SiYuan
SiYuan's 3.8 line is eight builds deep and still stabilizing the same four features
A side-by-side editorial comparison of CommaFeed and EGroupware — release velocity, themes, recent moves, and the top alternatives to consider.
A self-hosted RSS reader that now spends most of its release notes on security.
CommaFeed is a self-hosted Google Reader replacement that has spent the 6.x-to-7.x stretch rebuilding its trust boundaries rather than its feature list. SSRF protection has been tightened in four consecutive releases, feed parsing now strips javascript: URLs before they reach the database, and OPML imports are filtered against import-time abuse. The 7.x line pairs that with client interop: Fever API, ReadKit compatibility, and now the Google Reader API.
EGroupware's feed has become a security treadmill, with the 23.1 branch days from end of life.
Almost every release in this window is a security release, and most ship as a pair — one build for 26.x, an identical-scope backport for 23.1. Feature work exists but arrives as single lines inside those security notes: WebAuthN two-factor moving out of the former EPL add-on, OpenID bumped to current upstream libraries, invoice XML gaining reverse-charge exemption codes. The recurring theme in the fixes themselves is authentication — OAuth bearer tokens, OpenID regressions, an inverted email_verified check, passkeys broken by a missing proxy config.
CommaFeed is a self-hosted Google Reader replacement that has spent the 6.x-to-7.x stretch rebuilding its trust boundaries rather than its feature list. SSRF protection has been tightened in four consecutive releases, feed parsing now strips javascript: URLs before they reach the database, and OPML imports are filtered against import-time abuse. The 7.x line pairs that with client interop: Fever API, ReadKit compatibility, and now the Google Reader API.
The project is positioning itself as a sync backend other people's apps talk to, not just a web reader. Each release adds another protocol or client fix while hardening the assumption that an instance may be publicly exposed with untrusted users on it. The 7.0.0 swap from JEXL to sandboxed CEL and the 7.3.0 flip of blockLocalAddresses to a secure default both trade self-hoster convenience for a safer out-of-the-box posture.
Expect continued client-protocol coverage and further SSRF narrowing; the recurring pattern in these entries is that every new fetch path gets a follow-up hardening release.
Almost every release in this window is a security release, and most ship as a pair — one build for 26.x, an identical-scope backport for 23.1. Feature work exists but arrives as single lines inside those security notes: WebAuthN two-factor moving out of the former EPL add-on, OpenID bumped to current upstream libraries, invoice XML gaining reverse-charge exemption codes. The recurring theme in the fixes themselves is authentication — OAuth bearer tokens, OpenID regressions, an inverted email_verified check, passkeys broken by a missing proxy config.
The project is consolidating everyone onto 26.x and has put a date on it: security coverage for 23.1 ends August 15, 2026, repeated as a warning in every 23.1 build since July. After that the twin-release pattern should stop and the cadence should halve. The authentication stack is the least settled part of the codebase right now — the July and August releases keep re-fixing OpenID and WebAuthN regressions introduced by their own predecessors.
Expect 23.1 builds to cease after August 15 and the feed to become single-track 26.x. On the evidence of the last six releases, expect at least one more OpenID or WebAuthN regression fix to follow the recent upstream library bump.
Other Collab products tracked by Sparkpulse, ranked by recent ship velocity. Each card links to a full editorial trajectory and lets you pivot into a head-to-head comparison with either CommaFeed or EGroupware.
SiYuan's 3.8 line is eight builds deep and still stabilizing the same four features
Teable is spending this release cycle making computed fields and admin recovery trustworthy
Copilot's model roster churns weekly while GitHub quietly rewires policy and billing plumbing
Governance and cross-workspace plumbing, not new AI surface
A nightly canary train where the AI platform work hides between dependency bumps
Self-hosted dashboard that finished its feature arc and stopped shipping in 2023.
See all CommaFeed alternatives → · See all EGroupware alternatives →
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
Both compete on the same themes — self-hosting — within Collab. CommaFeed and EGroupware are shipping at a similar cadence (velocity 5.0 vs 5.0, both within Sparkpulse's "active" band). See the at-a-glance table above for a side-by-side breakdown of velocity, recent sparks, and editorial themes.
Sparkpulse doesn't pick a winner — we score release velocity, not feature parity. CommaFeed and EGroupware are shipping at a similar cadence (velocity 5.0 vs 5.0, both within Sparkpulse's "active" band). For your specific use case, the alternatives sections above list other Collab products to evaluate alongside.
Top CommaFeed alternatives in Collab are ranked by recent ship velocity. Browse the "CommaFeed alternatives" section above for the current picks, or visit /alternatives/commafeed for the full list with editorial commentary on each.
Top EGroupware alternatives in Collab are ranked by recent ship velocity. Browse the "EGroupware alternatives" section above for the current picks, or visit /alternatives/egroupware for the full list with editorial commentary on each.