I drafted an NIP; could anyone review it before I publish it? Motivation: many web clients render the same articles, causing duplicated content in search engines. Adding a single canonical reference c...
Two pieces of prior art worth citing in the draft:
1. nips PR #856 (Oct 2023, closed): "For improved SEO of web clients, add canonical tag to NIPs 23 and 52". Same motivation as yours. tyiu and erskingardner pushed back on tying events to a single webpage, and the author closed it saying a plain ["r", "<url>", "canonical"] tag does the job. That's exactly your tag shape, so it helps to link it and say what's new this time (your replaceable-only rule and the full-content requirement both answer the old objections).
https://github.com/nostr-protocol/nips/pull/856
2. NIP-48 proxy tags: ["proxy", "<url>", "web"] already marks an event as originating elsewhere on the web, and the NIP says clients may use it to reconcile duplicates. A line on when to use which would avoid confusion (proxy = content not born on Nostr; canonical = author's preferred web URL).
On fiatjaf's point: Google's duplicate-URL docs list rel=canonical as a strong signal, not a rule (redirects rank above it), and Google has said since 2008 that duplicates get filtered, not penalized, unless they're deceptive. So "helps the original get picked as the version shown in search" is safer wording for the Motivation than "prevents penalties".
https://developers.google.com/search/docs/crawling-indexing/consolidate-duplicate-urls
1. nips PR #856 (Oct 2023, closed): "For improved SEO of web clients, add canonical tag to NIPs 23 and 52". Same motivation as yours. tyiu and erskingardner pushed back on tying events to a single webpage, and the author closed it saying a plain ["r", "<url>", "canonical"] tag does the job. That's exactly your tag shape, so it helps to link it and say what's new this time (your replaceable-only rule and the full-content requirement both answer the old objections).
https://github.com/nostr-protocol/nips/pull/856
2. NIP-48 proxy tags: ["proxy", "<url>", "web"] already marks an event as originating elsewhere on the web, and the NIP says clients may use it to reconcile duplicates. A line on when to use which would avoid confusion (proxy = content not born on Nostr; canonical = author's preferred web URL).
On fiatjaf's point: Google's duplicate-URL docs list rel=canonical as a strong signal, not a rule (redirects rank above it), and Google has said since 2008 that duplicates get filtered, not penalized, unless they're deceptive. So "helps the original get picked as the version shown in search" is safer wording for the Motivation than "prevents penalties".
https://developers.google.com/search/docs/crawling-indexing/consolidate-duplicate-urls