OpenSocial – Developer’s View
Been head-down in the newly-announced OpenSocial API most of the weekend. (note to readers who don’t live, eat, breathe, and sleep web stuff: The OpenSocial API is technical stuff plus lots of pr released by Google with the support of major social networks like MySpace and LinkedIn aimed at diverting web developers attention from enhancing Facebook which has been on a tear lately even if you don’t know what it is and don’t care. This is like the Democratic primary candidates targeting Hillary. The fear that Facebook and Hillary have inspired is a testimonial to the leading positions they both occupy.)
Even wrote a trivial OpenSocial Hello World application that works and struggled with something more ambitious that still doesn’t work right. All part of making the kind of decision developers have to make about how much energy and resource to put into this new platform and whether the new platform changes the amount of energy we are already putting into old platforms.
To cut to the chase, FWD International (of which I’m CTO and in which I’m an investor) will do our kind of service enablement for OpenSocial networks and their users sooner rather than later BUT we will not take away from the resources already allocated to Facebook. Instead, we will spend less time than we would have on supporting networks which are neither Facebook nor OpenSocial compatible and presumably still end up with more supported networks. That’s not bad.
Nerd’s (bleary) eye view:
No question this thing was thrown together fast; someone gave the order “get it out the door NOW!” no matter what the planned release date had been. Even their own sample apps – except for the most trivial - don’t work with Internet Explorer (do work with FireFox) but no one from Google on the developer forum has (as of Sunday nite acknowledged this). Wasted a lot of time trying to debug my code (I know, I know, I should use FireFox – but most of my users don’t).
Once you stop trying to get IE to work, developing for this is pretty easy and sort of fun. The more I did, the more ideas I got. The basic structure is Google Gadgets (which have been around for a while). If you’ve been doing them, you have a real headstart.
Social networks like Orkut (owned by Google and VERY popular in Brazil) and all the networks spawned by Ning are called containers because they contain your app. With some very significant exceptions, apps written to run in one container will also run in others. Obviously a good thing since we developers can’t pick which social networks which haven’t even been developed yet are going to be hits but would like to have our apps available in all of them. Super network MySpace has promised support for OpenSource but there is no sign that they have done the technical stuff necessary to actually provide any support.
Orkut has a sandbox open where you can test your apps (once they let you in but that can take a day or so); it works at least some of the time because I’ve used it; but other developers complain about it intermittently failing. However, you can’t make your Orkut app available to anyone except other people with sandbox accounts so you won’t get rich there quickly. Ning, on the other hand, is actively supporting the promulgation of OpenSource apps according to the very well done blog on their website. Haven’t tried it yet so can’t vouch for it but will try tomorrow probably.
The documentation for how to make your website into a container for OpenSocial apps is MIA with no announced release date. Another sign of rushing this out the door, I think.
Big disadvantage compared to Facebook is that the authentication of who is actually using your app to access their data in the container (like info about their friends) appears to be so weak that it is unusable. This has just got to get fixed or the apps can’t really be social – they’ll just be games and eye candy. Good explanation of that problem in Miron’s Weblog.
Dave Winer, who knows more than a thing or two about social networking at both the nerd and user level, blogs “Standards devised by one tech company whose main purpose is to undermine another tech company, usually don't work.” Come to thing of it, I posted about last generations’ VIM v. MAPI API wars. Microsoft won that one against a pile on and it mattered. But, IMHO, OpenSocial is better than VIM.
Fred Wilson writes about ganging up on the leader.
Ning founder Marc Andreessen (who also had a lot to do with inventing the modern web browser) blogs about Ning’s support for OpenSocial.