HIGHSTREET‑ME1

• ■ Running an ETL pipeline to turn unstructured HTML and wildly inconsistent civic data sources into a clean, typed schema. You'd think the council could manage a spreadsheet. Reader, they cannot.
• ■ Rendering a fully interactive High Street map in nothing but HTML and CSS — no canvas, no SVG library, no framework. Turns out that means building a relational data model from scratch just to plot shop positions. Who knew retail geography was basically a graph problem… wtf!!! i dont even have GCSE's
• ■ Hunting for a lower-latency tide feed. The nearest live gauge is downstream at Sheerness, and the APIs with real precision all want a paid subscription. Not doing that.
• ■ Closing coverage gaps in the P102 shop dataset one business at a time. If that's you, get in touch and become a data point.
• ■ Scouting new data sources to feed into the pipeline — more feeds, more signal.
• ■ Figuring out how to proxy and obfuscate the origin URL and IP for our funky little weather station, so scallywags and bots can't just wander in and start querying a Raspberry Pi directly — which, incidentally, we have yet to actually bolt to the roof.

← Back to P101

• ■ A live ticker-tape feed streaming real-time coffee prices across every shop on the High Street.
• ■ A gamified leaderboard — because nothing motivates quite like public, unnecessary competition.

← Back to P101