Sunsetting the beta channel #964
Labels
effort/hours
Estimated to take one or several hours
exp/intermediate
Prior experience is likely helpful
kind/maintenance
Work required to avoid breaking changes or harm to project's status quo
P1
High: Likely tackled by core team if no one steps up
TLDR: switch to the Stable channel ASAP
Rationale
v2.17.1 illustrated problems with Chrome Web Store when publishing bugfix releases.
Below is a plan to improve current state of things.
Current setup
We have a dedicated Beta channel (which has separate ID) and we published a fix release there first.
After a day, we decided the fix is important enough that we also submitted it to the Stable channel.
Stable channel got reviewed within 1-2h, while Beta was still pending review, which defeats its purpose as a platform for fast-tracked testing. Based on our observations during 2020, the number of users impacts the priority in the review queue (~40k stable vs ~200 beta). With the user count on the Stable channel, reviews do not seem to ne an issue, and we should leverage that.
Proposed change
Deprecate Beta channel.
Chrome Web Store supports staged rollouts, where published can define % of users to get autoupdate.
This enables us to have a single publishing channel (current Stable) and initially publish to 1% of userbase (~400 atm), then do all the QA we need. It also enables us to bump it to 10%, 25% 50% 75% each following day and monitor github/forums, increasing our confidence and making big changes much less stressful.
Added value
TODOs
ipfs block get|rm
)The text was updated successfully, but these errors were encountered: