2009/5/14 Damien Elmes <[email protected]>: > >>> I've also contacted Damien from Anki, but he does not seem to be too >>> interested... >>> (http://groups.google.com/group/ankisrs/browse_thread/thread/410997e06be22302). >>> >> We can start from looking at Anki's implementation. If it's good >> enough and only requires a bit of changes then >> we can use it and Damien might be more interested in joining us. >> Anyway it makes sense to look at working implementation first in >> order to not reinvent the wheel. > > The bulk of it was written over a year ago, so I'm sure there's plenty > of room for improvement. One thing I need to tackle is the initial > sync or a full update - it performs quite well for daily syncs, but a > full sync on a deck of 30,000 cards takes up a large amount of memory, > unpacking the JSON into a python object tree. I will probably address > this by sending across the compressed deck verbatim, instead of > bundling it up into a sync message. Of course, this approach would > never be compatible with other implementations. An alternative would > be an incremental JSON parser, but that may be complicated to > implement. > I like the idea of using JSON. What do you think about using one card as minimal amount of sync data? This would allow us to synchronize in a lazy way, one card at a time. On top of this we can implement grouping cards into blocks and/or even compressing them(cards or blocks) to save bandwidth, which would allow us to quickly perform initial sync on big decks.
-- BR, Ed --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "mnemosyne-proj-users" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/mnemosyne-proj-users?hl=en -~----------~----~----~----~------~----~------~--~---
