Our $10,000 US Schools bounty just completed. We assembled a cleaned, free database of US schools. Please use it and let us know what you think.

How did we do?

This bounty was a little strange because it turned out multiple relatively large public datasets of US schools already existed. The first bounty participant who uploaded those took the lion's share of the bounty. Other contestants spent the rest of the bounty cleaning up and standardizing the data. This isn't ideal but it proved our core of bounty participants are better at finding open data on the internet than we are at this point. We've attracted monsters!

What the above means for consumers is this dataset likely has the most accurate websites, addresses, and geo locations of any open schools database. Check out this diff for an example of the improvements.

The assembled database includes 137,589 schools broken down into the following distribution by category:

We had five bounty participants with the top contributor earned $7,854. Three of the four other contestants earned over $500.

Conclusion

Bounties continue to inspire us even when they don't go as planned. Bounties show off the power of a database designed for collaboration. Bounties are a fun, motivating way to learn scraping, databases, data cleaning and a host of other marketable skills. On the DoltHub side, we love to find and fix the issues bounty participants find with our tools.

If you'd like to participate in bounties, we are running a bounty to collect a database of US businesses hosted by a previous bounty winner. If you want to talk to us about bounties or Dolt, come join us on our Discord.