4 ms·

Generating websites with SPARQL and Snowman, part 1

50 points by davidrupp 4y ago

kinghuang 4y ago
What great timing! I was just thinking about wanting something to generate a static website for graph data I've gathered in AWS Neptune. I'm definitely going to play around with this when I have some time.
alexchamberlain 4y ago
Nice to see some RDF projects evolving outside the Java (RDF4J) ecosystem- RDF is vastly undervalued in industry IMHO.
- melony 4y ago
  It is undervalued because it only offer value to third party search engines and web scrapers. The site owner does not derive any returns from implementing RDF which is often a non-trivial amount of work. The biggest search engine on the planet has enough scale that RDF offers it negligible value. Academia won't be able to convince anybody soon until they can show improvements in $.
  - p_l 4y ago
    BBC thought differently - they don't expose it externally, but BBC website uses semantic Web technologies on the backend to automatically link relevant information including making it easier to author new content.
    - mro_name 4y ago
      I wonder what happened to RDF at BBC. There must have been an earthquake around 2014? I was deeply impressed and inspired to my toy-project radio recorder. Check $ rapper -i grddl -o turtle 'http://rec.mro.name/stations/b2/2022/05/23/1005.xml http://rec.mro.name/stations/b2/2022/05/23/1005.xml'
  - FrankyHollywood 4y ago
    Not completely agree, I work at a company who makes money with semantic web consulting. It is really powerful technology, which we use to index and unify customer data, but also for dynamic React front-end generation. I was a bit hesitant to start working at this company, because of the many negative notes here on HN (https://hn.algolia.com/?q=why+the+semantic+web https://hn.algolia.com/?q=why+the+semantic+web) This contrasts with what I experience, we have a growing customer base, both in healthcare and finance. Just do a LinkedIn job search with a keywords like 'sparql' or 'semantic', you'll see many jobs.
- jansommer 4y ago
  Completely agree. I'm hoping to one day see Jena [0] compiled to a native image [1]. Having a persistent triple store with transactions, and an inference api in owl/rdfs/shacl with a prolog-like "logic programming engine", running in-process like SQLite, would be awesome. [0] https://jena.apache.org/ https://jena.apache.org/ [1] https://www.graalvm.org/22.0/reference-manual/native-image/ https://www.graalvm.org/22.0/reference-manual/native-image/
abbe98 4y ago
Cool to see a project(Snowman[0]) I'm the creator of mentioned here on HN. I created Snowman to allow one to use RDF/SPARQL even in the user-facing parts of ones stack(and to learn Go). [0]: https://github.com/glaciers-in-archives/snowman/ https://github.com/glaciers-in-archives/snowman/
- westurner 4y ago
  E.g. Hugo-academic has Microdata RDF (Go) Templates; and supports Markdown, Jupiter, LaTeX: https://github.com/wowchemy/starter-hugo-academic https://github.com/wowchemy/starter-hugo-academic "Build pages from data source" https://github.com/gohugoio/hugo/issues/5074 https://github.com/gohugoio/hugo/issues/5074

50 points by davidrupp 4y ago

What great timing! I was just thinking about wanting something to generate a static website for graph data I've gathered in AWS Neptune. I'm definitely going to play around with this when I have some time.

alexchamberlain 4y ago

Nice to see some RDF projects evolving outside the Java (RDF4J) ecosystem- RDF is vastly undervalued in industry IMHO.

melony 4y ago
It is undervalued because it only offer value to third party search engines and web scrapers. The site owner does not derive any returns from implementing RDF which is often a non-trivial amount of work. The biggest search engine on the planet has enough scale that RDF offers it negligible value. Academia won't be able to convince anybody soon until they can show improvements in $.
- p_l 4y ago
  BBC thought differently - they don't expose it externally, but BBC website uses semantic Web technologies on the backend to automatically link relevant information including making it easier to author new content.
  - mro_name 4y ago
    I wonder what happened to RDF at BBC. There must have been an earthquake around 2014? I was deeply impressed and inspired to my toy-project radio recorder. Check $ rapper -i grddl -o turtle 'http://rec.mro.name/stations/b2/2022/05/23/1005.xml http://rec.mro.name/stations/b2/2022/05/23/1005.xml'
- FrankyHollywood 4y ago
  Not completely agree, I work at a company who makes money with semantic web consulting. It is really powerful technology, which we use to index and unify customer data, but also for dynamic React front-end generation. I was a bit hesitant to start working at this company, because of the many negative notes here on HN (https://hn.algolia.com/?q=why+the+semantic+web https://hn.algolia.com/?q=why+the+semantic+web) This contrasts with what I experience, we have a growing customer base, both in healthcare and finance. Just do a LinkedIn job search with a keywords like 'sparql' or 'semantic', you'll see many jobs.
jansommer 4y ago
Completely agree. I'm hoping to one day see Jena [0] compiled to a native image [1]. Having a persistent triple store with transactions, and an inference api in owl/rdfs/shacl with a prolog-like "logic programming engine", running in-process like SQLite, would be awesome. [0] https://jena.apache.org/ https://jena.apache.org/ [1] https://www.graalvm.org/22.0/reference-manual/native-image/ https://www.graalvm.org/22.0/reference-manual/native-image/

abbe98 4y ago

Cool to see a project(Snowman[0]) I'm the creator of mentioned here on HN. I created Snowman to allow one to use RDF/SPARQL even in the user-facing parts of ones stack(and to learn Go). [0]: https://github.com/glaciers-in-archives/snowman/ https://github.com/glaciers-in-archives/snowman/

westurner 4y ago
E.g. Hugo-academic has Microdata RDF (Go) Templates; and supports Markdown, Jupiter, LaTeX: https://github.com/wowchemy/starter-hugo-academic https://github.com/wowchemy/starter-hugo-academic "Build pages from data source" https://github.com/gohugoio/hugo/issues/5074 https://github.com/gohugoio/hugo/issues/5074