[lizen_one]

That is a very interesting idea. As far as I understand, you generate the information purely read only.

Column statistics: Sample data randomly and calculate estimator from that.

Data Lineage/usage: From the queries (and 'similar' stuff like materialized views. It seems like that currenltly have only lineage on table level, i.e. which table depends on which tables. Do you plan to implement data lineage on column level, i.e. which column depends on which column? This sounds challenging and very interesting.

I tried to watch the demo video on your website, but it was too small and maximizing the video is disabled (Firefox). So I had to click through to youtube. I'm not sure if this is intended.

[zurfer]

Thanks for your thoughtful comment and pointing out the video problem! It should be fixed now.

Column Level Usage is something we are looking into. We also have some ideas. It seems like a solvable problem but as you said challenging. :)