A panel from Google. Social objects are anything that promotes a social activity. For example kids bringing home a stuffed animal for the weekend and you do activities with it, take photos and tell the story the next week in school. The question is how do we socialize objects online without having to create yet another social network? There are 18 platforms sighed up for OpenSocial and has access to over 200 million users. OpenSocial is just to the point that it is useful and is live. Most sites are launching with very limited functionality and will continue to add more as time goes on.

OpenSocial is one API and 2 parts.

  1. client side javascript
  2. server side rest

3 core sets of service. People (who am i and who are my friends), activities (post what is happening), persistence (storing the data on the platform).

Caja – when gadgets go bad. Caja is spanish for box and is meant to put apps in a box so they are controled. Gadgets can be a new way for spam and viruses if we are not careful. That is where caja come into play. It is a javascript sanitizer. It is optional but recommended for opensocial containers. Eventually will be secure enough to run gadgets inline instead of in iframes. That is the end goal of caja. Some platforms will require the use of caja. Users could also get a security warning if caja is not used.

Shindig – Open source project that is the reference implementation of opensocial plus gadgets stack. The goal is to launch a new simple containter in under an hour’s work. Contains a javascript container, gadget server, opensocial container javascript, and the opensocial gateway server. Java and javascript version are running now. PHP gadget server contributed by Ning and was recently rewritten to match java. They are looking for other platforms.

Stop waiting for OpenSocial to settle down. Now is the time to start coding. Everything is ready for production!

Jai Shen from Rock You:

Launch an application with a marketing and validation test. Test the audience, messaging, and channels (different viral channels on each platform). News feed, notifications, email, profile, invites, profile action, non user pages and profile are different channels to think about. Understand your application vertical like channel, content, quiz, games, gifts, and self expression. Messaging is all about the call to actions. Do users like it and does it work. Once you have the app built and messaging working well focus on groath. Viral groath is when one user causes at least one user to install the app. The largest part of app development is tuning. Track everything, Graph it and trend it. Manage your notification and invite allocations and notification spam blocking. Rock you suggests to be consurvative at the beginning until you understand your response. Then get more aggressive when you have confidence in your invites. This will ensure that your app does not get penalized in the beggining while you are building your app. Once you have grown your app focus on user engagement. Saturated social circles and tune for a better user experience. Rock You has apps that have had invites that have gone to every user in facebook. That is saturated!

Virality. grow outside the core group of users. Get users to spread to others. Dont hurt your user experience. Many people forget that a good user experience will be the biggest factor in making an app viral. Non user pages have worked well for rock you to engage users before they install. Many users want to see the app before they install.

OpenSocial has some challenges. The audience is different. It has a huge audience and has a promiscuity factor. Open social has a different user model. Scrapbook is big in Orkut. OpenSocial channels – news feeds (most people on open social are not really using yet), Profile, main page, bulletins, messaging and invites and non user pages. There is still a lot in the air with Open social. From a business perspective, Rock you sees not difference in opensocial as facebook. Focus on apps that are not dependent on viral channels.

Tyler Ballance from Slide. Facebook platform

Before you start understand the platform. What data is available? Understand the audience of the platform. What can the API do? Beebo still has just a subset of the functionality that facebook has to their developers.

Going cross platform. One approach is sharing the exact same code base. This has worked for top friends. Another option is to copy the code and change up the UI. This is what fun wall did. Use consistant branding when you go cross platform. Users start to recognize top development companies and this gets them more trust with the users. Use FBML so that you get the look and feel that matches the site.

The google computing cloud model is about getting the computer out of the way so that we can be fore productive. The social cloud is about getting the computer out of the way so that we can interact more easily and be more productive. People are the killer app on the web. Get the barriers out of the way so that people can just connect.

How do you know if it is social? Add …with your friends to the end of the sentence. You need a new insurace agent … with your friend. You need to shop for a new car … with your friend. You will find new music today … with your friends. You will plan a trip … with your friends. That means they can be social.

Some hard problems left to solve

  • Who are your? Authentication is important and how do you get access to your data and your friends. How many of us used the same username and password on sites that dont need security and ones that should be secure? All of us.
  • Will you be my friend? Invitations are out of control and people are getting more and more hesitent to share their friends info.
  • will you build an app for me? Building an app again and again and again. This should not be necessary.

The goal is to make the social web better. Googles answer to these problems is OpenSocial. It seems like this ideology is really focused on making the user experience better on the web.

Step 1. OpenSocial. If you can build a web app, you can make it social, and reach more than 200 million users. 1. invent it, 2. build it, 3. run it.

What has Google learned about social computing? There are classic social networks like beebo, myspace, hi5, nicng, orkut, plaxo. There are business networking sites like linkedin and viadeo. Then there are enterprise software like oracle and salesforce. Lastly there are communities like hopy sites, interest groups, marketing networks etc. There is a large breadth of interest.

Next, open source is a good idea. Shindig is an apache incubator project. For this to work you have to get some things right. Clear mission, open license, engaged community and real world use.

Will other google apps offer suppor for opensocial and get social feature? David Glazer from google kind of dodged this question but said that they have had conversations about adding such features to MANY of their apps. This type of functionality will be really exciting.

EcoSystem history – MySpace has allowed application for years. Myspace is about self expression and is about high levels of customization. The new developer system is more robust and allows for more functionality.

Myspace developer platform supports 3 sets of APIs:

  1. Open social apis – full support
  2. REST apis – server to server communication and includes oath authentication.
  3. ActionScript apis – flash support

Opensocial – Why Opensocial? Myspace says they have a commitment to open standards. They opened up profiles over 3 1/2 years ago and they still believe in being open. They also believe that portability is important and allows developers to spend more time building apps and less time on rebuilding the same app for each platform. Portability also allows data to be shared between platforms. Also will have full support for the OpenSocial public spec. They will also provide additional MySpace specific extensions like bulletins and additional attributes for bands.

Platform surfaces – The development surfaces are different on myspace than on other platforms and provide some new functionality for app developers that has not been available on facebook. These surfaces include the profile surface, canvas surface, user homepage surface, application gallery, and application profile. The user homepage is private to each user and no one else can see. This will allow apps to display information that is important to the user but not to everyone else. Like an ebay bid widget or a twitter app that shows all your friends twitters.

Security, privacy, and safety – Applications will go through a safety review processbefore going live to our members. Applications will be governed by the same privacy controls that are in place for members. They will be employing new technology designed to ensure that applications are safe for end users. This includes proprietary MySpace technology and also includes Caja which is a Javascript sanitizer created by MySpace and Google.

Balancing vitality and the user experience – They are taking a long term approach to growth and distribution and ensure a clean application experience. Artificial or spammy growth is not necessary. Initially applications will be able to initiate the workflow for sending a message on a 1 on 1 basis. Basically this means that the user will send any invites just using their standard messaging system. It also means that viral growth will be very difficult. He said that this may change in the future.

Presented by Chris Messina from Citizen Agency. He went to civicspace to spreadfirefox to flock to citizen agency.

Moving from a system centric type system to a citizen centric model. The web citizen has an ID, provenance, friends, agency (choice is good for everyone).

Aspects of a distributed social network:

  • Something that can identify you
  • Manage your profile form one place. Push everything out to the web. The web should update from me not the other way around. This makes it easier for the user
  • Flexabile permissioning system
  • Based on open formats and protocals

Benefits of the distributed network

  • Cross pollination of data
  • Spend more time updating profile and keeping it up to date.
  • Can compete outside of the individual platforms. This increases competition.

Problems with the distributed social network

  • How do you find your friends. Facebook does this pretty well but it is hard on the web in general.
  • Maintaining a common interface. The further you get out from the trunk the more inconsistant the experience is.
  • Interoperability. How do you upgrade?
  • Addoption
  • Who do you fix and troubleshoot problems
  • Identity theft is a problem

Components of a DiSo site

  • People, friends and identities. Publish your friends list so you can move it around.
  • Messaging and notifications. Too many and people are not reading them. There is no way to prioritize them.
  • Limiting and giving various levels of permission. Plaxo does a good job at this.
  • Groupings and events. Viddler – join a group. Upcoming. Need openid for events and groups.

What can be done today? Allow users to say where else they are on the web using micrformats (microformats.org) . Use openID. Use oauth.net. Use jabber.

Current status. diso-project.org. or diso.googlecode.com. Creating design patterns and how things should be done (look at flickr photos).

Users feel frustrated because they have to re-register and give their access info every time. But companies are scared to make data portable and want to protect their piece of the pie.

One issue is who actually owns the content on a social network. Most agree the user profile and any data they enter belongs to the user. But what about things like comments? Who owns those? Myspace thinks they own any content that their network enables to happen.

There is a data portability initiative that will allow a user to take their data with them. All the big names are signed up. It will take time and is a big project. The big question is will the companies actually do something in this area. There needs to be a standard the everyone uses so that this works.

How do you fix these issues when most users don’t understand the problem? Give users the option of who should have access to their data. This will let people do what feels comfortable to them. It really comes down to functionality to users and what they want to do. Also users need to be a bit more savy when their online presence is concerned. Be aware of what you want online and what you don’t.

How does privacy and portability effect business models? Most people expect that if they can add things to a site that they can pull them out. Business models that take this into account will be more effective. I think of the walmart photos and I can upload them and then download the original as well. We need to do a better job of this on WorldVitalRecords.com. If a user sees they can upload a gedcom file and can’t download their work it may make people not even start their family tree.

What are some things companies can do now to help privacy and data portability? Let users link to external sites. Make sure you can produce and consume the content you create on your site. Use meta data from the web like google profile data if you can. Dont restrict user from doing the things they want to do. How do you manage identities on your site? OpenID is a good way to do this. Yahoo has given 250 Million openids to all their users.

Many challenges of apps and widgets

  • difficult to monetize
  • low barriers to entry
  • metrics and analytics inconsistencies
  • spammy
  • cost per installation
  • disposable and low value
  • copy cat apps. many are the same

How do you measure the ROI of a widget or with a distributed app? Many things cant be measured and the items that can be measured are hard to quatify the ROI. No direct correlation between ad spend and revenue yet.

How do you protect your brand with widgets? This can be difficult because users can add widgets to any site. Ultimately the consumer owns the brand. You cant control where the widget can be placed. So just let customers promote your product.

Viral widgets are very hard to build. That is one of the big problems with widgets. So how to you promote your widget? You have to reach out to users where they are looking for widgets. I just installed netvibes.com today and started playing with it.

How do you make money with widgets? Social app market has matured quickly. Widget space has grown more slowly but continues to have a huge reach of users. If you distribute your widget with the idea that it is an ad you can see success. One model is
CPI (cost per install). This is how many apps and widgets are making money now. The problem with this is currently apps are promoting apps and it doesnt seem sustainable. One comment was this is VCs paying VCs. Most cpi companies want to get to more of a cpm model. Ultimately it comes down to the developers coming up with unique revenue models.

What makes a good widget? Understand from end to end what a widget campaign entails. From conseption to construction and distribution to analytics and tracking on the back end. Make sure the entire thing is well defined before you start. Start with an objective. Is it to make money or to get brand awareness or to install on their page. Identify who you are and who your audience is before you start. Then try and identify and try and connect with your audience.

What are the biggest mis conceptions with widgets? People assume virality. Make sure that virality is truly built in. Be true to who you are and what you do. Ebay is a good example in that their facebook widget was a flop but using widgetbox has been successful. Because the sellers want to put the info on their own sites. If you build it they will come doesn’t always work. Also don’t mix desktop and social networking functionality. Follow the “does this make my day” test. Does it put a smile on someones face? If so then it may be viral. Have real expectations when you start.

Panel with meebo, friendster, netvibes, beebo and is moderated by mashery.

Tomorrow netvibes launches their new social network called ginger. They also use uwa (univeral widgit application).

Why are there so many platforms and do we need them all? There is not alot of overlap between the different platforms. Check out the la monde graph that shows the popularity of the different social networks. Social networking is still new and there is now winner yet. The problem I see is that all these platforms are saying they are open but then they have their own custom features and api sets. I get the feeling that a lot of the openness in these companies is all talk. We will support open social, “some day”.

The platforms all seem to be promoting this idea that they want to build a “social ecosystem”. This means that they understand that for them to succeed they need both a good experience for the users and a good experience for the developers. Including a way to monetize the traffic of the developers. This is great for app developers like me. They also know that apps that do bad things hurt the platform and all apps in general.

The idea of having a news feed that is in reverse chronological order is the foundation of how you navigate the social network. The feeds provide an easy way to see exactly what is happening with your friends and when.
Friendfeed. Share things you find on the web with your friends.

What is the benefit of feeds to the user? Re broadcasting what you are doing to others. Feeds bring the ability so share your life. A blog is getting more complex and is becoming more dynamic incorporating things like twitter, diggs, livejournal, google notes ect. Almost more of a life feed. The flow of the feed helps people put the information in context. I need to add some of these widgets to my blog.

There are some risks with feeds. If it is very easy to make something that is personal, like an email thread, into a public feed then someone will make a mistake. Once private info is public you cant take it back. The user should have control of all their info and whether is it public or private.

One problem is if you have multiple systems that are agregated into a single feed how is each unique item identified. An example is installing the twitter app and the friend feed apps you get 2 copies of the twitters in facebook. How does each one know to add an item to a feed or not.

Syndication formats are improving and more things are types of data are getting added to the feeds.

How do you make money with feeds? Friendfeed just closed a 5 Million dollar round of funding. The combination of social context and content is what got friendfeed their funding. There is value this combination. You can dynamically add affiliate links to the feeds to make money but you need to be careful with this and make sure that the content is best for the user and not for the business. Mybloglog only provides the last few items through the feed. If you want the history you have to go back to the actual site and that is how the monetize that traffic.

Presented by Ian Kennedy from Yahoo – This is a widget that tracks which sites you go to and shows others who has been there. You can also become a member and it will track your usage and give you a homepage to view your activity. There is a new “with me” feature. This allows you to easily embed community features into your site.

I like their tagline : you are what you feed.

They have a new API for developers to get information from mybloglog. An example is Raven. Takes info from mybloglog and builds a page about it. m.mybloglog.com reads my bluetooth ID and tags you with it. It then will tell you what other mybloglog members are around you.