OpenSocial's Failed Promise: Only 0.7% of Apps Have Crossed Networks (Updated)
When the Google-led OpenSocial campaign launched in October 2007 it aimed to give developers a common environment that application publishers could publish widgets to with one set of code, deployable across Google sites, MySpace, Hi5 and numerous other social networks.
A directory of OpenSocial Apps launched today and the reality is even further from that goal than we expected. Out of 12,456 apps listed, only 83 are running on two or more "containers." That's 0.7% or one out of every 1500. Update: See this reply below from Google's Kevin Marks. Marks says that cross-network presence was counted manually and is actually larger than it appears in the directory.
OpenSocial was intended to help everyone else keep up with the huge success of the Facebook platform. If there's one clear market leader, everyone else has a common interest in creating a standard that will help scale the market opportunity on their platforms vs. what Facebook can offer. It hasn't worked out that way, though.
Outside observers hoped that OpenSocial would allow for user data to be transmitted from one site to another. People thought this was the coming of Data Portability, though OpenSocial advocates quickly said that wasn't the intent. It was just for apps to port, not user data. There is a lot of innovation going on in OpenSocial - it's a shame the platform isn't better appreciated.
Below: I discuss OpenSocial in March of last year on G4TV, forced to break the host's heart about data portability!
Why hasn't cross-network development happened though? There are a few theories. The most common is that though there is a common bed of code across all the different social network containers, each of them is also tweaked just enough that it's not that easy to "write once." When that became apparent, OpenSocial advocates started saying that the standard still made it a lot easier to develop for multiple networks. So if not "write once, deploy everywhere" then perhaps it was "write once and then take a lot less time to write for elsewhere than you'd have to otherwise."
Clearly developers haven't taken advantage of that opportunity.
Google's Kevin Marks, one of the leading public faces for OpenSocial, told us today that there are serious cultural differences between the networks and that this could help explain why there hasn't been more cross pollination. One look through the directory makes it clear though that while the countless "hot bikini girl" apps may not translate from MySpace to LinkedIn well, all the other networks have their own isolated versions of many of the same insipid apps.
The most viable explanation could be that Facebook is in fact the only game in town for the most sophisticated developers. That's a real shame, because it's never good for innovation for there to be only one game in town.
Share
Facebook
Comments
Subscribe to comments for this post OR Subscribe to comments for all ReadWriteWeb posts
Hi Marshall, I spoke to the developer behind this directory, and he said that at the moment the apps that are labelled as multi-container have been correlated by hand, so the number of multi-container apps has certainly been undercounted. As the FAQ says:
http://wiki.opensocial.org/index.php?title=OpenSocial_Directory
The goal of this directory is to help app developers and container developers to find each other and enable more applications to move between containers, to help address the cultural and linguistic barriers between porting applications.
As for your 'Data Portability' point, with OpenSocial REST APIs, and their compatibility with OAuth and PortableContacts, we are seeing a sea change in how people can move their data across the web - in particular the OpenID + OAuth +PoCo hybrid login has been adopted by Friendfeed and Facebook recently, joining Plaxo in providing a much smoother sign-up process for new users and friend-finding for existing users.
In addition, Google Friend Connect is building on these same open standards to enable any web site to become a social site and support delegated login, friend and activity connections, and embedding gadgets on the page.
Thanks for the comment Kevin. Will update post with a note referencing your reply .
I think the problem is people just don't really care if they run an app inside a social network or on an external domain. So rather than work around all the differences between each OpenSocial container developers are simply building stand-alone websites to interface with Google Friend Connect and Facebook Connect. I was really excited about OpenSocial when it was introduced, but now my excitement has diminished and I'd almost rather open up a plain vanilla OAuth API than an OpenSocial container. I don't need apps running within the confines of my website, they can run anywhere.
Marshall,
There is another benefit of opensocial apps that you have to consider. Say, I learn the opensocial way of building apps.
One benefit, is of course, to be able to port the same app to multiple platforms/containers, which as you pointed is not happening a lot.
The other benefit (although comparatively smaller) is that I can use my opensocial skills that I just learned for my one-container app to make another one-container app in the future for any of the other opensocial-participating platforms. So, my skills become portable.
This second benefit I am sure helps in the way of increasing the total number of apps built via Opensocial for their target platforms.
There are many shades to open social implementations. Twitter's OAuth requires the third-party applications to sit on their server. We do not want the over-capacity complication and decide on the say-hi-and-bye method to let Twitter users sign into isayusay.net as guests, create live social mail in a snap, and post them to Twitter with custom messages.
Opensocial still has great promise for building social networking apps, but it just can't seem to reach that critical mass.
I like your point about the cultural differences between networks and how apps that appeal to members of one social network won't necessarily be as popular in another.
The best apps meet the needs of a specific market. Different networks attract different "markets" of people so this will be a huge factor in what catches on and what doesn't, regardless of the technology.
@Scott:
You can still be excited about OpenSocial, and build apps targeted at any website via Google Friend Connect. There are 5 million 30-day active Google Friend Connect enabled sites and blogs out there.. and they all have the capability to host any OpenSocial application. However, keep in mind that the types of applications which work well embedded on a variety of websites are different than those which work well on traditional social networks. :)
Well I can simply explain why we haven't embraced it...
Because it does not support being present at the 'content creation page' along the editor.
As OpenSocial is currently structured it is more or less advanced widget with insight into the network, but not actually plugged into parts of the platform developers would like to be plugged in...
This is very similar to what Ryan Boyd says: "keep in mind that the types of applications which work well embedded on a variety of websites are different than those which work well on traditional social networks."
I hope this will change soon! :)
bye
Andraz Tori, Zemanta
@Ryan Boyd I'm more interested in OpenSocial from a container standpoint. I run a social network that focuses on developers, so it seems like a logical fit. What I'm considering now is using it more as an OAuth API interface to open up my data.
I wonder whether Google Wave will breath new life into OpenSocial? If the environment Google is building enables network site owners to build an app which runs inside Wave and enables the user to see activity relating to their account and contacts, then should Wave revolutionise e-mail, IM and socnet participation an OpenSocial involvement may become mandatory.
As a network owner I'll be watching this side of things closely.
Ian Hendry
CEO, WeCanDo.BIZ
http://www.wecando.biz
OpenSocial's Failed Promise: Only 0.7% of Apps Have Crossed Networks http://bit.ly/ltZJw :( sad [from http://twitter.com/marshallk/statuses/1940079370]
Posted by: Marshall Kirkpatrick
|
June 10, 2009 8:57 AM
There are many shades to open social implementations.
wonder whether Google Wave will breath new life into OpenSocial? If the environment Google is building enables baless
Jerry, What a small world! That's incredible. Hopefully our paths will cross again (and more explicitly) one day... Best wishes yerss
Jerry, What a small world! That's incredible. Hopefully our paths will cross again (and more explicitly) on onem u
wonder whether Google Wave will breath new life into OpenSocial? If the environment Google is building enables baless
I like your point about the cultural differences between networks and how apps that appeal to members of one social network won't necessarily be as popular in another.
The best apps meet the needs of a specific market. Different networks attract different "markets" of people so this will be a huge factor in what catches on and what doesn't, regardless of the technology.
Great article - thanks! It is so important for people to do their homework when starting domain development. If you're interested in learning even more, here is another great site to visit http://BeyondDomaining.com
Tnank a lot![url=http://www.kscastings.com/]Investment Castings[/url]
Most apps would have to meet the current market needs, apps are much better than old days and more people started studying application design courses in london where i study it myself. There is good money in the sector.
thanks for this good article. Its true!
I like your point about the cultural differences between networks and how apps that appeal to members of one social network won't necessarily be as popular in another.
I was really excited about OpenSocial when it was introduced, but now my excitement has diminished ..Thanks for that nice article..
The best apps meet the needs of a specific market. Different networks attract different "markets" of people so this will be a huge factor in what catches on and what doesn't, regardless of the technology. Thanks.
Thanks for nice article...I like your point about the cultural differences between networks and how apps that appeal to members of one social network won't necessarily be as popular in another...
thanks for this good article. Its true!
You've got a real problem with comment spam as people seek links from your site.
Great article - thanks! It is so important for people to do their homework when starting domain development.
Want to see Open Social Doing Big things!
Visit the # 1 Open Social - TV Social Network
TVFanSpace.com - 90 days after launhc its going wild with signups and traffic!
If the environment Google is building enables network site owners to build an app which runs inside Wave and enables the user to see activity relating to their account and contacts, then should Wave revolutionise e-mail, IM and socnet participation an OpenSocial involvement
Great article - thanks! Its true!
Wonder Google Wave will breath new life into OpenSocial?
I totally agree with this point that Only 0.7% of Apps Have Crossed Networks (Updated). Even I had seen many social applications that are been crossed.
thank you very muhch. bunuda yayinlamazsiniz taa:))
Good post. Will watch it. Maybe you have also have a look to Gioco di Bowling
bu bilgiler için teşekkür ederiz.
Thanks for nice article...I like your point about the cultural differences between networks and how apps that appeal to members of one social network won't necessarily be as popular in another...
thanks
I like your point about the cultural differences between networks and how apps that appeal to members of one social network won't necessarily be as popular in another...
thank you very muhch. bunuda yayinlamazsiniz taa:))
Hi,
Many of Facebook's users are online and interacting with the site for hours each day and no doubt have a personal connection to what goes on there.They are my trade magazines, and while I'd never cancel my subscriptions, I am as guilty as most people of ignoring them. On the other hand, I never fail to read Tampa's major daily newspaper every morning (with coffee cup in hand) and monitor certain news websites...
opensocial apis create great opportunity for web 2.0