Files
Renderive/render_3D/datoviz/spec/release/COMMUNICATION.md
T
2026-08-14 01:54:38 +08:00

92 lines
3.2 KiB
Markdown

# Release Communication
Release communication should make the public release easy to understand and attractive to the right
audience without hiding limitations.
## Public Messages
Refine these messages for release notes, README, website, blog, and social posts:
1. Datoviz v0.4 is a modular C visualization engine for scientific GPU visualization.
2. The main user model is retained scene rendering with GPU-backed visuals and interaction.
3. Lower layers are available for advanced users who need DRP2, Vulkan runtime, canvas, stream,
offscreen, video, or backend integration.
4. The generated Python binding exposes the C API; high-level object-oriented plotting belongs to
VisPy2/GSP.
5. v0.4 prioritizes architecture, performance, and release proof over long-term API lock-in; v0.5
may still break API where needed.
## RC1 Positioning
Present RC1 as a controlled public testing milestone, not a broad launch.
Use language such as:
1. first release candidate;
2. early technical feedback;
3. C-first Vulkan rendering core;
4. retained scene layer;
5. DRP2 rendering contract;
6. low-level Python bindings;
7. offscreen and app rendering paths;
8. experimental WebGPU/WASM and optional provider slices.
Avoid language that implies:
1. final v0.4 stability;
2. production-ready plotting;
3. Matplotlib, VTK, or VisPy replacement;
4. VisPy 2.0 availability;
5. complete WebGPU parity;
6. fully stable public API.
## Blog And Announcement Assets
Generate launch assets from current examples, not mockups:
1. one short launch video;
2. several 10-20 second clips from gallery examples;
3. high-quality screenshots for README, website home page, blog post, LinkedIn, X, and release
notes;
4. architecture diagram showing `scene -> frame plan -> drp2 -> vklite/canvas -> stream/video`;
5. comparison-style examples inspired by strong scientific visualization galleries while keeping
Datoviz's own visual identity.
Public communication should link to exact examples, source files, docs pages, known limitations, and
release artifacts where possible.
## Channels
Use these channels when the release artifacts and docs are ready:
1. GitHub release;
2. project website;
3. release blog post or news page;
4. LinkedIn post;
5. X post or thread;
6. relevant scientific Python, visualization, and GPU communities;
7. direct notes to early adopters, collaborators, and selected dataset authors.
For RC1, prefer soft announcement channels first: GitHub release, project docs, direct early-adopter
messages, selected VisPy/GSP and scientific Python contacts, IBL/internal channels, and selected
graphics/Vulkan contacts. Delay broad channels such as Hacker News, large generic Python audiences,
and comparison-driven launch posts until final v0.4 unless there is a deliberate maintainer decision
to widen RC visibility.
## Feedback Policy
Each release candidate and final announcement should include a concrete feedback request:
1. what the project wants tested;
2. which APIs or examples are considered supported, experimental, advanced/unstable, deferred, or
external/GSP;
3. where users should report build, driver, docs, example, or correctness issues;
4. which known limitations are already tracked.
Known limitations must remain visible in the release notes, docs, and website.