Tracking changes in Discord Canary updatesCurrent build hash: 60a5604b8b9cab4219b3b0d5c1141a70a9839f07
60a5604b8b9cab4219b3b0d5c1141a70a9839f07
This experiment reduces the space around stickers in the sticker picker. Its entry was removed in this build, while the inspected picker code uses the smaller padding directly.
This is a spacing adjustment in the sticker picker. Removing the experiment did not remove the smaller padding from the inspected code, which still supplies 1px directly.
Records the removal of an experiment comparing video players for expanded attachments.
The experiment could change playback controls and presentation when opening an attachment. It did not establish a measured quality or reliability improvement.
Records the removal of an experiment that preselected a plan during Nitro gift checkout.
The earlier flow could shorten gift checkout while limiting the ability to switch plans before paying.
Records the removal of an experiment for formatted Markdown attachment previews.
The experiment aimed to make README and note files easier to read inside chat without downloading them.
Records the removal of an experiment connecting app decisions to server AutoMod rules.
The experiment connected app-defined decisions to AutoMod. This record does not establish an app's exact moderation criteria or the data it handles.
Records the removal of an experiment for choosing among dismissible notices.
The change concerned which notice appears first when several compete for attention. Actual changes in exposure frequency or selection outcomes have not been measured.
The experiment record for private-profile guidance to an additional group was removed.
This preserves the earlier evidence about profile-visibility guidance; the current screen state needs separate confirmation.