[Cosmo-dev] simplifying eimml

[Phillip J. Eby]

At 12:11 PM 12/1/2006 -0800, Brian Moseley wrote:
>On 12/1/06, Phillip J. Eby <pje at telecommunity.com> wrote:
>
>>That is, even if Cosmo and Chandler were sending true diffs, it's still
>>necessary for Chandler to match the incoming diffs against the previous
>>baseline record set - which means the baseline record set has to be either
>>in-memory or indexed by record keys.  Using the item "envelopes" allows us
>>to load only a subset of the baseline records at any one point in time.
>
>then you're in favor of grouping records into "item states" in the
>xml? if so, i'm fine with it too.

Yeah, we pretty much have to do this until/unless we can use relational 
storage for the sync baseline.  (Well, we could make it work now anyway, 
but the performance would likely be abominable.)