Zed is our office

Original Article ↗ Hacker News Discussion ↗

Collaboration as Core Concept

  • Many were surprised to learn Zed was built around real‑time collaboration from day one, not as a bolt‑on.
  • Supporters see integrated shared docs, channels, and cursors as a powerful medium for remote teams, training juniors, and code walkthroughs without screen sharing.
  • Others view it as better suited to shared note‑taking than serious coding.

Pair Programming and Mass Live Editing

  • Strong divide on pair programming: some find high‑bandwidth, shared‑cursor work invaluable; others “hate” it and prefer async review.
  • Multi‑cursor “dozens of people editing a file” provokes anxiety; critics call it distracting and chaotic, defenders say tools still allow turn‑taking and selective use.

“Slack in the Editor” & Attention Concerns

  • A big worry is turning the editor into yet another chat client, fragmenting comms across Slack + Zed and creating pressure to follow continuous streams.
  • Some argue this undermines thoughtful, discrete communication (commits/PRs, documents) and feeds attention‑economy dynamics.
  • Others say it’s opt‑in, easier to mute than Slack/email, and can be a lighter‑weight way to jump into focused joint problem‑solving.

AI Integration and Product Direction

  • Several commenters say they lost interest when AI features arrived, seeing it as a shift from a clean editor toward hype‑driven bloat.
  • Others think AI is commercially necessary, works well in Zed, and can be fully disabled.
  • Mixed experiences: some prefer Zed’s Claude integration to raw Claude; others find CLI‑based Claude more effective than Zed’s UX.

Self‑Hosting, Security, and Enterprise Use

  • Strong demand to self‑host collaboration for privacy/compliance; self‑hosting existed, was dropped during infrastructure changes, and is promised to return later.
  • Until then, many doubt enterprises will route code and comms through Zed’s servers.

Version Control and DeltaDB

  • Some extrapolate Zed’s model to “living” codebases: continuous edits, testing, and deployment with less Git ceremony.
  • Others insist on human‑curated commits and stable checkpoints; fear of noisy auto‑commits and multi‑agent editing is common.
  • DeltaDB (operation‑level version control) intrigues some but raises lock‑in and complexity concerns.

Editor Fundamentals, Performance, and UX

  • Zed is widely praised for startup speed and responsiveness versus VS Code and JetBrains, especially on weaker hardware.
  • At the same time, multiple users report rough edges: flaky collab/voice, file‑watch desync, container/remote dev friction, Windows terminal issues, blurry text on many displays, missing or buggy basics (wrapping, LSP stability, devcontainers, multi‑monitor ergonomics).
  • Some feel core reliability and extensibility should be prioritized over new collab/AI/VCS layers.

Ecosystem, Standards, and Adoption Barriers

  • Lack of a standardized, editor‑agnostic collab protocol (like LSP) is seen as a major barrier; current tools assume everyone uses the same editor.
  • Historical tools (SubEthaEdit, Gobby) and existing solutions (VS Code Live Share, JetBrains Code With Me, external pair‑tools) are frequently referenced for comparison.
  • Many teams are entrenched in VS Code/JetBrains + Slack/Jira/Confluence + devcontainers, making “Zed as the office” feel unrealistic outside Zed’s own company.

Who This Seems to Be For

  • Commenters infer Zed is optimized for a VC‑funded, remote, developer‑heavy org that lives in the editor and wants deeply integrated code‑centric collaboration.
  • Indie developers, small polyglot shops, and people who prize minimal, distraction‑free tools often feel they’re not the target audience and stick with Neovim/Sublime/VS Code.