On Thu, Sep 5, 2013 at 11:43 AM, Chris Karlof <[email protected]> wrote:
> On Sep 5, 2013, at 10:00 AM, Eric Rescorla <[email protected]> wrote: > > On Thu, Sep 5, 2013 at 9:04 AM, Lloyd Hilaiel <[email protected]>wrote: > >> What are the tradeoffs of dictating protocol and implementing support at >> the os level, vs a JavaScript api and leaving the proto decision up to an >> updatable app (plus implementing or promoting an awesome js carddav is lib)? > > > I'm not sure what "at the OS level" means in this context. We ship > a contacts app. That app should (somehow) implement carddav. > > -Ekr > > > I think what Lloyd is referring to here is "do we need a general sync > framework that supported by the OS?" > > E.g., Android has one [1], but iOS doesn't. > > SyncAdapters help enable orderly, scheduled background syncing. I don't > know the capabilities of FXOS apps to do background tasks, but I think > having OS coordinated background syncing for apps would be desirable. > > I'm not sure we want this service to be a dependency of v1 contact syncing > (e.g., contacts app that speaks CardDav, only syncs when it's loaded), but > it would be useful in the near future. > If I'm reading this Google doc correctly, there are no restrictions on what the SyncAdapter can do. It's just a way for the application to get itself run in the background on some Sync schedule. Is that correct? No doubt it would be useful to offer such a facility to apps such as the contacts app, but it seems like it's orthogonal to what protocols we should be implementing and from which such protocols should be built into the OS. -Ekr -chris > > [1] > http://developer.android.com/training/sync-adapters/creating-sync-adapter.html > > > > >> Lloyd >> >> Eric Rescorla <[email protected]> wrote: >> >> On Thu, Sep 5, 2013 at 5:41 AM, Lloyd Hilaiel <[email protected]> >> wrote: >> >> > The difference is, users must have choice. Sure we can support the big >> > guys, but we should also make it so little guys can build ffx support >> in an >> > absolutely seamless manner and directly compete. >> > >> > This is an area where we can give the market a fighting chance to do one >> > thing, do it extremely well, and win the hearts of users away from large >> > silos. >> > >> > Why not contact sync by ffxos apps? (Give the user a choice, feature >> the >> > most popular) >> > >> >> I'm all in favor of giving users a choice, but I'm not sure how >> much there is to decide. >> >> There are standardized technologies for both calendar sync and >> contact sync (CalDAV and CardDAV) respectively. As far as >> I can tell, both Apple and Google support both. >> >> https://support.google.com/mail/answer/2753077?hl=en >> https://discussions.apple.com/thread/3407527?start=0&tstart=0 >> >> Assuming that's correct, if we just do CardDAV we should be >> mostly done >> >> -Ekr >> >> >> >> >> > >> > Mark Finkle <[email protected]> wrote: >> > >> > ----- Original Message ----- >> > >> > > On Sep 3, 2013, at 5:11 AM, Christophe Brocas < >> > [email protected]> >> > > wrote: >> > >> > > > Hello >> > > > >> > > > Sync has always been Firefox oriented and so it put the focus on >> > syncing >> > > > booksmarks/tabs/passwords informations. >> > > > >> > > > But with the move to Mobile with FirefoxOS, I think users will first >> > want >> > > > to >> > > > sync their contacts and calendar between their mobile/PC/tablet. >> > Because it >> > > > is >> > > > what first matter on a phone. >> > > > >> > >> > > We should probably provide our own contact sync service eventually, >> but >> > in >> > > the spirit of "not building another silo", what about syncing to other >> > > contact service providers (Google, Yahoo, Facebook, etc)? >> > >> > > Maybe we should support third-party contact sync before we build our >> own. >> > > When we launch support for contact sync, exactly 0 zero users will >> have >> > > pre-existing contacts in the Firefox OS contact syncing service, which >> > makes >> > > the value prop less compelling. However, I suspect at least some users >> > will >> > > have pre-existing contacts at these other providers, so supporting >> > contact >> > > sync with them is a compelling value prop. >> > >> > I can't help feeling a bit sad about this plan though. We use FirefoxOS >> to >> > give people a great entry platform to the mobile internet, then we >> > introduce them to Google or Facebook and proceed to get them locked into >> > those vendors. >> > >> > I can see the rationale of the plan: If you want Sync, then you must >> have >> > more than just your phone. If you have more than just a phone, you >> probably >> > already have your data locked into a vendor. >> > >> > Unless these people, like many others, mistakenly see Sync is a way to >> > backup their contacts and other data. >> > _______________________________________________ >> > Sync-dev mailing list >> > [email protected] >> > https://mail.mozilla.org/listinfo/sync-dev >> > >> > > >
_______________________________________________ Sync-dev mailing list [email protected] https://mail.mozilla.org/listinfo/sync-dev
