@darius It would be helpful to non-devs to know how rigorously safety was considered and to what extent it's mitigated before implementation. A few thoughts

• Why is it important to more easily know how the data is shaped?

• Who benefits, and how could/might the resulting info be misused, even if 'anonymized'?

• How much trust is the user expected to have (could PII scraped data be leaked or hacked before it's tossed, for example)?

• What are the unintended consequences of enabling this?