GitP/How To/Prepare Podcast Metadata
Prepare Podcast Metadata
- Prepare the episode in this garden, then mirror it into Person/codekiln/GitHub/gitpa for publication. The session page supplies the episode text and directly embeds the recording’s MP3 URL.
Prepare the session
- Create or update the recording's episode page under
GitP/A/Session/YY/MM/DD-Day. Set logseq-entity:: [[Logseq/Entity/Podcast/Episode/GitP/A]], [[Logseq/Entity/Podcast/Episode]] in its frontmatter. Link the music-making log and add the supported session details, preset downloads, and recording asset. - Give the page one H1 episode title and a short description as its first child. Embed the MP3 beneath it with
. GitP/A/Session/26/10/06-Tue is an example. - Write chapter headings as plain timestamps and titles, such as
- ### 14:42 Adapters and input modes. Keep one full recording on the episode; the published player turns timestamps into seek controls. Separate chapter recordings and waveform embeds are unnecessary. - Follow Logseq/Entity/Asset for the recording's asset page. The asset page holds file metadata and the audio link; the episode also embeds the MP3 URL directly.
Check preset research
- Every saved preset linked from the session needs upstream notes covering saved export provenance, oscillator controls, filter, envelopes, LFO, modulation matrix and performance context. Include cited evidence and supported units, distinguishing the archived or factory baseline from observed session edits. Logseq/Entity/Preset/Synth/Microfreak describes the evidence limits.
- Run
mise run gitpa:preset:audit --episode 'GitP/A/Session/YY/MM/DD-Day' --check from this garden before publication. Resolve missing pages, stubs and unresearched sections; keep unsupported values and unavailable evidence explicit. The audit checks recorded coverage rather than proving every interpretation. - Use
mise run gitpa:preset:audit --all --json > /tmp/gitp-preset-audit.json for an advisory historical report. It does not gate the whole site's deployment.
Mirror into Gitpa
- Run Logseq/Entity/Proxy/Page/mise/Task/sync from the Gitpa checkout with this garden's checkout path,
--destination gitp-garden, the session page name, and --apply. It copies the session as a Logseq/Entity/Proxy/Page instance. Sync the episode’s audio and MIDI asset pages separately by their exact page names so Gitpa includes their file metadata. - Sync every preset referenced by the session separately by its exact page name after enriching its notes upstream. Keep the preset research in this source garden and re-sync later corrections.
- Edit the episode in this garden and sync again. Sync copies the episode body and
public:: selection from the source. Gitpa keeps tags::, podcast-guid::, and podcast-published-at::.
Prepare the publication PR
- On the source Logseq/Entity/Podcast/Episode/GitP/A page in logseq-encode-garden, set
public:: true to select publication or public:: false to exclude the episode from RSS and the website. Re-sync copies that selection to the publishing repository’s proxy. On the proxy, set podcast-guid:: and podcast-published-at::; these identify the RSS release and remain destination-owned. - Run
mise run rss:build, mise run rss:check, and mise run rss:test in Gitpa. RSS reads the proxy title, description, and recording link directly; enclosure size comes from the public media response. - Open a Gitpa PR with the episode title, a direct MP3 link, and the reason to listen: decide whether to publish this recording. The human merges the PR after one listen and a go decision.