Dave McClure (who helps teach the Stanford Facebook class, and runs the Graphing Social Patterns conference), Justin Smith (of Inside Facebook), Rodney Rumford (of FaceReviews), and myself brainstormed last night trying to understand the current state and direction of the emerging widget industry.
The widget universe is vast with a lot of variation. So although the initial goals was to try to categorize the widgets into clean buckets, and we found that to be an impossible task. Widgets can often have multiple attributes, so instead we focused on attributes and characteristics, rather than groupings. We developed a couple of models, (none that are perfected) but noticed a few ways to look at the attributes:
Levels of Data Interaction
Highest | Application uses data from your social network | iLike
High | Application uses data from your preferences | Pandora
Low | Application pulls data from source | Audio stream (like a radio station)
None | Static widget, display badge | Widget links to other website
We also explored a model of self-expression/vanity/entertainment vs utility and communication. And also found that some apps have different lifecycles from simple one time users (disposable widgets) or to those that got more value over time as network usage increased.
Later, Ro Choy from RockYou came by and presented me with a very clear definition of the differences between widgets and applications. Essentially (and I have a video of him to come) widgets are limited in functionality (due to limitations in size), usually resulting in the widget creator trying to get the user to go another website, vs an application that has full functionality, is on multiple pages, and therefore the widget creator doesn’t need to lead the user off the application.
So in summary, this was one of the first cracks at trying to segment the landscape, we’ll have to have subsequent meetings to drill down even farther. I’ll be mulling this whole thing over for the next few days/weeks to try to make sense of all the learnings. It’s a much larger universe (over 13,000 widgets exist, which will likely double this year) developing a taxonomy will be challenging and fun.
I was planning to live stream the event, but had issues with my wireless, although Rodney was taping and I’m sure he’ll post it. Chances are, you’ll find it very boring and dry, as we weren’t playing to the camera, we just had it on from the other side of the room as we discussed.
Hi Jeremiah,
Great post – thank you for sharing your insights with us. When we’re talking to our clients, we discuss five major attributes of widgets that are possible to build:
1. Games
2. Competitions
3. User Interaction
4. Tools
5. Badges
I was inspired by your article to write about these:
http://socialmarketingstrategy.blogspot.com/2008/01/five-major-attributes-of-widgets.html
Most applications and widgets combine at least 2 of these attributes. If anyone agrees/disagrees, would add more etc then let me know.
Thank you for the great post Jeremiah.The widgetsphere is growing at an incredible pace. Causing great fragmentation of media into a million little widgets and apps. We are focusing on applying Apple design, a business model and a bit of good ols innovation to our widget strategy.This post is very helpful in adding perspective to our strategy at Mediatrust.
Hi Jeremiah, thanks for sharing. That’s an interesting way to segment the landscape.
One other major difference between widgets and apps is the means of distribution. Because apps are tied into the social graph, they can be pushed – via invites, app interaction, etc. Widgets rely on more passive distribution – a user sees a widget on a blog or profile page, and grabs it for their own use. This social graph tie in plus push capability causes apps to be 7X as viral as widgets (source: RockYou).
Just found another great article on this – check out Facebook Applications Trend Reports 1 by Asi Sharabi, interesting analysis of the top 100 apps.
Thanks for your comment, completely appreciate that there are way too many different types of widgets/applications to classify. Unfortunately we need some kind of framework to give to clients!
Hi Jeremiah, thanks for sharing. That's an interesting way to segment the landscape.
One other major difference between widgets and apps is the means of distribution. Because apps are tied into the social graph, they can be pushed – via invites, app interaction, etc. Widgets rely on more passive distribution – a user sees a widget on a blog or profile page, and grabs it for their own use. This social graph tie in plus push capability causes apps to be 7X as viral as widgets (source: RockYou).