GCRI · Registry
Registry の相互運用性
Registries are only as useful as the links between them. A record that cannot be referenced stably, exported faithfully or read by another system becomes an island, and islands drift apart. This page sets out how stable references, persistent identifiers, metadata crosswalks, export formats and interface semantics let registries exchange records without losing meaning, and how interoperability fails through link rot, lossy mapping and undocumented change.
作業の進め方。
Interoperability console
Registry interoperability · Crosswalk · Profile v3
Field mappings
| identifier | id | copy | 合格 |
| evidence_label | quality.label | enum map | 合格 |
| licence | rights.licence | SPDX id | Ambiguous |
| version | version | semver | 合格 |
適合性
図Interoperability console: example names and illustrative values, not a record or a live system.
Registry interoperability: how the risk develops.
Stable references, exports and API semantics. The dossier below sets out how the risk develops, what is measured, where the data come from, how it is modelled and who decides.
Link rot and content drift
Location links break when sites are reorganised, and links that still resolve may point to changed content. Without persistent identifiers and archived snapshots, references silently stop meaning what their authors intended.
Lossy crosswalks
Mapping one schema to another drops fields with no equivalent and flattens distinctions, such as several contributor roles folded into one. Round trips lose information, and each exchange makes the copies less alike.
Undocumented interface change
An interface changes a field’s meaning, format or default without a version or notice. Consumers keep parsing successfully while reading the wrong thing, and the errors surface far from the change that caused them.
Identity collision across registries
The same organisation, person or work carries different identifiers in different registries, or a local identifier is reused for different things. Joins across システム then merge what is distinct or split what is one.
How registry interoperability is modelled, measured and tested.
モデル、その仮定、データ、不確実性を、科学者やエンジニアが確認する形で示します。構造は公表された手法に従っています。描かれた値は例示であり、知見ではありません。
データセットと来歴
The data classes behind registry interoperability and what is recorded about each before it is used.
| データ区分 | 来歴を記録済み | 解像度を明示 | ライセンス確認済み |
|---|---|---|---|
| 衛星画像、光学とレーダー | センサー、処理レベル、取得日 | 画素サイズと再訪周期 | シーンごと |
| 再解析と気候モデルの出力 | モデル、実験、メンバー、バージョン | グリッドと時間ステップ | リリースごと |
| 現地センサーと計測器 | 観測所、機器、校正記録 | 地点とサンプリング・レート | ネットワークごと |
| 人口と資産のエクスポージャー | 手法、作成年、入力ソース | セルの大きさと単位 | リリースごと |
| 境界と参照地理 | 作成者、版、日付 | 縮尺 | 版ごと |
- 系譜
- 情報源から図までの W3C PROV 記録
- 原則
- FAIR：見つけやすく、アクセス可能で、相互運用でき、再利用できる
- ギャップ
- 欠損として表示し、黙って埋めることはしない
Interoperability lets records from each domain move between systems.
Interoperability methods are designed for the 手法と技術シリーズ. Crosswalks to external standards are the subject of NS-14 in the proposed Nexus Standards, with no conformity claimed.
- Crosswalk proposals.
- Interoperability test results.
何を、どの程度うまく測定しているか。
各指標には、その単位、更新頻度、背後にある出典の種類、値が持つことになるエビデンスのラベルが付いています。値は表示しません：プログラムが自らの権限のもとで値を提供します。
| 指標 | 単位 | 周期 | 出典の種類 | ラベル |
|---|---|---|---|---|
| Identifier resolution success | share of resolution probes | 日次 | Resolver probes | 観測 |
| Interface availability | share of time available | 継続的 | Availability monitoring | 観測 |
| Export schema conformance | share of exported records validating | リリースごと | Schema validation | 観測 |
| Crosswalk coverage | share of fields mapped to the target schema | リリースごと | Crosswalk register | 申告済み |
| Round-trip fidelity | share of records returned intact | リリースごと | Round-trip test | 観測 |
| Broken outbound references | share of references | 月次 | Link checking | 観測 |
| Deprecation notice period | 日 | per interface change | Change log | 申告済み |
| Records linked to external persistent identifiers | share of records | 月次 | Identifier reconciliation | 観測 |
エビデンスの出所。
名前の明示された公開データセットと機関の記録。それぞれにライセンス区分があります。合意に基づく出典は、その条件のもとでのみ使用し、再公開することはありません。
| 出典 | 提供するもの | ライセンス区分 |
|---|---|---|
| DCAT version 3 | A vocabulary for catalogues, datasets, data services and versions in exchange | W3C document licence, free to implement |
| Schema.org | A shared vocabulary for embedding structured metadata in web pages | CC BY-SA for the vocabulary |
| OAI-PMH | A protocol for harvesting metadata from repositories | Open specification |
| Scholix link exchange | A framework and services for exchanging links between literature and data | CC0 for link metadata |
| re3data registry of research data repositories | Descriptions of repositories with their standards, identifiers and access policies | CC0 for metadata |
| FAIRsharing | A curated registry of standards, databases and policies and their relationships | CC BY-SA for content |
| OpenAIRE Graph | An open graph linking publications, data, software, projects and organisations across sources | CC BY |
| OpenAPI 仕様 | A standard description format for HTTP interfaces | Open specification, Apache licence |
誰が決め、エビデンスが何に役立つか。
Registry prepares evidence. The decisions below are taken by the authorities named, under their own rules, and never by the platform.
| 誰が決定するか | 質問 | このページが提供するエビデンス |
|---|---|---|
| Registry and repository managers | Which schemas, identifiers and interfaces to support, and on what deprecation timetable | Crosswalk coverage, round-trip losses and the consumers depending on each interface |
| Research infrastructure funders | Whether to fund integration with an existing registry or a new parallel system | Interoperability measures, standard conformance and existing link coverage |
| データ・スチュワード | Whether imported records can be merged with local ones | Identifier matches, crosswalk losses and conflicts between sources |
| National statistics offices and public data portals | Whether registry exports can feed published catalogues | DCAT conformance, licence metadata and the stability of identifiers |
このトピックが生み出す成果物
- Stable references for every record, resolving to a current view and its versions
- Exports in documented formats with schema files and published crosswalks
- An interface change log with version numbers and deprecation notices
- Round-trip tests against external registry schemas with every loss listed
What experts ask about interoperability.
- How should identifiers from different registries be linked without copying records?
- Which open formats preserve meaning, provenance and rights when records are exported?
- How are field meanings documented so that two システム agree on what a value means?
- How do interfaces signal that a record has changed or been withdrawn?
- How are schema changes made without breaking systems that depend on them?
Why exchange fails.
Interoperability fails at three levels. Technically, formats and protocols do not match. Semantically, the same field name means different things, so a status of active or a date of publication is silently misread. Organisationally, no one is responsible for keeping links current, so references decay. Semantic failures are the hardest to see because the data moves without error and is simply wrong at the destination.
Shared vocabularies address these. Linked-data formats such as JSON-LD carry meaning with the data, catalogue vocabularies such as DCAT describe datasets, and W3C PROV carries lineage. Methods from データ・エンジニアリング、相互運用性、スチュワードシップ cover these standards and their trade-offs. The Registry’s planned interfaces are described on interfaces.
Designing for exchange.
- 01
Link, do not copy
Records refer to external registries by identifier, preserving aliases so an external reference still resolves.
- 02
Documented semantics
Each field’s meaning, allowed values and version are published alongside any interface.
- 03
Versioned schemas
Schema changes take a new version, and older records keep the schema they were made under.
- 04
限界
Interoperability depends on partners maintaining their own records. Mappings between vocabularies lose nuance, and an export carries rights obligations that the receiving system must honour.
Where this page’s references lead.
このページの読解の基準となる機関と文書。状態と版は変わります。各リンクは発行機関へ移動します。
質問と回答。
What makes a reference stable?
Why not just use web addresses?
What is a crosswalk?
How are interface changes announced?
Is there an interface I can use now?
Will exports include restricted records?
Can my registry link to registry records?
このページからエビデンスを提供する。
報告書 is the Consortium’s quarterly publication. What you know about this subject can become evidence, a case or a correction in the next edition, credited to you and published under CC BY 4.0 where it is your own work.
- メールゲートウェイ GW-GCRIこのページから送信するエビデンス、訂正、事例、手法、レビュー、翻訳、プロファイルへの入力。このページのタグが付いて届くため、担当の編集者が受け取ります。
- Zenodoオープンな寄託寄託してコミュニティに提出するORCID iD を使って Zenodo に自分の記録を公開し、そこから Reports コミュニティに提出してください。キュレーターが審査し、DOI は引き続きあなたのものです。
- 報告書GCRI貢献のすべての方法を見る貢献の種類、それぞれの審査方法、受け取るクレジット、ライセンス、決して受け付けないもの。
個人データ、重要インフラの機微な詳細、共有する権利のないものは決して送らないでください。寄稿は利用前にレビューされます。送付によって著作者としての権利や編集上の権利が生じることはなく、スポンサーシップで調査結果を買うことは決してできません。
手法、データセット、または問いを持ち込んでください。
情報提供と問い合わせは本日から受け付けています。問い合わせには参照番号が付き、次の手順が示されます。申請、支払い、計算資源の割り当て、登録は、担当チーム、規約、審査体制が整ってから開始します。
法的文書。
このサイトの地図、データ、図表は、背景と一般的な情報として示すものです。助言、予報、警報ではありません。情報の限界は免責事項とデータに関する注意事項に記載しています。