Remix.run Logo
peheje 2 hours ago

the data is dominated by big unique TEXT columns, unsure how that can much compress better when grouped - but would be interesting to know