Re:

I've been thinking about this some more and I'm starting to think we should just build something like this out for all top tier partners and launch them. Initially I thought about it as a way to gain leverage to get folks to do the right thing, but now I actually just think it might be a better and easier way to get more people to sync their data and for us to be in control of our product experience for the most important social apps out there. Right now we're struggling to convince these companies to do the right integrations with us, but we have the power to just do them ourselves. I basically think we should build out a timeline store as part of our strategy and have these integrations be in there. We can how you the right apps to connect with based on which ones we already know you've authed or even shared data from recently. One other advantage of this is we can perform operations on the data to make it better and more useful on FB. For example, we could have a Twitter importer (which I've discussed with [redacted] and think they'd allow) which rewrites @names to links to user profiles. We could also download images from Pinterest and just display them inline if we wanted. I'm not 100% sure we'd end up doing that, but it's something we've discussed. If we don't need to provide a ton of distribution back to these apps to convince them to prioritize these features since we're building them ourselves, we actually have a lot more flexibility to do them well. Arguably, if you look at the incentives in the market, Twitter and perhaps Instagram (if they keep going) will never prioritize having the perfect FB integration since they're trying to build parallel networks. If we want them to do this well, doing it ourselves makes sense and it might be the only way. Given the amount of time we spend discussing them and their strategic importance to us, it seems very reasonable for us to take this approach to have the integration we want with them. For other companies beyond the top 5-10, we could just do an oauth implementation for the timeline store as well as supporting everything in Open Graph like we're moving towards doing today. I don't really understand why you want to build this as an employee-only tool. What does that get us rather than building it as consumer product and planning to launch it?