I’ve been drilling OpenSocial for most of the year, talking about why our platforms (Mix and Connect) will be containers, explaining how it’s good for users and even better for developers.
Now, I have a couple examples that should stick.
Ed commented on my last mockups post:
OpenSocial is now officially my new latest favourite thing!
1) When’s Connect going to roll out it’s OpenSocial support?
2) If I want to leverage an existing app into a Connect OpenSocial app (say; I want to embed OraTweet into a Connect app) then how will authentication work, OAuth?
Thanks for getting me all excited so early in the week with this magical geekery.
As you may know already, we’ve had OpenSocial 0.7, the Shindig reference implementation, running for several months on Connect, but it’s not all gussied up and ready for primetime. Rich is working on the new UI, and Anthony recently upgraded to OpenSocial 0.8 to take advantage of new stuff.
Even so, we have working environments that people can use to build apps, and we have a few that Anthony and Jesse, who was on loan from APAC to help with OpenSocial, have built including the obligatory “hello world” app, a FriendFeed app and a horoscope app.
So, we sent Ed the information on how to access the environments late last night. He had already messed around with the sandboxes provided by iGoogle and orkut. So, he was already savvy.
This morning, I get an email from him that he’s already got a “Very very rough first test of OraTweet integration” working. So, between 11:22 PM and 7:57 AM, he whipped up an OraTweet integration. That’s amazing to me.
Ed is a rockstar, and the system works.
Rich has been saying all along that OpenSocial was the way to go for us, and he’s right as rain. We can’t possibly scale to meet the needs of our users. We are four. Mix has more than 27,000 users, and Connect probably has 37,000 or so. Sure these are daily users, but even so, you get an idea of the problem.
OpenSocial allows us to meet the needs of our users. I had an ephipany this week: the app marketplace.
A Mix user asked if we had email topic capability, to send an idea or question to someone who could help or would be interested. I like the idea, and it would help with engagement and adoption. Problem is, Rich and Anthony are at capacity working on higher priority features, and so is ENTP.
Enter the app marketplace.
OpenSocial plus a marketplace connects consumers (users who want features) with producers (developers who want to build apps).
Consumers: Want a feature added to Mix/Connect, but it’s not on our radar? Add it to the app marketplace and see how many other people want it.
Producers: Want to build some OpenSocial apps, but don’t know where to start? Pick up an app from the marketplace and work directly with the users who want it.
So, implementing OpenSocial allows us to expand our feature coverage and leverages the hordes of rock star developers, like Ed, who are looking for the next cool challenge. It’s OpenLab 2.0.
Incidentally, while we’re talking OpenSocial, congrats to Friendster for going live on OpenSocial 0.7 this week. App developers out there have another place their apps will run and access to 75 million Friendster users.
I guess Dossy Shiobara, saw a referral from OraTweet in his logs. Anyway, short story long, I guess they collectively figured out that OraTweet was Noel’s baby, not Orat Weet, some person who worked for Oracle.
While we’re on the topic of OraTweet, there’s a skunkworks project going on internally to use OraTweet for productivity updates and such inside the firewall. Thanks to Meg for the tip. It’s good to see Noel get some traction for his side project.