id: meta-news
name: Meta data center news
domain: about.fb.com
kind: news
mode: rss
priority: 2
license: "Publicly published corporate newsroom feeds; headlines, dates and facts only, attribution kept"
attribution: "Meta — about.fb.com/news ; datacenters.atmeta.com/blog"
homepage: https://about.fb.com/news/
coverage: { global: true, operators: ["Meta"] }
implementation: cloud_newsroom
fetch:
  level: 1
  maxLevel: 2
  rpm: 10
  concurrency: 1
  respectRobots: true
  maxCreditsPerRun: 0
discovery:
  rss:
    - "https://about.fb.com/feed/"
  sitemap: ["https://datacenters.atmeta.com/post-sitemap.xml"]
  include: ["about\\.fb\\.com/news/20\\d{2}/", "datacenters\\.atmeta\\.com/20\\d{2}/\\d{2}/"]
  exclude: ["/feed/?$", "\\?share=", "/wp-json/"]
  classify:
    - { pattern: "datacenters\\.atmeta\\.com/20\\d{2}/\\d{2}/", group: newsroom, pageType: press_release, priority: 75 }
    - { pattern: "about\\.fb\\.com/news/20\\d{2}/", group: newsroom, pageType: press_release, priority: 65 }
schedule:
  newsroom: daily
  rss: daily
  sitemap: daily
params:
  include: "\\b(data ?cent(er|re)s?|datacenter|AI infrastructure|campus|invest(s|ing|ment)? (of |more than |over )?(\\$|€|£)\\d|(\\$|€|£)\\d+(\\.\\d+)? ?(billion|million)|gigawatt|megawatt|\\bMW\\b|\\bGW\\b|Hyperion|Prometheus|compute cluster|break(s|ing)? ground|(we are|now|comes|is) online|cooling|substation|clean energy|nuclear|geothermal)\\b"
  exclude: "\\b(Instagram|WhatsApp|Threads|Messenger|Quest|Ray-Ban|Horizon|Reels|creators?|teen accounts?|election)\\b"
defaults:
  operatorName: Meta
notes: >
  about.fb.com/feed (10 items, mostly product news) filtered on title/summary; the Meta Data Centers blog
  (datacenters.atmeta.com — "Kansas City, we are online!", "Hello, Bowling Green!", grants) has no working RSS
  (/feed/ returns the HTML home page) but publishes a Yoast post-sitemap.xml (~68 dated posts). discovery.include
  keeps only dated article URLs so static pages (community-grants, funds) are not turned into news. Both hosts allow crawling.