2,000 Public Datasets Opening Up
The state will open two thousand datasets to everyone. The data target in Türkiye’s AI Action Plan reads like this: at least 2,000 public datasets published in the National Data Library in machine-readable format with standard metadata and open licences. At least 80 percent are to be updated on schedule. At least 100 will be opened as high-value data products through APIs.
Open data is not a new idea. Two things here are new. Quality will be measured. And the rules of sharing will be written down.
What the Number Says
BU BÖLÜMÜN ÖZETİ
- Data comes in two layers
- Every institution gets a “data product owner”
- Quality scores will be published
Three structures sit behind the two thousand.
Data comes in two layers
The first layer is open to everyone: public-interest datasets, anonymised and licensed. The second is controlled: data in sensitive fields such as health, defence, finance and public security opens only to authorised actors, through secure data spaces. Access rules and liability boundaries are defined separately for each layer.
Every institution gets a “data product owner”
This is one of the plan’s quietest but most functional clauses. A named person in each public institution will be responsible for keeping the data inventory current, deciding which sets are shareable, tracking quality and coordinating access requests. In other words, someone owns the data. Most data problems in any organisation begin with the absence of that someone.
Quality scores will be published
Datasets will be tracked by quality, not just by count. The criteria are named: machine readability, metadata completeness, currency, API accessibility, licence clarity, anonymisation compliance, user feedback. For high-value sets this score will be reported publicly.
What the Headline Misses
BU BÖLÜMÜN ÖZETİ
- Companies will be able to share data too
- Data providers get incentives
- A “data factory” service is planned
Three details concern businesses directly.
Companies will be able to share data too
The Data Spaces programme is not only for the public sector. Companies will also be able to share data in controlled fashion, on a voluntary and mutual-benefit basis. Common rules, secure working rooms, standard contracts and legal safeguards are being built. Your data can move from a closed safe to an instrument of exchange.
Data providers get incentives
Several incentives are listed for organisations that open their data: tax and R&D deductions, compute and data credits, and a formal trusted-provider status. For a business that already keeps orderly records, that can become a new revenue line.
A “data factory” service is planned
A shared service will handle preparation, cleaning, standardisation and conversion into machine-readable formats. This is where most businesses get stuck. The methods and standards built on the public side will set an example for the private side too. Privacy-preserving methods such as anonymisation and synthetic data generation are also on the list, so data can be useful without being handed over.
What a Business Should Do
BU BÖLÜMÜN ÖZETİ
- Cost: half a day for the inventory, months for the order
- First step: write down what you hold
- Next step: scan the library for your own work
Data work is long-term. Its beginning is short.
Cost: half a day for the inventory, months for the order
Drawing a data inventory takes half a day. Keeping data orderly takes continuity. Do not confuse the two. The inventory finishes within a week; the order is a habit that spreads over months. The good news: without an inventory you cannot use any public data opportunity, and with one, every door opens.
First step: write down what you hold
One table. Which data, stored where, in whose name, updated how often, accessible by whom. Most businesses are surprised the first time they draw it. Forgotten records, files sitting with one person, lists nobody has updated in years. The table is also page one of your data title.
Next step: scan the library for your own work
When the library opens, look for sets relevant to your sector. Market analysis, location choice, pricing, demand forecasting. All of these get stronger with public data. Free data is often more current than paid research. The scanning method sits in a separate piece.
Two thousand datasets are opening. You cannot use anyone else’s data before ordering your own.