Almost 2.5 years ago, I wrote a longread (in Russian) about Uzbekistan’s open data portal—“What’s Wrong with Uzbekistan’s Open Data Portal?”. Recently, I decided to take another look at the portal: data.egov.uz. Unfortunately, not much has changed.

Yes, the number of datasets has grown—from 6,623 to 10,412. That sounds impressive. But here’s the reality:

➡️ In 2023, there were 2,823 single-row datasets. Today, there are 5,207 of them—50% of the entire portal.
➡️ Only 114 datasets contain more than 1,000 records—that's just over 1% of all published datasets.
➡️ The total uncompressed volume of the portal’s data (in JSON) is around 426 MB (compared to 284 MB last year).

Why create thousands of datasets that each contain just a single row? The answer is simple: to boost quantity, not quality. For real data users, such datasets are virtually useless.

Has anything meaningfully changed with open data in Uzbekistan? Sadly, no. The number of datasets is not a true indicator of openness when the majority are artificially fragmented like this.

At Dateno (dateno.io), we’ve chosen not to index this portal—at least for now. It uses non-standard software, isn’t easily crawlable, and more than half of its datasets lack meaningful value.

👉 What do you think?
Do such open data portals provide any real value?
Is it worth talking about them at all?

#OpenData #Transparency #DataQuality #Dateno #DigitalGovernance #DataPortals