Skip to content

Latest commit

 

History

History
84 lines (66 loc) · 4.47 KB

lib-collab-crdt-community-pm-showcase.md

File metadata and controls

84 lines (66 loc) · 4.47 KB
title tags created modified
lib-collab-crdt-community-pm-showcase
community
crdt
showcase
2023-11-01 03:15:08 UTC
2024-08-11 07:21:11 UTC

lib-collab-crdt-community-pm-showcase

guide

discuss

  • Show HN: Briefer – Multiplayer notebooks with schedules, SQL, and built-in LLMs | Hacker News _202407

  • Can you tell me how it compares to Deepnote.com please?

    • The products are similar in the sense that both are cloud-based notebooks, but we have different approaches to building them.
      1. We want to go beyond just notebooks. We want to centralize all data tasks in Briefer, including traditional BI so we’re building specific features for that, like dashboarding and everything that comes with it.
      1. We’ll enable people to use their own compute for the notebooks.
      1. We want to allow people to manage notebooks, dashboards, and apps “as code” so they can version it whenever necessary.
    • We unfortunately didn’t finish items 2 and 3 yet because we’ve only started working on Briefer six months ago, but we’re fast and we’ll get there soon.

  • Architectures for Central Server Collaboration

  • https://news.ycombinator.com/item?id=40579892

  • 📓 My presentation about "Collaborative editing in Jupyter" at PyData Paris has been accepted _202406

  • https://x.com/davidbrochart/status/1796838314386866396

    • I'll talk about the latest developments and how CRDTs are making their way and redefining the architecture of Jupyter(Lab): chat system, server-side execution...

  • I'm not sure how collaboration plays with JupyterHub, since you get a dedicated Jupyter Server/Lab instance for each user. I don't know if there is a way to share a server between multiple users with JupyterHub?

  • Interested to know how collaborative editing in jupyter can play a roller in a jupyter hub

    • If it's not possible now, I think we should find a way to fund the dev work to bring this to jupyterhub. It feels like a no brainer since the hub is the most common multi-user technology in the jupyter ecosystem. Would have a huge impact

  • RTC on JupyterHub is documented here

  • https://x.com/davidbrochart/status/1796138236516991392

    • CRDTs are making their way into JupyterLab, with a new collaborative input widget
    • This allows multiple users to type into the input box, and to recover the widget after closing/reopening the window, preventing the notebook from deadlocking.

  • A long-standing issue in Jupyter is about to get fixed: getting back a notebook state when reconnecting.

  • https://twitter.com/davidbrochart/status/1722564382813516093

    • It will be available in jpterm and jupyverse first, then in JupyterLab.
    • This demo shows that outputs are restored but it will also work with widgets. It's all powered by CRDT.

  • I think jpterm will be the playground for most innovations in the Jupyter ecosystem, before they land in JupyterLab. The reason is the same as the one behind Textual: it's easier to use than web technology.

  • 🌰 分享我设计的基于CRDT的软件架构 hamsterbase

  • https://twitter.com/hamsterbase/status/1590005075581497344

    1. 以 CRDT 文件作为 single source of truth, 保证向前向后兼容。
    2. 以 sqlite 数据库作为缓存。 sqlite 的数据库格式可以按照需求任意设计，可以随意修改。
    3. 不同客户端直接通过点对点同步 CRDT 文件，支持完全离线可用，同步无冲突。

  • 数据库的crdt同步怎么设计的

    • 一个网页对应一个 CRDT 的文件（类似于 JSON）
    • 每次同步的时候，获取对方的文件列表、本地的文件列表
    • 对列表进行 DIFF, 发送本地多的，拉取本地少的。

  • 你可以在本地跑 hamsterbase 试试看。 文件结构都在本地的

    • 可以把文件的版本号放在前面，这样可以在不读取完整文件的情况下获取当前的所有 CRDT 文件。

  • 可以把 CRDT 同步抽象出来，同步服务不需要关心 CRDT 的格式

    • 这个是我从零设计的。 CRDTSyncDir 可以用 webdav , fs , oss , http 实现。
    • 每次同步就是拿本地 fs 实现的 dir 和 http 实现的 remote dir 同步。
interface CRDTSyncDir{
  listAllFiles();
  fileStat(id,version)
  listVersions(id)
  getVersion(id)
  read(id,version,filter)
  write(id,content)
}