking_token. This field should only be used in conjunction with a Composer-side configuration to use it for value selection. It is not appropriate to use this field as free-form metadata. This value must match [[:alnum:]][[:alnum:]-]{0,127} (i.e., [a-zA-Z0-9]+[a-zA-Z0-9_-]* and must be less than 128 characters in length) , or else its containing triple will be considered malformed and will be rejected by Livegraph at ingress. As of 2018-02-14, this field is used within Livegraph's Composer and is exposed in Livegraph's Lookup APIs to aid debugging, but it is not indexed or served by TopicServer.

When specified as part of triples input to Livegraph (go/livegraph), this indicates that the triple needs to be triangulated by 3 different sources before it can be served in production. A triple's source is identified by its URL(provenance.source).host(). Please see go/baike-triangulation for more background on this. WARNING: If you're a new client trying to enable triangulation for your feed, please contact lg-composition@/kashk@ before using this feature.

Freebase & KG: if the triples were extracted from the web, the source URL where the assertion was found. (generally empty in freebase-sourced triples) TopicServer will serve/populate/retain if REQUIRES_CITATION is also set, or if the process is explicitly allowed. http://g/topic-server/vn9PBWtVKqI/arQEDqKTAgAJ