Enterprise RSS

Open Social Developers Journal - Show Me the Money

Sometimes it’s nice to be old. Well, it’s at least nice to be able to remember mistakes so that we can hopefully avoid repeating them. Back in the “Web 1.0” days, there was a period of euphoria where everything related to the internet was certain to be incredibly profitable. And now we are in the era of “Web 2.0”…

So how does anyone make money in the OpenSocial world? We’ll be chatting more about the business perspective for this in an upcoming webinar, but let’s just focus on the developer’s side for a moment here. Advertisements stand out as the clearest path to dollars in most scenarios. What does it take to include advertisements in an OpenSocial Gadget?

First, here’s the world’s shortest primer on web-based advertising. Ads make money when the end user sees them or clicks on them. The ads that make the most money tend to match the other content on the page (e.g. put sports memorabilia ads next to sports news articles) and/or target characteristics of the user (e.g. offer Denver Broncos tickets to a person who lives in Denver).

Since most social sites require a login, the ad networks can have trouble spidering the pages to know what the page is really about. Some networks have options (which are not available to all accounts) that allow the request for an ad unit to say what the content is about (“this page is about automobiles – send me car ads”). Of course, this only works if the application somehow has insight into what the page is displaying. Clearly an application should know what kind of content it is showing. If I write a dating gadget, it’s probably safe to request ads about dating (hold that thought for a moment…)

Targeting by user seems really intuitive for social gadgets. Given the success of Facebook, several ad networks have appeared that focus exclusively on these opportunities. It seems likely enough that someone will have a fairly robust offering. Most likely it ends up being a “least common denominator” offering like Lookery which just uses Age, Sex, and Location. But what happens when a container site can’t tell the application the user’s gender and whose problem will it be to translate all of the possible ways that “location” might be described in different social sites into something the ad network understands?

In addition, OpenSocial has the concept of the user and a visitor. When I browse my friend’s profile, I’m a visitor. But the application doesn’t have any rights to get my profile information, so what kind of ad should be displayed in that situation?

But the real challenge here is understanding the rules of the container site. If I make an application that’s about dating, and a dating site decides to support OpenSocial applications, I really doubt they would like to see ads for their competitors showing up in my application on their site. Today there’s no API that conveys any rules for advertising. Pragmatic social sites will probably screen every application to see if it meets their standards. But what prevents developers from modifying the application later? And just how quickly will social sites review applications?

Remember the early days of getting an application into the Facebook directory. Right now, I’m waiting for one of the OpenSocial launch partner sites to respond to my request to add Didja Hear to their approved list.

Advertising within the gadget isn’t the only path to money for OpenSocial Applications. But other monetization options like affiliate sales programs or cross-promotion fees for other application developers have both technical and business challenges as well.

In many ways, this is very similar to the interior decorator problem. Application developers can build unique, highly-customized versions of their applications for each target social site. Or they can take their chances with some very basic advertising strategy which will work everywhere (but probably not perform well anywhere).

The ideal solution seems like it would be an ad network designed specifically for OpenSocial. It would need a really large inventory of ads along with an API that would allow container sites to express some basic concepts (including forbidden advertisers and targeting rules). It’s a good bet there are some bright minds considering this problem right now…

I’m not old enough to have experienced the 1849 gold rush, but I’ve heard that the best money was made in supplying the prospectors. Clearly in the “web 1.0” world, a lot of companies profited by solving the difficult problems in a new emerging landscape. There’s gold in these OpenSocial hills, but we haven’t yet seen whose going to profit the most from it.

P.S. While we’re waiting for the container sites and API’s to get stable, you can check out this page to catch a quick screencast about Didja Hear!? We’ll be posting links and instructions on how to add Didja Hear to different social sites on this page as well.

Brian Kellner is VP of Products at NewsGator

3 Things You Should Know About NewsGator

Most people identify NewsGator with our market leading desktop RSS client software and as one of the first online RSS reader services. What most people don't realize is that the consumer client software and services are only one part of our business. Here's 3 things you should know about NewsGator:

1) We have over 39 separate products, including Facebook and Google OpenSocial applications, a comprehensive enterprise products business, and a content syndication services business. We have a product called Social Sites that runs on top of Microsoft Sharepoint to deliver a "facebook behind the firewall" capability, and we work with other enterprise 2.0 companies to deliver suite offerings, including Intel's SuiteTwo.

2) Our datacenter processes over 1.8 million feeds daily, resulting in over 7 million new content items. We archive content and have 3 terabytes of content that we make available to our partners. The archive content is increasingly being used to power features for discovery and relevancy of content.

3) Some of the largest media brands in the world use NewsGator's syndication service to "widget'ize" their content, including USAToday, and The Discovery Channel. We deliver over 100 million widget impressions a month and our media clients typically achieve a clickthrough rate that is double what they get through other channels. You can use the free version of this service to create your own widgets.

OpenSocial Developer Journal : Blind Interior Decorating

My wife is a huge fan of HGTV. It’s actually pretty cool to see how different a room can look when a skilled designer goes to work on it. But how would this turn out if the designer couldn’t see the room? Building applications for OpenSocial is somewhat like this.

Container sites are supposed to provide some space on the “profile” page for a user and another “canvas” page that has more space. But just how much space will there be? What color is the site? What fonts are used? Can the user modify size, color, font, etc?

These aren’t new problems for web design. Users run their computers at different screen resolutions for example. But now we have an additional layer of complexity in trying to understand how the container might vary from site to site.

In general, this pushes application developers to go for simpler implementations. The blind interior decorator is likely to pick neutral colors and small pieces of furniture. But this doesn’t create the best user experience. Another alternative is to build an application solely for one container site. This certainly produces a good user experience, but it doesn’t really leverage OpenSocial. It’s really no different than building to a site-specific set of API’s.

The alternative that we are pursuing for the moment is to make a version of Didja Hear for each container site. This way, we can provide the best possible experience in Plaxo which might look significantly different than the best experience in Orkut. We might size the preview of a video differently. We might show more or fewer videos in the view. We might change the colors or fonts. This gives the best user experience, but at scale we will have a very large number of variations of each application. Developers don’t like this because it makes maintenance much harder.

The other approach is to have some set of API’s that inform the application about how it should look. Discussion started on this in the weeks before OpenSocial was announced. It’s not a simple problem, but it’s necessary in the long run. Otherwise we end up with 100 different Didja Hears and 1000 “beige room” applications.

OpenSocial Developer Journal : Have We Met Before?

NewsGator is all about presenting the right view of your content when and where you want it. An important part of that concept in our RSS readers is synchronization. If I mark a post read on my mobile NewsGator Go! reader, that post is marked read everywhere else. We do this by tieing all of the user activity to a NewsGator account.

Now suppose I were to add Didja Hear (our OpenSocial application) to Plaxo where my user ID might be based on my NewsGator email address. I set some content preferences so Didja Hear only shows videos about movies, television, and music. I start sending videos to friends, making comments, getting videos from them, etc. Later on, I add Didja Hear to Orkut where I log in with my gmail address.

As an application developer, this scenario raises some questions. Should we prompt the user whenever they add our application to see if they have added it before? If we do that, we could link the two accounts. Then we can provide a benefit of keeping the same preference information.

We could take it a step further though as well. In my scenario above, I could possibly see a video sent from a friend on Orkut while viewing Didja Hear in Plaxo. That sounds like a nice benefit for the user – wherever you look at Didja Hear, you see all of your content and interaction from friends. But in addition with us needing to ask the user to connect the dots and the user feeling comfortable with connecting the two accounts, we have another couple of issues.

If my friend Sue in Orkut puts a comment on a video she sent to me, we show that as “Sue says: …”. We don’t store Sue’s actual name in our database. We store her Orkut ID. Now if I’m looking at Didja Hear in Plaxo, the only way we could show “Sue says: … “ is if we store that information. The alternative is for us to display the considerably less social message “Somebody said: … “.

Facebook has an explicit policy forbidding the storage of a user’s personal data for more than 24 hours, and I could see different container sites creating different usage policies around personal data. As an application developer, it would be ideal if there were some standard restrictions that could be applied to profile and friend data consistently from all container sites. For example, a user setting that says “Share my first name only in views outside this site” would be very helpful piece of information.

One small twist remains to be explored in this story, but that’s a subject for a future post. In the meantime, we need to get back to some real OpenSocial development.

OpenSocial Development Journal – Know Your Friends

At NewsGator, we’re happy to have been selected as one of the early developers in the OpenSocial program. It’s been exciting to build our Didja Hear!? application, and we wanted to share some of our thoughts on social networking and the opportunities, questions and challenges of OpenSocial from the application developer point of view. This is the first in a series of posts that cover the issues we are currently addressing.
 
One of the OpenSocial API’s lets us ask who are the friends of a user. This is a powerful and basic capability of any social networking solution. The primary challenge for us as a developer is understanding what a container site really means when they tell us who are the user’s friends.  
 
Imagine you asked ten random people to name all their friends. Each of them is likely to use different criteria in defining what a friend really is.  Some people might answer the question with a statement like, “Well, I only have two really close friends, but I stay in touch with fifty people.”  
 
Within a single social networking site, the answer is likely to be very consistent. But different sites may choose to answer the question differently. When we were building the Didja Hear application, the rule for Orkut’s answer to the question  “Who are Bob’s friends?”, was to respond with Bob’s friends who already have Didja Hear installed! If we contrast this with Facebook, we can get a list of all of a user’s friends and the information about whether or not they have our application installed.  
 
From the point of view of viral distribution, it’s much harder to prompt Bob to send Sue an invitation to add Didja Hear if we don’t know that Sue even exists.  We’ll have a bit more to say about invitations in upcoming posts…
 
The filter that Orkut was considering around providing friend names is just one example. If we are working with a social network that allows more granular concepts than “friend” or “not a friend”, how will they answer when Didja Hear asks? The most precise answer would be for the site to send us the users along with extended data that indicates how close they are (best friend, friend, business associate, friend of a friend, etc). But if every container site does this, it could be overwhelming to deal with the complexity.
 
OpenSocial is very cool and very powerful. It is already making an impact. But underneath powerful, open concepts there are always important details that need to get ironed out. Over the next several posts, we’ll show you more of the questions we’re considering right now.

-- Brian Kellner is VP Products for NewsGator

Tags: NewsGator, Google, OpenSocial, Didja Hear

Live From Defrag!

Day 1 at Defrag turned out to be well worth the trip out here. With approximately 250 people in attendance it is the perfect size to meet pretty much everyone and wtih people like Doc Searls, Brad Feld, Phil Windley, Andrew McAfee, Esther Dyson, and Jeff Clavier in attendance, well you are certain to have some good hallway conversations.

Dave Weinberger opened with a keynote that was insightful and philosophical, my favorite takeways were that "links are not information, links do what information doesn't" and ""words are not carriers of meaning, but pointers to shared understanding" (UPDATE: read Dave's comment on this). What I found fascinating about Weinberger's keynote is that he makes the web and what we are doing with it meaningful in a big picture kind of way.

The panel discussion that followed the keynote featured Jerry Michalski, JP Rangaswami, Joshua Schachter, and our own JB Holston. This discussion covered a lot of ground but one thing that caught my attention was how there is a tension in enterprise organizations between open vs. closed, and young vs. old. It's nothing new to suggest that there is a generational divide that dictates ideology and tools.

Dan Farber posted a summary of the first two keynote sessions.

Michael Barrett has a presentation on security which can be summed up as "web 1.0 security is broken and web 2.0 just adds more stuff that will break". Phil Windley captured the presentation in an appropriately named This Stuff Scares the Hell Out of Me post.

We had an open session slot on the agenda where we broke up into working groups to discuss various topics, such as who owns the data, and platforms, which I participated in. The platform open session was interesting because it illustrated the dramatically evolving nature of what it means to be a platform. One thing we all agreed on is that platforms don't declare themselves to be a platform, what happens is that success with customers creates the conditions by which a application evolves to be a platform.

There was a lot more going on in the platform open session, which could have spawned several blog posts but one thing is clear, the platform game is changing and if you want to play in that world you had better understand the new world order.

Next up was a vendor presentation from BEA, which is a sponsor of the conference. All speculation about their future aside, I really like their Aqualogic product line but wonder if the evolving nature of portals chips away at their market opportunity outside of the large enterprise segment.

I saw Andre Durand and Kimbal Musk, was a great conversation and then Jeff Clavier walked up and we went out for coffee.

I moderated a panel on enterprise mashups with Adam Gross from Salesforce.com and John Crupi from Jackbe. We covered a wide range of topics related to enterprise mashups, including security/governance, APIs, monetization, start page apps, social networking, and user interfaces. I enjoyed this discussion and I believe we may be on the cusp of delivering some cool mashup capabilities to enterprise users.

My mashup panel would have had a larger audience were I not going up against another panel on "social networking the enterprise". It's pretty tough to compete with that session right now so we had maybe 40 people in it, however at the end a couple of guys from a company doing mashups came up to me and commented that we covered some good topics not normally talked about in mashup presentations so I'll take that as a success.

We next had a sponsor challenge from NewsGator CTO Greg Reinacker, who I thought did a great job presenting RSS in the enterprise. He's actually very funny, you should book him as a speaker at your next conference.

A highlight of the day is Dick Hardt fragging identity at defrag. It was a new presentation and, as would be expected, very good. I'm sure a slideshare version or video will be up in due time. Esther Dyson gave a presentation about stuff that was way to deep for me to appreciate.

Doc Searls gave a very entertaining presentation on "customer reach vs. vendor grasp", which can be summed up as how little control over your personal profile data you really don't have. Doc is insightful and pointed in his presentation, which like Weinberger, usually takes days of thought to fully grasp. Smart people like Doc think at a different clock rate.

One slide that said a lot more than the words allowed: "we lost more than our identities at the end of the Enlightenment, jobs replaced crafts". Doc makes an important point about identity should mean something more than a name and attributes, identity should reflect a transactional relationship between systems and people.

Lastly, Doc's project around Vendor Relationship Management (project VRM) addresses the way in which we interact with companies. VRM is misnamed though given that in enterprise software circles the term has been used for years and means something completely different, although upon second thought it's really the same thing with different players (b2b instead of b2c)

Last up today is Ross Mayfield, who had great news this week about Socialtext raising more money and getting a new CEO. Ross is never at a loss for entertaining presentation titles is closing out the day with "What to do in Denver when your corporation is dead". I'm running out of battery so I'll have to deprive you of Ross' wisdom for now.

PS- Here's some images on Flickr taken through the day.

--- Jeff Nolan is VP Corporate Development for NewsGator.

OpenSocial, Didja Hear!? and Enterprise 2.0

Some days I imagine the internet must want to take a deep breath and just rest for a while. With Google announcing OpenSocial, a huge stream of blog posts have been pouring down the "pipes".

Within the flood of blog posts, you might have seen NewsGator's name bobbing along with something called Didja Hear!? We were very pleased when Google asked us if we would like to be an early participant in this program, and we used the power of our Widget Framework and our experience from building NewsFriends for Facebook to create Didja Hear!? which pulls the best video and graphic content out of the seven million posts we process each day and makes it easy for users to share these videos and their comments. Here's how it looks in Orkut:

This application ends up being something like the Web 2.0 equivalent of a water-cooler chat with a great conversation starter. We recommend content in categories like movies, music, sports, celebrities, televesion, and gadgets and users send them on to friends or reply to videos and podcasts shared by their friends. So we leverage NewsGator's back end to select great content, and we leverage your social network to highlight the content that is most compelling for you.

So what does all this have to do with Enterprise 2.0? In some ways, Didja Hear!? has more in common with our Social Sites for SharePoint than it does with NewsFriends for Facebook. Just as Social Sites allows for an administrator to select content to make the experience more compelling, Didja Hear!? uses our editor's tools from the Widget Framework to get the best content by automatically filtering and sorting the content. We can also promote specific stories and eliminate unwanted ones.

But the real power happens when user behavior and relationships are added to the mix. The explicit actions of my friends to share content are the most powerful and accurate filters available. But we also take advantage of less explicit feedback. By tracking user actions including clicking links, forwarding articles, saving articles, etc, we can improve relevance. Even if my friends don't actually send me a story, the fact that they spent time interacting with it suggests that it's more valuable to me.

This is true in both consumer and enterprise applications. OpenSocial exposes information about a user, her friends, and her activities. This is exactly the same kind of information that is available in SharePoint or Connections. We leverage that information to provide the best content for a user, and we extend that information by adding in the reading, tagging, subscribing, saving, and other attention data to provide better connections of people and content.

This same story applies in our Syndication Services business where we leverage the same social content intelligence in our platform to provide the best possible widget experiences.

At the end of the day, all of this "2.0" stuff comes down to taking advantage of data that users provide for their own benefit to produce more benefit for them and others. Social and content consumption data are both extremely powerful - combined they produce a truly exceptional value in both business and individual user scenarios. And after the flood of posts has subsided and the "pipes" of the internet groan in relief, getting real value from OpenSocial will be the center of attention.

Brian Kellner
VP, Product Management

P.S. If you want to understand OpenSocial better, read Marc Andreesen's post. He gives a great overview of the technology as well as broader thought about its impact.