model-writtenrun 6151 - may be wrong, links go to the source messages. Commit-by-commit LLM analysis is in the Commits dialog.
- Daniel Gustafsson opens the thread with a documentation-only series derived from an NSS discussion: standardize encrypted connections as SSL/TLS, reserve individual protocol names for specific protocols, add acronym and glossary entries, and clean up OpenSSL markup #1.
- Michael Paquier lands the OpenSSL markup portion as f80979f659, with text reflow and the accidentally included terminology changes removed; his follow-up identifies that misplaced change and questions the SNI link and separate SSL/TLS acronym tags #2.
- Daniel first adds an SNI glossary definition in the remaining series #3, but Michael proposes keeping these general networking terms on the acronym page with external references instead #4; Daniel adopts that approach and a single SSL/TLS acronym tag in #5.
- Michael commits the acronym portion of #5 as 15ff5401d1, making small link and markup adjustments, and confirms the landing while expressing support for the remaining terminology change #6.
- Peter Eisentraut objects that pervasive SSL/TLS wording adds tediousness and argues for either retaining familiar SSL terminology or switching fully to TLS #7; Jacob Champion points to Apache's SSL/TLS titles #8, while Daniel emphasizes consistency and suggests explanatory text and TLS index entries as a smaller alternative #9.
- Euler Taveira favors SSL/TLS as a bridge between existing configuration names and the current protocol, reports that the patch looks good, and submits a rebased remainder containing the broad replacement plus small leftover acronym edits #10.
- Peter maintains that existing documentation largely uses SSL consistently #11 #12, and Daniel acknowledges that retaining SSL with an explanatory paragraph would avoid the disputed churn #13.
- In September, Daniel replaces the broad rewrite with a three-file patch explaining that SSL terminology includes TLS and adding TLS index entries; the widespread replacements and residual acronym edits are left out #14.
- After a six-month gap, Robert Haas accepts the narrower clarification #15, and Daniel verifies that it still applies with offsets and builds successfully before announcing his intention to commit it for PostgreSQL 15 #16 #17.
- Daniel lands #14 unchanged as 860ea46ba7 and confirms completion in the final message #18; the thread ends with the markup cleanup, acronym additions, and explanatory clarification committed, while the wholesale SSL/TLS replacement has been abandoned.