Notes from today's migration meeting. In attendance: Asa, Deb, Karen, Crystal, Tauni, mfinkle, Lloyd, Warner, ckarlof, myself, and I think I saw Vlad in the room.
Strategy: * We want to encourage existing Sync users to migrate to a Firefox Account, without harming the experience of users who aren't ready to move. * We're aiming for an overlap period between Sync and Sync.next. This will reduce the risk of users having partitions in their devices, reduce problems with upgrade/downgrades and staged releases, etc. The exact duration of this overlap period will depend on several factors: complexity, cost, time to market, and market parity. * Delayed upsell. We want to sell Sync.next and Firefox Account once users are ready to upgrade -- that is, they have the right version of Firefox on each of their devices. * Forcing function. As we ramp up, we'll remove the ability to set up an old Sync account from each product. Eventually we will force users off the old service and decommission it. See overlap period. Migration mechanism: * Detect specific states. * Non-email Sync account name. No migration possible. * Single device. Encourage migration! No partitioning possible! * Self-hosted. Offer to migrate to Mozilla's system if there's no locked pref to indicate that that's not desirable, provide docs to support continued self-hosting, inform the user of decommissioning plans so they can make an informed choice. * Offer to set up a new Firefox Account using their same email address. Enter the normal account setup flow, which will take care of email verification. Don't reuse the password -- it's gone over the wire, many users won't remember it, and it's probably neither strong nor memorable. * Preserve data syncing preferences, also allowing user to make new decisions at this point. * Write decommissioning sentinel into the old Sync account. (Bug 895526, Bug 895518.) * Disable or delete local Sync.old account. * Clean up local prefs for sanity. * Configure to intercept old account hooks e.g., Send Tab intents. Non-goals: * Migration of data. * Simultaneous syncing to Sync.old and Sync.next. Open issues and action items: * [lloyd] How long is the overlap period? Need to know some operational costs. * [rnewman, mfinkle] What's the client engineering burden to having an overlap period? * [rnewman] Send Tab isn't in the PICL IdP. We're trying to find out how important a feature this is for parity. * [product] We know we plan to disable the account setup UI during ramp-up. At what point do we want to disable *login*, for those poor suckers who reinstalled Windows and want to get their data back from Sync? * [product] When should we kill the Sync promo on Android? Depends on how valuable it is and how long our roadmap is for Sync.next. * [?] Add-ons that provide sync engines will be screwed. Original etherpad for notes: https://services.etherpad.mozilla.org/sync-migration _______________________________________________ Sync-dev mailing list [email protected] https://mail.mozilla.org/listinfo/sync-dev
