香港紀事館Hong Kong Story · A sourced knowledge base of Hong Kong

Terms · 使用條款

內容使用條款

Terms of use — what you may do with this site

呢頁講嘅係「使唔使用得」：邊啲可以直接做、邊啲要事先取得書面同意、機器要點取用。 各類內容嘅權利歸屬（官方資料、編採圖片、本站正文）只寫喺編輯方法 § 內容權利 一處，呢頁唔會另寫第二份。

條款版本：二○二六年十月二十五日。以本頁最新版本為準；版本日期一改，即代表條款已更新。

繁體中文

一、可以直接做，毋須事先詢問

  • 閱讀、列印、下載任何一頁，作個人研習之用。
  • 由任何頁面連到本站，或把網址分享出去。
  • 引用句子或節錄（包括學術論文、教材、新聞評論、簡報），條件是同時註明出處：站名「香港紀事館」、 該頁完整網址，並盡量保留原文嘅統計時點（asOf）同來源機構名稱——數字離開時點就無從核對。 引用範圍要合乎適用嘅版權法（引用權／合理使用）。
  • 用搜尋引擎索引本站，或由 AI 代理按第三節取用本站內容作答。
  • 程式讀取 /api/v1/index.json、/llms.txt、/openapi.json 等機器出口， 作研究、核對或引用；引用格式同取用次序見 /llms.txt 嘅代理合約一節。

二、要事先取得書面同意

  • 全文或大部分轉載，不論媒介（網站、App、電子報、紙本、影片字幕）。
  • 商業再利用：付費產品或服務、廣告素材、收費課程教材、顧問報告、資料庫轉售，或把本站內容 當成自身產品嘅主要賣點。
  • 改寫、翻譯、重組後對外發布而唔標明來源，或把本站編輯判斷（例如「已核實」「缺口」） 講成自己嘅結論。
  • 大量抓取後另建鏡像、衍生資料集或「平行版本」對外提供（包括換個介面、換個名再發布）。
  • 移除或遮蔽出處、統計時點、缺口聲明（caveat／gaps），令內容睇落比原頁更權威。
  • 使用本站名稱、標誌或版面，暗示本站認可、合作、認證、代你背書，或令讀者誤以為內容由你發布。

要取得同意，請經營運者網站聯絡Day Dream Production HK，講明用途、範圍同會唔會涉及商業成分。未取得書面同意前，請假設「唔可以」。

三、機器取用（爬蟲、AI 代理、API）

  • 「可唔可以爬」睇 robots.txt；「爬完可以用嚟做咩」睇同一個檔每個組嘅Content-Usage 行。呢兩件事嘅現行取態只喺嗰度公布一份，本頁唔另寫數值—— 第二份數值一定會同 robots.txt 漂移。
  • 「應該點用」睇 /llms.txt 嘅代理合約（取用次序、引用格式、時點要求）， 同 編輯方法 嘅「AI 代理如何使用本站」一節。
  • 本站冇 hosted MCP 伺服器、冇 API key、冇查詢語言；支援 OpenAPI 工具註冊嘅代理可以直接用/openapi.json。
  • 請守禮貌：唔好令本站或其他讀者受影響（例如並發過高、無視 Disallow、重複抓取同一頁）； 遇到錯誤或限流請退避再試。惡意行為（例如繞過 robots.txt 嘅限制、把本站當成免費算力或代理轉發） 一律視作違反本條款。本站會對異常爆發流量回 429；呢個只係濫用防護，唔改變robots.txt 對搜尋引擎同 AI 代理嘅允許政策。
  • 本站會隨資料同政策更新 robots.txt 同機器出口；代理每次取用前應重新讀取，唔好快取政策。

四、更正、下架與投訴

  • 發現事實錯誤：本站以官方原文為準，會先核對來源再改；改唔到嘅會寫成缺口，唔會靠估補一個數。 更正原則見編輯方法。
  • 來源機構、圖片權利人或其代表認為本站取用方式超出範圍：請經營運者網站聯絡，附上具體頁面、 涉及內容同權利依據。本站會先暫停展示相關內容，再核對來源同授權記錄（圖片嘅授權、作者、時點同 sha1 全部有登記），並在合理時間內回覆。
  • 涉及私隱、個人資料或安全嘅請求，同樣經上述渠道處理，唔需要先公開討論。

五、呢頁唔係咩

  • 呢頁係使用規則，唔係開放授權，亦唔係法律意見。它唔會把本站正文變成 CC 或其他 SPDX 授權；結構化資料嘅 license 因此維持留空——唔會編一個我哋無權授予嘅條款出嚟。
  • 官方資料第三方再用時，條件以原發布機構公布為準（例如「資料一線通」嘅授權條款）；本站唔會代為授予 任何權利，亦唔代表該機構認可本站。
  • 本頁中英兩版如有歧義，以中文版為準；同任何適用法律有衝突嘅條文，以法律為準。

English

1. You may do this without asking

  • Read, print or download any page for your own study.
  • Link to any page, or share its address.
  • Quote sentences or short excerpts (in papers, teaching materials, news commentary, slides), provided you credit the source: the site name “Hong Kong Story”, the full page URL, and — wherever possible — the reference date (asOf) and the publishing body. A figure without its date cannot be checked. Quotation must stay within whatever copyright exception applies to you.
  • Index the site in a search engine, or use it through an AI agent under section 3.
  • Read the machine endpoints — /api/v1/index.json,/llms.txt, /openapi.json — for research, verification or citation. The access order and citation format are in the agent contract section of/llms.txt.

2. You need written permission for this

  • Republishing a page in full, or most of it, in any medium (websites, apps, newsletters, print, subtitles).
  • Commercial reuse: paid products or services, advertising, paid course materials, consultancy reports, database resale, or making our content the main selling point of your own product.
  • Rewriting, translating or rearranging our text for publication without attribution, or presenting our editorial judgements (for example “verified” or “gap”) as your own findings.
  • Bulk-crawling the site to build a mirror, a derivative dataset or a parallel edition for others (including republishing under a different interface or name).
  • Removing or obscuring the source, the reference date or the gap statements (caveat / gaps) so that the material looks more authoritative than the page it came from.
  • Using the site’s name, logo or layout to suggest endorsement, partnership or certification, to imply that we vouch for you, or to mislead readers about who published the content.

For permission, contact Day Dream Production HK through the operator’s website, stating the purpose, the scope and whether there is any commercial element. Until permission is given in writing, assume the answer is no.

3. Machine access (crawlers, AI agents, API)

  • Whether you may crawl is set in robots.txt; what you may do with what you fetch is set in the Content-Usage line of each group in the same file. Both are published there only, once. This page deliberately does not restate the values — a second copy would drift from robots.txt.
  • How to use the site is in the agent contract section of /llms.txt (access order, citation format, date requirements) and in “How AI agents should use this site” on theeditorial method page.
  • There is no hosted MCP server, no API key and no query language. Agents that support OpenAPI tool registration can use /openapi.json directly.
  • Be considerate: do not degrade the site for other readers (excessive concurrency, ignoringDisallow, repeatedly fetching the same page). Back off on errors or rate limits. Abuse — including circumventing robots.txt restrictions or using the site as free compute or an open proxy — is a breach of these terms. Abusive bursts may receive HTTP 429; that is abuse protection only and does not change the allow policy for search engines and AI agents inrobots.txt.
  • robots.txt and the machine endpoints change as data and policy change. Re-read them before each run rather than caching the policy.

4. Corrections, takedowns and complaints

  • If a fact is wrong, the official text prevails: we check the source first, then amend. What cannot be sourced is written as a gap, never filled with an estimate. See theeditorial method.
  • If a publishing body, an image rights holder or their representative believes our use exceeds the permitted scope, contact us through the operator’s website with the specific page, the material concerned and the basis of the right. We will pause display of the material while we check the source and the rights record (licence, author, date and sha1 are logged for every editorial image) and reply within a reasonable time.
  • Privacy, personal-data and safety requests go through the same channel; no public discussion is needed first.

5. What this page is not

  • These are rules of use, not an open licence and not legal advice. They do not turn our text into a CC or other SPDX licence; that is why the license property in our structured data stays empty — we will not invent terms we have no right to grant.
  • Reuse of official material remains subject to the publishing body’s own terms (for example the DATA.GOV.HK licence). We grant no rights on their behalf, and our citation does not imply their endorsement.
  • If the Chinese and English versions differ, the Chinese version prevails; where a provision conflicts with applicable law, the law prevails.
↑