Reviving Praderas (Day 7) — Phase 4: SEO, social metadata, and date archive

Reviving Praderas (Day 7) — Phase 4: helping Google and humans find things without drama

After stabilizing navigation, content metadata, and series, the next roadmap block was Phase 4: SEO and discoverability without compromising Pico's flat-file simplicity.

This phase does not replace infrastructure-level domain decisions. Instead, it documents them and keeps site behavior coherent with canonical configuration in config/config.yml.

What we implemented

  1. Unified HTML metadata in themes/bootstrap-blog/page-meta.twig, included from index.twig, post.twig, and blog.twig:

    • consistent title pattern (... | Blog Praderas, with tags-page variant),
    • meta description and robots when present,
    • canonical link using current_page.url,
    • Open Graph (og:title, og:description, og:url, og:type, og:locale, og:site_name),
    • Twitter cards (twitter:card, twitter:title, twitter:description),
    • og:type as article for blog/* and website elsewhere.
  2. Better Spanish search behavior: configured PicoSearch.low_value_words with Spanish stopwords so multi-word queries are less dominated by filler terms.

  3. Date archive page: added content/archivo.md (Template: archive) and themes/bootstrap-blog/archive.twig, grouping blog/* entries by year and month (Spanish labels), plus sidebar link ("Ver archivo ->").

  4. Minimal discoverability copy: added Description to content/search.md so search page metadata is useful when indexed or shared.

Honest out-of-scope notes

  • Root domain vs blog subdomain is still a DNS/hosting decision. In code, canonical base_url remains https://blog.praderas.org.
  • Highlight snippets, heavy JSON-LD, and hreflang were intentionally deferred to future iterations.

Day 7 closing

Phase 4 left the site with stronger HTML signals, a human-friendly archive, and better Spanish query behavior while keeping Pico lightweight. The formal next step was Phase 5 (multilingual).