Sexy Widget

Small Tweak to the iLike Widget

One of my favorite widgets has just gotten a small face lift.  The iLike widget has introduced a couple of tabs that allow the user to toggle between recently played tracks, top bands, all time songs, etc.  More interestingly, the widget is now set to rotate automatically between these various views.

I can understand why they've made this change.  As widgets start to pack more and more functionality and content, you are increasingly seeing these widgets make use of multiple tabs.  The problem is that users are not used to looking for a widget's navigation.  So how do you show the user that there's more to your widget than one view?  You set it on auto-rotate.  Scroll down on the right sidebar to see what I'm talking about.

Random Friday Burrito Widgets

I think the title is self-explanatory:

The Promise of Distributed Social Networks

Over the past couple of days, I’ve been alerted to a couple of new MyBlogLog-like distributed social networks that are attempting to provide a meta layer of social media functionality across existing online communities. 

I find this to be a fascinating model with a ton of potential.  As I wrote about MyBlogLog back in October of last year:

When I think of the hundreds of destination site social networks (many of them funded) vying for visitors attention, and a look at the tiny handful of businesses trying to enhance community functionality for the millions of blogs out there, I see a big disconnect.  If I were investing in these sorts of opportunities, I’d be a lot more interested in hearing about MyBlogLog’s plans that some “MySpace of ______” site.

But it’s not just the relative lack of competition that makes this such an interesting space.  Just as the link friendly nature of the blogging platform has bound together like minded blogs and content, distributed social networks have the potential to make connections between the people that frequent these communities across multiple domains.

Social networks provide great tools for interacting with others, as long as the folks that you want to interact with are members of the same site.  And as great as the blogging platform is for connecting content with links, it is a lousy format for building community.  Communities that have arisen around specific blogs have done so in spite of the blogging platform, and not because of it.

There is clearly a need to extend the social media revolution to the edge – to enable interaction and facilitate content and people introductions across domains.

The primary obstacle that this new breed of distributed social networks will have to overcome is to convince folks to embed their code into the existing community’s page.  If you are trying to build a meta layer of community across blogs, you will need to convince the blog owners of the value of your embed.  If you are trying to tie disparate social network profiles together, you will need to convince influential members of the various communities to give your embed a whirl.

MyBlogLog overcame this hurdle in dramatic fashion, by convincing a couple of influential bloggers of the benefit of their service (community!), and letting things snowball from there. 

As well as MyBlog has done, they have not yet won this space.  As I said back in my original MyBlogLog review:

In my opinion, the three features that are absolutely fundamental to online community are; 1) enabling user to user communication; 2) displaying a user’s historical site activity; and 3) and providing a reputation / recognition system.

Currently, MyBlogLog is only providing the first one.  Users can message each other through the MyBlogLog site, but they can’t browse through comments that they (or others) have posted, or get a glimpse of members’ community and overall reputation.

What will be the next distributed network service to start snowballing through the blogosphere and SNS communities?  Over the next few days, I’m going to be reviewing some of the players in this space.  If you’re working on something like this, or know of a cool distributed social network, please drop me a note.

Widget Platforms and Start Pages: On a Collision Course

If anybody needed more evidence that the widget platforms and the online start pages are on a collision course, check out this announcement from Netvibes by way of Techcrunch:

The new product will be called the Universal Widget API and will be available at eco.netvibes.com/uwa (this site is live now with a landing page, more information will be available next week). Once launched, any widget created for Netvibes, Krim says, will work on the Vista, Google, Mac and Opera platforms as well. Support for Yahoo Widgets and other platforms will follow soon after.

A single javascript embed code will add the widget to any of the supported platforms. The code will recognize the platform and run the appropriate code for that platform within the widget. Once Coriander has launched, sites will be able to create and promote a single widget embed code for most platforms.

Netvibes is already claiming what looks like critical mass in terms of consumer adoption - 10M users, with about 1/3 of them spending at least an hour a day on the site.

If this isn't a consumer facing widget gallery and distribution play, I don't know what is.

So why is Netvibes doing this?  When your product is a glorified RSS reader that prides itself on being ad free, you'd better keep your options open in terms of business model.

Making Sense of the Widget Platform Landscape

I’ve been obsessing a lot about the widget platforms recently (Widgetbox, Musestorm, Snipperoo, Clearspring), and how they might make money. 

In order to get my brain around the business model question, I’ve taken a shot at listing out the various services that these guys are performing with an emphasis on the things that either publishers or end users might be willing to pay for.

For those with the endurance to read all the way through this post, there will be a handy little chart that will attempt to show who is offering what services.

For now, I’m going to focus just on the widget platforms like Widgetbox, MuseStorm, ClearSpring, and Snipperoo.  I’ll save the start pages for another day.

The core services being provided by the widget platforms seem to be:

RSS Feed to Flash Widget Conversion (feed by feed)

This is exactly what it sounds like – an automated way for small publishers to convert their RSS feeds into a snazzy Flash widget.  The two best examples of this are Musestorm and Widgetbox Blidgets, which both offer automated tools that can convert any RSS feed into a customizable widget.

This doesn’t have much potential as a revenue model by itself, as a number of Flash RSS Reader widgets exist, all of which are free.

Content to Flash Widget Conversion (enterprise)

I know for a fact that several of these widget platforms are being inundated with requests from big companies to help them widgetize their content.  A great example of this is how Clearspring recently helped the NBA widgetize player profiles for each of the league’s 300+ pro basketball players.

While big, old school companies can be nice clients because of their marketing budgets, my sense is that none of the widget platforms want to get too hooked on this type of  one-off revenue model.  Unless a widget platform is going to see attribution or be allowed to host the widget configuration tool, I just can’t see these venture funded businesses getting too excited about becoming consulting shops.

Enterprise level widget conversion doesn’t have to be of the hand-holding, consulting business type, however.  Media companies with large amounts of widgetizable content are also looking for automated ways to convert this content into feeds, maximize its potential uptake, and track the results.  More on this to follow.

Distribution Facilitation Tools

Distribution facilitation, for lack of a better term, is about reducing the obstacles that stand between your widgets and their uptake.  For example, there are tools that make it easier for the widget consumer to embed your widget in the community platform of their choice (TypePad, MySpace, Netvibes, etc.).  Widget publishers have the choice between building these tools themselves, and letting a service like Musestorm or Widgetbox handle it for them.  These tools can be a hassle to build, and in some cases, require a business development relationship with the community platform.

Along with these embed tools, I would also include the provision of a high quality widget configuration tool as a factor in driving distribution.  It makes sense to me that the better the customization options that you offer your widget consumers, the higher your uptake will be.  Once again, the widget publisher has to make the build or outsource decision.

So would widget publishers be willing to pay to improve their rate of widget uptake?  As a widget publisher with a small development team and a project list a mile long, I would say yes – as long as it was on a pay for performance basis, and I was able to secure a few “have to have” attributes such as widget to widget distribution, and a crawlable backlink.  I don’t see any reason for my developers to try and recreate the good work that has been done by Widgetbox and Musestorm in building tools that make it easier for end users to consume my company’s widgets.

Widget Analytics

Widget analytics pose some unique issues not dealt with by major site and blog analytics packages.  Specifically, knowing where your widgets are showing up is important information for getting a complete view of how folks are consuming your widgets.

That being said, I’m not sure if basic page view and location data will be enough to incent widget publishers to pay.  Through services like Google Analytics and Statcounter, publishers are used to getting their basic stats for free.  The door may be open for more sophisticated stats to be monetizable, or for enterprise level stats capturing thousands of widgets serving millions of impressions.

Widget Aggregation / Directories

Another piece of the distribution puzzle is the active promotion of your widget via a directory.  Just as Amazon has built an incredibly successful destination site where consumers can take advantage of neat tools like the Amazon recommendation engine and user reviews, companies like Widgetbox and Snipperoo hope to build similar destination sites for users to find widgets.

For the directory driven business model to work, a directory would have to attract a large user base.  With a large user base, business models such as advertising and featured listings could be possible – however, to date none of the widget directories are showing signs of breaking out.

A second potential revenue model associated with widget aggregation could be the sale of premium widgets, with the widget platform sharing revenue with the widget publisher.  There are examples in the gaming and social networking verticals of sites that have been able to sell digital goods to their user bases – World of Warcraft and Habbo Hotel come to mind.  However, these digital goods have been sold in the context of a larger experience, and not in a standalone marketplace.

I remain skeptical that widgets will ever be consumed via a centralized, Amazon-like service.  So far, at least, widget consumers have looked to each other for widget recommendations, or as Tim Post says, have consumed them in the wild.

Widget Panels / Containers

This is a consumer facing service offered by both Snipperoo and Widgetbox that allows consumers of widgets to manage their widgets on their blog or personal site via a centralized control panel.  The benefit of this service is that once the consumer has installed a widget container on their blog, they’ll never have to wade through HTML again.

It’s unclear to me that enough consumers are looking for this type of service to make it a significant market.  Even if I’m wrong about this and widget management panels take off, it seems unlikely to me that consumers would pay for this service.

Summary

So there we have it.  Of the so called widget platforms, we have one company (Snipperoo) that is focused entirely on the widget consumer, two companies (Musestorm and Clearspring) that are focused (almost) entirely on the widget publisher, and one company (Widgetbox) that is providing services to both widget publishers and consumers.

At this early stage of the game, I would rather be facing the publisher than the consumer.  Both the widget directory model and the widget container value proposition seem suspect to me in terms of their ability to attract a large audience.  A consumer emphasis also places you squarely in the path of the well funded start pages like Netvibes and Pageflakes, who are basically providing the same services but in the context of a consumer behavior with more proven demand.

Conversely, facing the publisher allows these companies to cherry pick the occasional large enterprise client and their associated marketing budgets, and perhaps piggy back some exposure from an exceptionally successful widget.  There may even be a long term agency play for the right company that is willing to do a lot of handholding with the big consumer brand companies.

It seems to me that the real winning formula is going to look something like this: build and host an entirely automated platform that widgetizes content, removes the barriers to widget uptake, and provides useful statistics about widget usage.  Charge publishers on a CPM basis so that only those clients that are seeing widget uptake have to pay.  If your platform is good enough, there’s a good chance that one of your stable of widgets that you manage will blow up, providing you exposure and case studies that you can be used to sign up other publishers.

Whether an actual advertising network, affiliate, or ecommerce model will ever emerge for widgets remains to be seen.  If it does, the widget platforms with large installed bases could find themselves in a position to earn a cut of whatever revenue is earned – not unlike Google AdSense.  While the details of how this might play out remain blurry (will host sites allow it?  How many ways will revenue have to be shared?), you have to think that there is value in building up a large footprint.

Anyway, this post has gone on way too long already.  Here’s my best estimate at the services that are available through each of the four players in this space.

Contextually Aware Widgets – Testing Criteo’s AutoRoll Widget

Criteo is a white label provider of recommendation engine functionality to ecommerce and social networking sites, similar to Aggregate Knowledge and Loomia.  Specifically, Criteo uses collaborative filtering powered by data gleaned from millions of consumers to crank out product, service, and content recommendations.

AutoRoll is Criteo’s attempt to pack this recommendation service into a widget.  Installed AutoRoll widgets collect data about the visitors who frequent the widget’s host blog.  This data is fed back to Criteo’s recommendation engine, which scours the AutoRoll network for other blogs that might be of interest to the reader based on the usage patterns of others.

It’s a neat concept, and the execution should get even better as more and more blogs give it a shot, feeding more and more data back to the Criteo mother ship, and expanding the universe of potential blog recommendations.

So how did it work out for Sexy Widget?  Just about perfectly.  If you look at the bottom widget in the right hand column, you’ll see that AutoRoll is recommending Widgets Lab, Widgetoko, Stickiwidgets, Technically Speaking, and Wii Rally.

The first three blogs are about as good a match for Sexy Widget that exist in the blogosphere.  Technically Speaking looks like a relevant, quality blog that I didn’t know about, and Wii Rally is a gaming blog which may or may not be interesting to Sexy Widget readers.  Overall, these are excellent results, and assuming that they are organic and not hand tweaked, they speak very well to the accuracy of the engine. 

Obviously, each of the recommended blogs is displaying the AutoRoll widget.

From a publisher perspective, there are two obvious benefits: 1) providing a nice feature for your users; and 2) putting yourself on the receiving end of some referrals from related blogs. 

From a referral perspective, it's important that Criteo find a way to "get credit" for the referrals they are sending.  I haven't noticed Criteo showing up in my log files - if there's a way for them to redirect the referral clicks that a publisher gets through Criteo.com, it could go a long way towards proving the widget's merit to the widget host, and guaranteeing continued real estate on the sidebar.

Taking a look at the big picture here, I think what’s really neat about this widget is the fact that it’s contextually aware.  Presumably, no two AutoRolls are the same, and the widget adapts itself to the readers of each blog.  This is next generation kind of stuff.

It seems to me that MyBlogLog could provide this service as well with the massive amount of blog usage data that they have collected, so I would imagine that Criteo is pushing hard for distribution.

Going forward, I haven’t decided yet if I’m going to keep the AutoRoll on Sexy Widget.  Maybe it’s just the arrogant blogger in me, but I sort of feel that if there’s a blog worth reading in the widget space, I’m going to find out about it and drop it in the sidebar myself. 

That being said, automation is a big plus.  It’s conceivable that the “blogs about widgets” space will get so big as to become unmanageable.  In that case, an AutoRoll widget would be invaluable.

LinkedInABox Widget – What the LinkedIn Widget Should Look Like

A while back I left a review of the somewhat Spartan LinkedIn badge in which I concluded:

LinkedIn has stuck to its guns of being a no nonsense professional service that offers the functionality you need to manage and extend your professional network, and not much else.  There are no user photos on the site, no html embeds allowed, and not much of the other functionality that most social networking services offer.

The LinkedIn widget is an extension of this highly focused approach.  As a self-proclaimed widget aficionado, this widget leaves me a little cold – there’s just not much to write about.

I’m happy to say that there now is a more interesting LinkedIn widget to write about.  A service called LinkedInABox has developed a JavaScript widget that does a nice job of capturing the key elements of your LinkedIn profile for display on your blog.

The LinkedInABox widget configuration tool lets you select a display name, the fields from your LinkedIn profile you would like to display (Summary, Education, Connections, Etc.), a color, and a background image.  You can see my LinkedInABox widget in the right hand column.

I think these guys did a nice job.  The widget is extremely easy to set up, and packs a lot of information in a small footprint.

There are two big stumbling blocks for LinkedInABox to overcome in order to gain broad adoption.  The first is convincing folks to share their LinkedIn login information with an unknown site.  If I were these guys, I would be plastering the home page with whatever third party trust seals that they can find, as well as linking to a comprehensive privacy policy that details exactly how my login information will be used, stored, and protected.  If I hadn’t been a reckless widget blogger, I probably would have been hesitant to share my LinkedIn login info.

The second, and far more serious stumbling block, is the lack of control they have over their own destiny.  This service is completely and utterly dependent on LinkedIn continuing to allow them access to the profile data.  Should LinkedIn decide that they would prefer that consumers of LinkedIn profile data click through to their site (where the traffic can be monetized), it will be lights out for LinkedInABox.

The business model here appears to try and get bought by LinkedIn. 

As a loyal LinkedIn user, I want the ability to display MY profile data wherever I want it.  The current LinkedIn badge doesn’t cut it, and the lack of a real distributed presence is a weakness that deserves to be exploited by entrepreneurs like those from LinkedInABox.

I hope that this widget gets some traction.  At a minimum, it should put the pressure on LinkedIn to set their our data free.

Yahoo! Pipes and Widgets

Man, is Yahoo! Pipes good news for the widget ecosystem.  This will take user generated widgets to a whole new level.

The launch of Y! Pipes is raising all kinds of questions... Why doesn't Yahoo! have a "widgetize this pipe" button?  Are Musestorm, Widgetbox, and Clearspring racing to get the RSS to Flash Widget conversion business from Y! Pipes?  Is Y! Pipes on its way to becoming a standalone social network in its own right?  Y! certainly has the social media infrastructure to make it happen - profiles, ratings, content modules, etc.   How can publishers position their feeds as a core module for Y! Pipes? Could there be a pay model built into this for access to certain premium feeds, with Y! earning a nice cut?   To me, that sounds like a business model to get excited about.

As Bradley Horowitz says, it's tough so early into this to be able to get your head around all of the implications of something so radical.

For more analysis on Yahoo! Pipes, see Flying Seeds, RogerD, and O'Reilly.

Widget Mail – A Review of Gigya

Gigya is a web based email service that allows you to embed widgets into the body of your email messages.  Widgets can be chosen from pre-existing categories (layouts, smilies, videos, etc.), or by posting code into an embed field. 

Gigya feels like social media applied to email, and it’s not just the widgets that cause that effect.  Your contacts are displayed as headshots / icons, not just email addresses, and you can upload your own photo as part of your profile.

I had no problems either selecting widgets from the Gigya menu, or embedding my own.  But here’s the catch: if you embed a Flash widget into an email, it arrives to (non-Gigya) recipients as an image.  In other words, recipients that are not checking their mail at Gigya.com are forced to click on a thumbnail of the widget in order to see the Flash version.

This causes no issues for widgets that are layouts or images.  But for more interactive widgets like video and RSS feeds, it’s not ideal.  Clicking through on the image of the widget takes you to a hosted Gigya page where the actual widget is displayed.  It’s a bit clunky.

Conceptually, Gigya is a killer idea.  When was the last time we saw a radical rethinking of email?  Email doesn’t have to be boring and stodgy.  Widgets could go a long way towards reviving this sputtering medium.  Furthermore, email has a built in viral factor which should give Gigya a good chance of getting some adoption. 

One minor point to continue my recent theme of widget distribution: if widget enhanced email takes off, services like Gigya could be a nice channel for widget aggregators to partner with.  For example, I’d love to be able to pull my Widgetbox or Musestorm widgets right from my Gigya control panel, for insertion into my email.

As reported on Mashable, Gigya has recently raised some money.  They're off to a solid start, and I can’t wait to see if the concept of widget mail has legs.  (somebody already snapped up widgetmail.com btw)

Webwag Widget on Demand

If you haven't tried out the new Webwag widget on demand (WOD) yet, you should.  It's neat.

Webwag is a start page similar to Netvibes, PageFlakes, and YourMinis.

The best way to think about the Webwag WOD is as a dynamic, screen capture tool whose output you can display as a module on your Webwag start page.  For example, you could use the WOD tool to capture the "Cool New Videos" section of the MySpace home page.  New videos showing up on the MySpace home page would also be reflected on the WOD module of your Webwag start page.

Why is this handy?  Because as far as I know, MySpace doesn't publish either RSS feeds or widgets of their "Cool New Videos" section.  Here's a screenshot of my Webwag start page with the MySpace WOD embedded:

Because these modules can only be placed on Webwag start pages, I consider them more "site specific modules" than widgets.  But given that YourMinis and PageFlakes have shown us that it's no big jump to make these sorts of modules exportable as widgets, you have to think this is coming for WODs.

Ready or not, the web is modulizing, fragmenting, and yes, widgetizing in front of our eyes.

See more on Mashable.