There are many different ways to do this. And it simply doesn't matter which one we chose. Losing a usage count is irrelevant. As long we are reasonably counting, that's perfectly fine. Complexity is our biggest enemy here. Precision is not a design goal. Fast delivery of a solution that reasonably approximates consistency is.
Andreas Sent from Mobile. On Jul 26, 2013, at 17:22, Richard Newman <[email protected]> wrote: >> Sounds like you're refining that theory. That conflict resolution is type >> specific. >> >> For passwords - most recent change wins. >> >> For bookmarks, tabs and history - favor duplication over deletion. > > Yes, and furthermore: > > I think the instances in which we have to fall back to actual copying or > overwriting should be slim: genuine conflicts of core fields. Changing a > password in two places at the same time, altering a bookmark title in two > places at the same time. > > Most 'conflicts' -- *records* modified in two places -- will actually be > either modifications of two different attributes (I rename a bookmark on my > phone and move it into a different folder on my desktop), or colliding > modifications of fields that are reconcilable (two devices both increment a > usage count). The act of reconciling is one of extracting and comparing > sequences of changes for congruence, and I contend that most sequences of > real-world changes will be congruent. > _______________________________________________ Sync-dev mailing list [email protected] https://mail.mozilla.org/listinfo/sync-dev
