Anu + Rufus
Anu + Rufus
- Get updates from Anu e.g. re vital strategies
- Vital strategies meeting coming up
- Very good BoE demo
- What do they want: confidence, sustainability (they can pay b/c they want to, and freedom and back-stop if someone disappears)
- Rufus updates
- Started DataHub Next and moved material over
- Question: what is product? (again!)
- Brainstorm directions and questions ✅2023-03-07 go with data canvas / data stories for now
- Data Project
- Data Canvas / Scratchpad
- Data Stories
- What's the MVP 🚧2023-03-07 see https://app.excalidraw.com/l/9u8crB2ZmUo/8RvBmMyyL38
- How would you edit the data stories locally ✅2023-03-07 maybe just use obsidian 😉 for now. see e.g. https://github.com/coddingtonbear/obsidian-csv-table
- Build my issue tree in the sky ✅2023-03-07 started at ../ultimate-issue-tree-2023-03-07.canvas
- AOB
- I'm hesitating to put half-baked stuff on datahub-next. what do we think?
- Option: stay with product repo and migrate things as more polished / finalized. => keep journal in product.
- 👍 free and simple. no concern of half-baked stuff. comments / critique etc.
- Option 2: go all in on datahub-next => keep most stuff there
- 👍 think in public
- 👍 potentially traffic etc
- 👍 eat our dogfood with a public digital garden
- Try it and see …
- Option: stay with product repo and migrate things as more polished / finalized. => keep journal in product.
- sveltekit: discussed and added to the great big question tree ✅2023-03-07 for now no compelling reason on our main projects to go with that
- Need a good brand name for "users creating their own content with your data" e.g. Data Labs, Data Stories, Data Canvas, Data Scratchpad
- I'm hesitating to put half-baked stuff on datahub-next. what do we think?
Specific questions: #todo move this stuff to ultimate issue tree
- Single file vs multiple files multiple but can start with single file
- Small data vs big data small to meduim?
Rufus
Plan
- Content base plan 🚧2023-03-07 in progress below
Scratch
- datasette for our simple data APIs and browsing …
- Loader: https://github.com/simonw/csvs-to-sqlite/
- datasette lite
- datasette itself
- Publishing data (instructions for deploying to google cloud run): https://docs.datasette.io/en/stable/publish.html
Ideas:
- Create a demo obsidian repo with data
- Essay: A modern lightweight data publishing workflow
- Steps: publish, store, present
- Essay: (even simpler) Use obsidian as your desktop data publishing app
- Question: non-technical user creating a dataset …
- Conversion of data and knowledge management in an org
Content Base
Architecture: https://app.excalidraw.com/l/9u8crB2ZmUo/1YKW635KsAU
- How would we do a blog with a good content base?
- What objects do we have?
- posts
- date
- authors: Reference(Person)
- people
- posts
- What components do we have?
- Index page (listing)
- Post page
- What objects do we have?
Essays
Why we are using markdown as a our base format …
Why we are betting on markdown …
Why is markdown so special
Subject: Markdown is amazing and here's why (Markdown is the unix of formats / Markdown is eating the world / Markdown has superpowers, here's a few of them)
Question: what's so special about markdown?
#todo dig out the stuff i've done before.
- Plain text yet has all key features for rich text (not binary) … which means what
- Extensible
- Lots of tooling
- Structured data/content via frontmatter and blocks and unstructured
- Just right: simple but not too simple
- Can combine structured and unstructured content in one file
- Human-readable
- Human writable (and forgiving)
Issues
- Bundling assets e.g. images
Asides
- ask S re naming for pyramid stuff
- https://blot.im - a flowershow alternative
- https://www.reddit.com/r/ObsidianMD/comments/u7r2a5/a_free_open_source_obsidian_publish_alternative/ - on google first page for obsidian publish
