// the take
A Flask + Redis dashboard server for pinning matplotlib/d3-style charts created in IPython notebooks onto a shareable, drag-and-resize grid, with an optional SQL editor view. Aimed at data scientists in 2015 who wanted to turn notebook plots into a live dashboard without standing up something like Zeppelin or Redash.
Decomposing 'push a chart from a notebook to a shared dashboard' into a small Flask app plus Redis for real-time updates was a reasonable, lightweight design for its era. The gridstack-based layout gives free drag/resize/arrange behavior instead of hand-rolled CSS grids. It also ships an actual test suite wired to Travis and Coveralls, which is more rigor than most notebook-adjacent tools of that generation bothered with.
The README itself is the maintainer announcing he abandoned the project years ago in favor of a blog-post workaround, and the last real push was in 2021 - this isn't just stale, it's explicitly unmaintained. There's no authentication or access control at all; 'login management' is still listed as an unimplemented future item in the changelog, which matters because the whole pitch is sharing dashboards, including ones meant to be private. Getting it running today means fighting Python 2-era tooling (nosetests, `mysql.server start`, Flask's dev server) and vendored-in dependencies frozen at jQuery 2.1.4, Bootstrap 3.3.5, and D3 v3. The project's own coverage output shows the parts someone would actually touch - SQL rendering, storage - sitting at 59-62% coverage while other modules show 0%, so the riskiest code paths are the least tested.
We read the README, the directory tree and a key source file in October 2026. Nobody pays to be listed. How we judge →