https://github.com/mui/mui-x/issues/5913
The current docs rating is not working:
It’s a UX problem. People don’t see the action (e.g. @Sycamore didn't saw it until @Olivier Tassinari showed it, it was reported by other community members too).
The solution here would be to move the rating out of the prev/next page.
Operations problem. The data we collect is very hard to use. It’s using DynamoDB which can barely be queried for returning stats. https://us-east-1.console.aws.amazon.com/dynamodbv2/home?region=us-east-1#item-explorer?initialTagKey=&table=feedback-prod
dateTime
field is a string and stored as a non-ISO format, so it can't be sorted.Scalability. The feature should be isolated inside the mono-repository. No configuration should be needed for MUI X, Toolpad, or Design kits to benefit from it.
?
Algolia https://www.algolia.com/doc/tools/crawler/apis/configuration/ignore-no-follow-to/#how-to-use
Google Cloud platform
Twilio — they’ve done a lot of research into this topic:
https://www.youtube.com/watch?v=-M7m7f5hcxU
Their solution
They use a 5-star rating widget at the bottom of all docs pages, that they wire to a Slack channel. Each time a community member leaves feedback, it creates a notification.
e.g. https://www.twilio.com/docs/flex/developer/plugins/migration-guide#material-ui
The extra time pressure this puts on the team is solved by a weekly duty rotation with the team members.
What I learned
Do not expect to get insights from metrics. Tracking a metric does not directly lead to action. It could be a way to prioritize: demo X has the lowest ranking, let takes some time to improve it. But that’s in a perfect world.
Better to have testimonials, from users complaining to know what they expected and did not find.
Notice that doc modification has a low technical difficulty and low risk. So improvement can be implemented on a daily basis
https://baseweb.design/components/checkbox/
powered by https://delighted.com/web
Figma