Product Backlog
> Maintained by: Product Scout
> Last intelligence update: 2026-07-13 (10:04 UTC)
> Trigger: What I want for the commenting experience to be — Gardener expanded with 5 requested UX improvements, full competitive research corpus (Slite, Coda, Linear, Google Docs), and bridge link to Remodeling the web thesis
---
NEW — Commenting UX Suite (trigger: 2026-07-13 @ 10:04 UTC)
Opportunity #1: Full-Window Discussions Panel
Problem: Users cannot focus on discussions because the panel is cramped in a sidebar. Reviewing many comments with constant context-switching is fatiguing.
Evidence: Point #1 in What I want for the commenting experience to be. The Discussions Panel Is Confusing doc shows broken thread tree with mismatched reply counts.
Users: Seed users doing document review, collaborative editing, async discussions.
Solution: Dedicated full-window discussion mode (toggle from sidebar). Linear's full-issue view as pattern. Document hides, thread tree fills viewport — collapsible subtrees, visible chains, block-level comment linking.
MVP: Toggle button in discussions panel header switching sidebar split-view ↔ full-window thread view. Rendering already exists — needs container + state machine transition: discussions: { sidebar | fullWindow }.
Distribution: Seed product UI. Announce via short post.
Pricing: Core Seed feature.
Competition: Notion (sidebar only), Coda/Google Docs (margin), Linear (sidebar + full-window — best reference).
Confidence: 9/10 — directly requested, UX pain confirmed, low implementation risk.
First experiment: Prototype toggle on one document. Measure click rate, time-in-mode, perceived fatigue.
---
Opportunity #2: Block-Addressable Comments (Seed Differentiator)
Problem: Comments are not first-class linkable blocks. Cannot copy a link to a specific comment, create block ranges across comments, or reply with precise context anchor. Discussions are ephemeral.
Evidence: Point #2 in What I want for the commenting experience to be. Competitive research corpus (Slite, Coda, Linear, Google Docs) confirms no platform treats comments as block-addressable, range-linkable primitives. This is a genuinely novel positioning for Seed.
Users: Seed power users, collaborative writers, knowledge base maintainers.
Solution: Comments become first-class HM blocks — stable block IDs, block ranges (hm://.../doc#comment-ABC..comment-XYZ), linkable/embeddable/citeable like document blocks. Extend the document machine block model to the comments resource.
MVP: Assign stable block IDs to comment blocks. Show a "copy link" action on each comment (like Linear's three-dot menu). Enable click-to-highlight when navigating to a block-linked comment.
Distribution: Seed product. Announce via comparative blog post: "What Block-Level Commenting Unlocks" — position against the 4 platforms researched.
Pricing: Core Seed feature. Becomes a key selling point vs. Notion/Coda.
Competition: None — no platform does this. Closest: Linear's per-comment permalink but not block-ranges. This is a genuine differentiator.
Confidence: 8/10 — competitive research corpus is complete, the author explicitly wants this. Risk: document machine must support comments as blocks (depends on data model).
First experiment: Give one comment a stable block ID, verify it can be linked to from another document and renders correctly. If this works, the architecture is sound.
---
Opportunity #3: Mobile Comment Editor Remediation
Problem: The mobile comment box is laggy, janky, and broken with links, embeds, mentions, and media. The drawer does not work properly. Users cannot trust the editor. This kills mobile engagement.
Evidence: Points #3–4 in What I want for the commenting experience to be: "mobile drawer does not work properly, very janky" and "comment box is laggy... makes users not trust the editor." The Mobile formatting toolbar project documents the broken floating toolbar, transparent slash menu, and iOS keyboard handling issues — but focuses on the formatting toolbar, not the editor core.
Users: Seed mobile users (iOS Safari, Android Chrome). Growing mobile web audience.
Solution: A targeted fix-release cycle focused on:
Replace janky drawer with a bottom-sheet (iOS native pattern) or full-screen composer
Fix ProseMirror editor rendering inside mobile comment context (lag with embeds/mentions)
Stable keyboard handling (sticky vs. fixed positioning, iOS Safari compatibility)
MVP: Three focused sprints: (1) drawer → bottom-sheet, (2) editor lag profiling + fix, (3) keyboard handling. Each sprint ships independently.
Distribution: Seed product. The fix-and-tell content pattern is proven — write "How we fixed mobile commenting on Seed" as a technical walkthrough after the fix.
Pricing: Core Seed feature.
Competition: Notion mobile (struggles with the same issues), Google Docs mobile (works because read-only review mode). This is a baseline expectation.
Confidence: 7/10 — clear pain, but engineering complexity. The document machine must not break. The mobile formatting toolbar project is already underway, providing momentum.
First experiment: Profile the editor on a mid-range Android device. Identify the bottleneck (rendering, state updates, or keyboard events). Fix the single worst offender.
---
Opportunity #4: Discussions Panel Thread Tree Redesign
Problem: The thread tree is deeply confusing. Reply counts don't match visible comments. Clicking "Reply (8)" shows only 3 comments. Nested replies are hidden under multiple expansion clicks. Users cannot see the full thread on one screen.
Evidence: Discussions Panel Is Confusing contains direct screenshots and analysis:
Comment shows "Reply (8)" but only 3 replies visible
Clicking expands but still shows wrong count
Horacio's comment shows "Reply (2)" but replies are actually 5 comments in 2 chains
User asks: "Don't we really want a way to see the entire thread on one screen?"
Users: All Seed users who read or participate in discussions.
Solution: Flat thread tree with collapsible subtrees (like HackerNews, YouTube, Reddit). Total reply count matches visible comments. All replies visible without multi-level expansion. Option to "show entire thread" on one screen.
MVP: Replace the current thread tree component. Use a simple recursive renderer: top-level comments with collapsible children. Count = children.length. Done.
Distribution: Seed product UI.
Pricing: Core Seed feature.
Competition: Standard pattern — HN, YouTube, Reddit, Linear all do this correctly. Seed is an outlier in the wrong direction.
Confidence: 10/10 — this is a bug, not a feature request. The thread tree is objectively broken based on the screenshots. Fix is well-understood and low-risk.
First experiment: In the Discussions Panel Is Confusing document, verify that the current version still exhibits the bug. If yes, the fix is straightforward.
---
Existing — Web Remodeling Cluster Consolidation
Remodeling the web update (See Also expansion 2→5 links) solidifies the Web Remodeling cluster at 9 docs across 5 accounts with narrative arc: broken internet → sealed containers → trust crisis → Seed solution → architectural expression. See Emerging Topics Topic #4.
---
Previously Tracked — Commenting UX Suite (cross-ref)
The Strategy Opportunity Backlog identifies 3 strategic editorial opportunities from this same trigger: "What Block-Level Commenting Unlocks" (editorial), "What We Learned from Studying Slite, Coda, Linear, and Google Docs Comments" (comparative UX analysis), and "Commenting That Works on Mobile First" (editorial). These are editorial/content opportunities. The opportunities above are product/engineering opportunities. They complement each other.
---
Changelog
2026-07-13 — Added Commenting UX Suite (4 opportunities) from Gardener trigger on What I want for the commenting experience to be. Preserved existing Web Remodeling note. Added cross-reference to Strategy Opportunity Backlog.
Do you like what you are reading? Subscribe to receive updates.
Unsubscribe anytime