The Buzzword That Ate the Smart City
If you’ve attended any smart city conference in the past two years, you’ve probably heard the word data space more times than the word citizen. Every presentation seems to promise one: “We are building a to enable collaboration, trust, interoperability, and… [insert your favorite buzzword here, or find a Gartner Magic Quadrant about players in this domain].”
When I went to the FIWARE Global Summit in Marrakesh, Morocco a few months ago, one of the FIWARE key people — thank you, Jason — literally grabbed me and said:
“You need to invest some time into data spaces. This is good. This is hot.”
I’ve been in this industry long enough to hear a new hot buzzword every year. My first reaction? “Nah.” But I listened.
The lecture itself was professional, knowledgeable, and convincing enough to make me write a mental note: research this thing later.
Convinced? Not yet. At that moment, data spaces still sounded like the next fashionable term that everyone would forget in a year.
Fast forward a bit, and here we are — still (for some parts even more) skeptical, but much deeper into the rabbit hole.
However, ask the same people what a data space actually is — and suddenly things get awkward. You’ll hear definitions that sound something like “a database in the cloud”, or “a web portal where you can download data”, or “some European regulation thing with Gaia-X”. Or, I really like this, "Open Data on the CKAN server". None of these are wrong, but none of them are the full picture either. Maybe in a galaxy far, far away...
At SenLab, we decided to cut through the fog — not because we like buzzwords, after all our unofficial motto is: "No bullshit!", but because we’re stubborn. And because, like most of you, we got tired of nodding politely while someone explained something that didn’t actually exist in practice.
So, What a Data space IS?
Let’s start with a clean, almost poetic definition:
A data space is a shared digital infrastructure that enables multiple organizations to exchange data, under clearly defined rules, with technical interoperability and legal trust built-in.
Sounds elegant, doesn’t it? In plain language:
- It’s not a database. Everyone keeps their data where it is.
- It’s not just an API. It’s a system that manages who can access what, under which conditions.
- It’s not open data. You don’t just publish your CSVs; you control who gets them, when, and why.
If you like metaphors (my team currently works on the pool automation project):
- A data lake is like a swimming pool — you dump everything in, and everyone splashes around.
- A data space is like a set of private pools connected by rules, doors, and lifeguards who check your membership card.
That’s the theory. Now, the reality…
Why Smart Cities Pretend to Have One
Here’s the uncomfortable truth: most so-called “city data spaces” today are just open data portals with better marketing. A CKAN instance, some APIs, a nice dashboard — and voilà, a data space! Except it isn’t.
- No identity management.
- No usage policies enforced.
- No contractual framework.
- No federation with anyone else.
On the other side, it does look like innovation, and it makes for a good press release. Cities love to say, “We have a data space,” because it signals modernity, sovereignty, and European funding alignment. The irony? Many of them wouldn’t notice if you swapped their “data space” with a Dropbox folder.
For the smart cities, it is a good start, so we have prepared our own "city data space" - one needs to start somewhere, so a CKAN instance, some APIs, a nice dashboard. Nice, we have a data space. HERE it is, just a first draft.
Where FIWARE and Gaia-X Come In
The EU didn’t invent data spaces just to make consultants rich. The idea is rooted in data sovereignty: making sure Europeans don’t hand over all their smart city data to Silicon Valley or Shenzhen.
- FIWARE offers the technical building blocks (context brokers, APIs, connectors).
- Gaia-X adds the governance and trust framework (rules, labels, identity).
Together, they should form the Lego kit for building real data spaces. The catch? Half the Lego bricks are missing, the instructions are outdated, and some of the pieces don’t fit together.
This is where most projects stall: great architecture slides, few working deployments.
Why Bother Then?
Because the idea of a data space still makes sense. Imagine:
- City A shares mobility data with City B — without giving away control.
- Utility providers share consumption data with municipal planners — but only for agreed purposes.
- Startups build services on top — with permission, not piracy.
It’s a network of trust, not a free-for-all. And in a world of rising costs, energy crises, and digital competition, trust-based collaboration is more than a buzzword — it’s survival.
Our Take at SenLab
When we started digging into data spaces, we quickly realized two things:
- The theory is solid. We do need new models for sharing data responsibly.
- The practice is painful. Almost nothing works as documented.
That’s why we built FIWAREBox: a sandbox environment where we could test, break, and reassemble the pieces of a data space without waiting for polished documentation.
It’s not perfect (and we’ll happily admit that), but it’s real. We can show a client what a data space looks like today, not in ten years. And we can do it with honesty — acknowledging the potholes instead of painting over them.
The Journey Starts Here
To conclude, what a data space really is? It’s an ambitious, half-built bridge between data silos and data sovereignty. It’s also a moving target, constantly redefined by Brussels, standards bodies, and software vendors.
And most importantly: it’s the next battlefield for smart cities. Pretending to have one will not be enough. At some point, citizens, regulators, and businesses will demand the real thing.
When that happens, those who have experimented early — who got their hands dirty with FIWARE, Gaia-X, FIWAREBox and the messy reality — will be the ones who will actually deliver.
At SenLab, we plan to be one of them.
... to be continued