Open·Parlamento isn’t a graph: it’s a universal tool
Direct answer: the network of colored nodes you see on Open·Parlamento is hype — a way in, not the product. The real asset is what you don’t see: the knowledge graph of relations between norms and the constellation of MCP connectors that queries Italian and European open data. It’s a universal tool — to study, map, cross-reference data and decisions — and, in time, a bridge toward a new kind of open data. The graphics are there to let people play; the value sits underneath.
TL;DR
- The graph is hype. The two views (Concepts and Relations) are there to make the thing approachable, not to explain it.
- The asset is the engine: a knowledge graph of relations + a network of MCP servers that pick the right source for every question.
- What actually sells: not the pixels, but the engine — via API or white-label — plus ~40,000 per-norm pages indexed.
- The long view: the connectors are a bridge over dated open-data infrastructure (think CKAN); in time, a self-managed, public and open system.
The graph you see is hype (and that’s fine)
Open·Parlamento has two visualizations: the nodes graph («Concepts») and the relations graph. They’re nice, the network moves, people play with them and say «cool». But let’s be honest: on their own they’re not understandable. Without a legend explaining what you’re looking at, you see a cloud of connected words and don’t know how to read it.
And that’s fine, because I didn’t build them for that. Those two graphs are the surface of a much bigger, heavier knowledge graph: the more nodes you add, the slower it gets to draw. I made the graphics only to make the thing tangible. The product is elsewhere — in fact, I almost shouldn’t even expose that stuff, because it’s my asset.
So what is the asset?
Two pieces working together.
- The knowledge graph of relations. Not a RAG answering at random: a network of norms (article by article, with ELI identifiers) connected by authoritative relations — amends, repeals, replaces, converts — extracted from Normattiva in Akoma Ntoso. ~6,445 articles, 86,672 relations at confidence 1.0. The relations are the value, not the text.
- The constellation of MCP connectors. A network of servers that brings law and public data into any AI client: RepublicMCP for the Chamber and the Senate (SPARQL),
open-parlamento-mcpfor IT/EU law, case law, parliamentary procedure, statistics, the NRRP, procurement. The agent picks the right source for the question: ask about healthcare, it queries the health source; ask about a decree, it goes to Normattiva.
| What the user sees | What the asset really is |
|---|---|
| A network of nodes that moves | A graph of citable relations between norms |
| «Cool» | An agent that picks the source and cites the original |
| A web page | An engine queryable via API |
| A demo | ~40,000 per-norm pages indexed on Google |
A universal tool, not a «law website»
Once the asset is the engine and not the page, the tool stops being «a site to read laws» and becomes universal: it serves to study a norm, map a sector, cross-reference data and decisions (what was decided vs what actually happened), tell the story of a legislative battle. The same engine, many different questions.
What actually sells (and what doesn’t)
I don’t sell the visualization. If someone pays me to make a powerful, crystal-clear graphic, I’ll do it — but it’s the last thing, not the product. What has value is the engine: the API to plug into your own systems, or a white-label integration for a newsroom or a law firm. Plus the long tail of SEO: for every norm there’s a sanitized, indexed page — around 40,000 — so a Google search takes you straight to the right article, with the official source alongside it.
A bridge toward a new kind of open data
And here’s the part that looks far ahead. Today the MCP connectors act as a bridge over open-data infrastructure that is dated and often badly run by the States using it. The symbolic case is CKAN, the software powering portals like dati.gov.it and data.europa.eu: it was born in the mid-2000s out of the Open Knowledge Foundation (repo on GitHub) and many keep it in terrible shape — endpoints answering empty, frozen catalogs, broken certificates (I wrote about it here).
The bet: once I’ve absorbed the useful data and the connectors make all the necessary calls, that fragile layer can be replaced — with a self-managed, public and open system, native to AI, that maps information intelligently and automatically for search, organization, indexing and legislative support. Not «another portal»: a new generation of open data. The bridge is there to cross; then you build the new shore.
FAQ
Why do you say the graph is «just hype»?
Because graph visualizations, on their own, explain nothing to whoever’s looking: they’re there to make the project approachable and «playable». The real value is the underlying knowledge graph of relations and the network of connectors that feeds it — what makes reliable, source-cited answers possible.
What does it mean that the asset is «an engine»?
That the product isn’t the web page but a queryable system: a knowledge graph + agents that pick the right source and cite the original. You use it via API or white-label, inside your own products, not by staring at a dashboard.
Do you really want to «replace» CKAN?
In time, yes: once the connectors cover the useful sources, the goal is a self-managed, public and open open-data layer, designed to be queried by AI agents — more reliable than the legacy portals that today often answer empty.
The product is the engine, not the pixels. See how Open·Parlamento works, read how you surface the inconsistencies and gaps in the law and how far it can go by cross-referencing all the open data, or let’s talk.
Related articles
- Understand a country by cross-referencing its open data, with AI agentsA statute is an idea; the data tell you what it actually became. An agentic system that cross-references every public database — law, cadastre, statistics, archives — to understand reality by analysing the data, not the opinions.
- Building at AI speed: the real bottleneck is decidingPersonal thoughts after a day spent building — articles, a manifesto, a page, even a team of automated fixes — at double speed. The lesson: producing isn’t the problem anymore. Deciding is.
- Why I Use Git as a Litmus Test When I Judge a TeamVersioning discipline is still the fastest way to tell whether a dev team is serious. Here is why — and where Git actually comes from.