Bruce MacVarish Notes

Twitter Graph Relevance

John Batelle provides further proof of the value found at the intersection of network graphs and conversation analytics.   

We now have a number of graphs defined by aggregating links and conversations around social objects: web graphs ( e.g. Google), social graphs (e.g. Facebook), interest graphs (e.g. twitter), location graphs (e.g. foursquare).  

"In short, Twitter knows who you are connected to, and what you (and they) are interested in.

Fashion that into a graph - the same kind of graph that powered Google's graph of web links, or Facebook's social graph - and all of a sudden you have a pretty powerful organizing principle for relevance in the Twitterverse."

John Batelle, Twitter's "Public Interest Graph"

We will see more network graphs being formed around a number of new and different social objects.  In addition, new innovative ways to derive relevance from the social conversations across various network graphs will continue to emerge.  

As MG Siegler points out in his recent take on the Facebook vs. Twitter battle...

"Social networks ... lose because they fail to innovate. Because they fail to stay ahead of the curve."

Expect to see continued innovation focused on ways to aggregate and analyze real-time conversations across network graphs... and... a consolidation across the layers of network graphs.

The Live Web of Interaction Signals for Business Users

Over the next 3 years, the ability to interact, analyze and filter our stream of interaction activities based on a live or real-time model will redefine our personal, work and mobile experiences with the web.  

Let's look at these two categories (i) the Live Web of real-time streams and (ii) Interaction Signals separately:

The Live Web of Real-time Streams for Business Users

The real-time theme continues to grow and it will increasingly define a new class of web.   While it has taken some time to get established, it is not a new idea.   With a hat tip to Stuart Henshall, I've found it very helpful to read the early insights and writings of David Gelernter  .    

 

Professor Gelernter is a Computer Science Professor at Yale and his research centers on information management, parallel programming, and artificial intelligence.  As early as the 80's, Professor Gelernter envisioned managing information based on a time paradigm and through a decentralized data storage system.  He is credited with setting the foundation for today's time based activity stream and cloud computing models so prevalent in Twitter, Friendfeed, Facebook, Salesforce.com, Google apps and a number of start ups.

In this 2001 interview, I like how Professor Gelernter describes the impact of moving from a space ( or file) based computing model to a time ( or activity stream) based model:

When I look at where software is heading and what it is really doing, ... what's happening and what will happen with the emergence of a new generation of information-management systems, and as we discard Windows and NT ... these systems that are 1960s, 1970s systems on which we rely today, we'll see a transition similar to what happened during the 19th century, when people's sense of space suddenly changed.

 

If you compare the world of 1800 to the world of 1900, people's sense of space was tremendously limited and local and restricted in 1800. If you look at a New England village of the time, you can see this dramatically, everything is on site, a small cluster of houses, in which everything that needs to be done is done, and fields beyond, and beyond the fields a forest.

People traveled to some extent, but they didn't travel often, most people rarely traveled at all. The picture of space outside people's own local space was exceptionally fuzzy. Today, our picture of time is equally fuzzy; we have an idea of our local time and what happened today and yesterday, and what's going to happen next week, what happened the last few weeks, but outside of this, our view of time is as restricted and local as people's view of space was around 1800. If you look at what happened in the 19th century as transportation became available, cheap and ubiquitous, all of a sudden people developed a sense of space beyond their own local spaces, and the world changed dramatically.

What's going to happen, what software will do over the next few years has already started to happen and will accelerate... our software will be time-based, rather than space-based.

 

We'll deal with streams of information rather than chaotic file systems that are based on 1940s idea of desks and file cabinets. The transition to a software world where we have a stream with a past, present and future is a transition to a world in which people have a much more acute sense of time outside their own local week, or month in which they now have a clear idea of what was different, why February of 1997 was different from February of 1994, which most people today don't have a clear picture of.

With the introduction of activity streams, a business user, a department, an enterprise will have a more acute sense of time and how things have changed over time.  They will have a clearer idea of what is happening now... live... in real-time... with the people, topics and activities most important to them... and what is different between this week vs. last,  this month vs. last, this year vs. last ... across people, projects, customers, product lines, businesses and markets.   This is an incredibly powerful shift.  

 

Interaction Signals for Business Users

The shift to the live web is underway in our personal (consumer) lives and will ... as with other consumer web innovations ... be adopted and integrated within our work (enterprise) lives.    If we describe the enterprise live web as a flow or activity stream of information tied to our live transactions, conversations and relationships ... then a number of new, live, time-based signals or metrics will be needed to help us optimize how we allocate time and attention in our daily interactions.

I think there are 4 classes of signals or metrics that are emerging to help business users understand how their interactions - and by interactions, I mean their transactions, conversations and relationships inside and outside the enterprise - are changing over time and will improve their productivity by embracing the Live Web. They include:

Interaction Activity Signals track the number, rate of change and scale of interaction activities for a business user over time.  Examples include:

• number of posts, emails, meetings, conversations, orders, tweets, cold calls over time
• count of views, votes, comment, responses, feedback, re-tweets over time
• active, inactive subscribers / followers over time
• total inbound/outbound connections, interactions, links over time
• number of unique interactions over time

Interaction Role Signals track the involvement, collaboration, leadership and consistency of roles that business users play through interactions over time.  Examples include:

• pattern and pace of interactions over time
• degree of replication and emulation during interactions over time
• variability in emphasis, rhythm and timing of interactions over time
• degree of openness to new ideas and influence over time
• inbound-to-outbound interaction ratio over time
• positive-to-negative interaction ratio over time

Interaction Influence Signals track the diversity, reach and influence of interaction activities for a business user over time.  Examples include:

• diversity of business user social graph in the enterprise over time
• conversation to broadcast ratio over time
• conversation reach across the enterprise over time
• degree of reference / Link over time
• diversity of reference / Link over time
• influence of your followers over time
• influence of your referencers over time
• new conversation generation over time

Interaction Results Signals track individual business user, team or enterprise deliverables resulting from interaction activity over time.  Examples include:

• new transactions (orders, contracts, demos, investment, etc. ) over time
• conversation-to-conversion ratio over time
• relationship-to-conversion ratio over time
• correlation ratios between interaction activity, role, influence and results metrics over time

As the Live Web of activity streams and interaction signals emerge in the enterprise, one could say we're about to step out of the quaint New England village of the 1800's as described by Professor Gelernter.   

The enterprise is shifting from a space/file software model to a time/stream based software model where the real-time interactions of business users have a past, a present and a future.   

New value for business users - and revenue for enterprises - will come from aggregating, analyzing and filtering these interaction signals enabled by the Live Web.

"Sensing" Voice

In my vision of future state for enterprise applications, the 6th element is "Sensing".   

I've used this term to capture how future applications will create new value for users by sensing relevance, context and personal preferences through analytics of voice, video, text, location, attention or other ambient and declarative data from the user.  The ability to capture, store, index, search and analyze voice recordings is fundamental to this future state vision.

Nuance, BBN, TellMe/Microsoft, Nexidia, CallMiner, Utopy, SER, IBM and others have invested to improve STT, TTS, ASR and Speech Analytics technologies that are all critical to this "sensing" end-state.

Recently, Microsoft announced at the Mobile World Congress in Barcelona - Microsoft Recite -  a Voice capture and search application for Windows Mobile devices.

To get a sense of the UX and VUI, check out this video clip...

 

With Recite users can record voice messages and then, based on a voice interface, search for specific terms or phrases to find earlier messages... and I suspect with time... earlier conversations. 

It appears as if Recite uses some type of voice pattern matching or phonetic search engine.  There is not translation from speech to text and the accuracy improves with longer search phrases.  Both of these characteristics points to phonetic processing.  

You can download the app here.

User Experience for the Business User in 2009

Four fundamental forces of change are driving the evolution of the user experience for the Business User in 2009:

(a) economic recession and its impact on the enterprise IT budget 

(b) enterprise governance, security and business policy requirements force "hardening" of consumer apps

(c) new sources of value and revenue continue to be found at the Enterprise Edge

(d) consumer innovations influence and shape business user expectations (see below)

Based on the above, I've outlined some initial thoughts on how the user experience for the Business User will evolve in 2009:

1. Acceleration of Enterprise SaaS and Cloud Computing: The economic downturn will encourage more businesses to explore both cloud and hybrid (premise + cloud) computing models.  As a result, the Business User Experience will improve with seamless integration between apps behind the firewall and those in the cloud.  Watch for hybrid applications that create a seamless premise + cloud experience for the business user.

2. Broader use of Voice interfaces: ASR, Speech-to-Text and Text-to-Speech will continue to emerge as primary interfaces to mobile and web applications.   

3. Visualization continues to emerge: As more and more user data is collected and integrated into business user search, filtering, communication and collaboration applications, visualization will simplify insights and analysis of user activities to improve user interaction and experience.

4. Business Apps look like Consumer Apps:  We will continue to see the blending of consumer and business user experience.  Business users will increasingly look for experiences that are shaped and informed by their experience with consumer applications.  From Search to Create, Communicate and Collaborate apps, business users will ask for experiences similar to what they experience in consumer applications.  

5. Move from Unifying to Threading & Filtering:  The last few years have seen services and applications that aggregate user content from across various networks and communities. FriendFeed, Facebook, Meebo and Tumblr are examples on the consumer side; Unified Communications and Messaging on the enterprise side New value for the Business User will be found at the next step beyond aggregation and unifying.  During 2009 we will see improved "threading" of conversations and relationships... and improved filtering... enabling Business Users to find, create and share the most important and contextually relevant conversations and relationships.

6. Business Users join the Flow:  Twitter-like applications that provide for near real-time communication, collaboration and sharing across networks of followers or subscribers will shape the experience for Business Users.  These flow apps will be device agnostic and sync across mobile, web and enterprise applications to provide business users with constant access to the flow of conversations, relationships and transactions within the enterprise

Mobile "Sensing" for Context-Aware Application

From Nokia's Conversation blog, we learn about their vision of turning mobile devices into sensors.  

For me, it serves as an example of Nokia's interest in the "Sensing Wave".

"Every phone is a sensor 
When you start thinking of the mobile device as providing context, you start to think of users projecting who they are, what they are doing, and where they are. But, when you begin to think of what could mobile devices do anonymously, you can see cases where sensor-filled devices can, in aggregate, provide information on traffic, weather, or health patterns."

Most people see how the mobile phone could power context-aware applications by aggregating, analyzing and acting on information "sensed" by the device.  

Here, Nokia highlights how anonymous data could provide context-aware applications for the collective.   In addition, capturing personal data for the individual user will provide context-aware applications and reasoning engines that are critical to helping each of us intelligently filter media and conversations to those that are the most relevant to our situation and preferences.

We're at the early stages of the "Sensing Wave".  Watch for more context-aware applications to emerge that "sense" based on data aggregation, analysis and acting.

Google Mobile = Chrome + Android + Cloud

Scott Webster of Android Guys highlights what I've been thinking.  Google Android may be aimed initially at Microsoft or Apple, but it really is a first step to a larger Google Mobile vision that uses Chrome, Android and Cloud based apps to drive new growth. From Scott:

"Alot of people think that Google’s mobile agenda stops at Android.  I think that this is where it begins.  Android is just a tool to help further things at a faster rate.  The boys from Mountain View are already putting out applications for other platforms like iPhone and Windows Mobile.  Android simply takes everything that Google offers and bundles it into a convenient package.  Why download all the individual apps and services when they come preloaded on a revolutionary handset?

It’s no secret that the “cloud” is where things are headed.  No longer being confined to a desktop or laptop, people are free to access their files from practically anywhere in the world, providing an internet connection is available.  Pull up your spreadsheets, vacation photos, and favorite music all from the same device, regardless of operating system.  Google has slowly been going cloud on us for the last few years with things like Google Docs, Picassa, and Gmail.  Android is the means to the end."

Watch for v2 of Android and more importantly, Android based devices and applications that fulfill the Trivergence model to move us all toward the Google Mobile Vision.

The Sensing Wave of Opportunity

Recently, I found myself considering three different proof-points of how we are entering the "Sensing" wave of innovations and technologies as described in the 10 year Forecasts published by the Institute for the Future ...

Proof-point #1:  Sensing Filters

The first proof-point is Clay Shirky's presentation at the Web 2.0 NYC where he made the case that we're not suffering from Information Overload but Filter Failure.   For filters to be successful, they need to "sense" what is important to me by becoming social-aware, context-aware, relevance-aware and location-aware.  Filters need "sensing" intelligence to correctly filter the flow of transactions, conversations and relationships that are important to me.   Here's Clay's presentation:

Proof-point #2: Sensing Signals

Sandy Pentland at MIT has a new book out titled Honest Signals.  I met Sandy a few years ago while doing some work with the MIT Media Lab.   At that time, Sandy was in data gathering mode... yet you could see the value of his research as it related to enterprise communications, collaboration and customer service applications. The new book outlines Sandy's findings from 5 years of research on how individuals communicate (send signals) beyond the conscious conversation... and the value of "sensing" those signals.

Here's an introduction summary:

"How can you know when someone is bluffing? Paying attention? Genuinely interested? The answer, writes Sandy Pentland in Honest Signals, is that subtle patterns in how we interact with other people reveal our attitudes toward them. These unconscious social signals are not just a back channel or a complement to our conscious language; they form a separate communication network. Biologically based "honest signaling," evolved from ancient primate signaling mechanisms, offers an unmatched window into our intentions, goals, and values. If we understand this ancient channel of communication, Pentland claims, we can accurately predict the outcomes of situations ranging from job interviews to first dates.

Pentland, an MIT professor, has used a specially designed digital sensor worn like an ID badge--a "sociometer"--to monitor and analyze the back-and-forth patterns of signaling among groups of people. He and his researchers found that this second channel of communication, revolving not around words but around social relations, profoundly influences major decisions in our lives--even though we are largely unaware of it. Pentland presents the scientific background necessary for understanding this form of communication, applies it to examples of group behavior in real organizations, and shows how by "reading" our social networks we can become more successful at pitching an idea, getting a job, or closing a deal. Using this "network intelligence" theory of social signaling, Pentland describes how we can harness the intelligence of our social network to become better managers, workers, and communicators."

Proof-point #3: Sensing Worker Effectiveness

Michael Moaz at Gartner highlights how aggregating and analyzing an employee's use of company and public applications will be used to sense the effectiveness of workers.

"What happens to this flotsam of data about us? Maybe nothing. Or maybe software systems begin to monitor the amount of time we spend on all of the media in our work lives, in the same way that your phone minutes are measured, your time in an application are measured at work, or your time in the office, or time stamps are captured for your credit card purchases. It isn’t too hard to project when these systems will evolve and become a standard part of your evaluation as an employee. I am employee 6746 with my company, and this week I spent 12 hours outside of the corporate productivity systems. Next we monitor how any of this time tracks back to achieving each of the metrics set for me in my capacity as an employee. Were you more productive meeting business goals? More efficient? If you were effective and efficient, it will be great. But what if you weren’t? Then how will your time invested in the social network be viewed?"

I believe the emerging wave of "Sensing" solutions and innovations will create new, fast growing companies and whole new industry segments just as the "Computing" Wave ( think Mainframe, Midrange, PC, Software) and "Connecting" Wave ( think networking, routing, switching, LAN, WAN, Wireless, Mobile) did over the last 25+ years.

... So watch for social-aware, context-aware, relevance-aware, attention-aware, location-aware, activity-aware technologies and related analytics applications as further proof-points of the growing "sensing" wave of opportunity.

The Future Desktop will be Flow-based and Attention Analytics driven

Flow based tools like twitter, identi.ca and Workstreamer are great examples of how applications and desktops will evolve to help end-users manage and analyze how their attention is being allocated to the various information streams they believe are contextually important and relevant to their work and personal lives.

Along this line of thought, Nova Spivack of Twine offers his vision for the future of the desktop and the role that flow-based, attention analytics will play.  

"The focus shifts from information to attention

As our digital lives shift from being focused on the old fashioned desktop (space-based metaphor) to the Web environment we will see a shift from organizing information spatially (directories, folders, desktops, etc.) to organizing information temporally (river of news, feeds, blogs, lifestreaming, microblogging).

Instead of being a big directory, the desktop of the future is going to be more like a Feed reader or social news site. The focus will be on keep up with all the stuff flowing through and what the trends are, rather than on all the stuff that is stored there already.

The focus will be on helping the user to manage their attention rather than just their information.

This is a leap to the meta-level. A second-order desktop. Instead of just being about the information (the first-order), it is going to be about what is happening with the information (the second-order)."

GNIP Data Innovation: Aggregating Social Web Data

A few days ago, I used the example of Peer39 to illustrate my belief that just as hardware and software innovation drove new growth over the past 20 years... over the next 5+ years, new value creation will increasingly come from Data Innovation.

Here I'm thinking about data innovation from a technology, customer and business model perspective when aggregating, indexing, threading, analyzing, sharing and visualizing data.  

The announcement of GNIP is a great example.

Today, Eric Marcoullier of MyBlogLog fame (acquired by Yahoo) has launched a new web service called GNIP that aggregates, translates and notifies users of new social data.   GNIP describes how it can dramatically change the translation, distribution and protocol translation of web data messages to improve application performance and efficiency by removing the REST polling of the social web. In their words:

Today, 3rd party data consumption means...

• API throttling & high latency

• Competing protocols

• One-off data integrations

• Identity discovery challenges

• More grunt work than any of us like to do

None of these issues are insurmountable, most are not even complicated. They're hassles. An exponentially growing set of hassles that distract you from the features of your service that your users actually care about. Gnip is here to help. We make it easy to aggregate user data so that you can focus on doing something cool with it

GNIP also offer the ability to translate from one protocol to an another through a "protocol bridge" so that consumption of the aggregated data is efficient and simplified.

 
I really like the assessment and vision that Marshall Kirkpatrick of RWW outlines for GNIP.

"Gnip wants to serve as the grand central station and universal translation service for the new social web."

and a little bit later...

"...we think that Gnip has a whole lot of potential to deliver huge value to the applications leveraging it, to the backers financing it and ultimately to all the users of the emerging class of social web applications. We love this kind of stuff."

(Enterprise) Relationships are Complicated

Chris Messina has been posting his thoughts on how to improve today's "social" web to make it easer to find and communicate with our real-life network of family and friends that are increasingly accessible online. 

"Put another way, it’s not good enough to simply dismiss the trend of social networking because our primitive technological expressions don’t reflect the complexity of real human relationships, or because humans are just one of kind of “object” to be “semantified” in TBL’s “Giant Global Graph“… instead, people are connecting today, and they’re wanting to connect to people outside of their chosen “home” network and frankly the experience sucks and it’s confusing. It’s not good enough to get all prissy about it; the reality is that there are solutions out there today, and there are people working on these things, and we need smart people like Greenfield and Berners-Lee to see that solutions that enable the humanist web (however semantic it needs to be) are being prioritized and built… and that we [need] not accede to the heedless restructuring of everyday human relations on inappropriate and clumsy models derived from technical systems."

...and later...

"Human relationships are simply far too complicated to be left up to assumptions and inferences made by technologists whose affinity oftentimes lies closer to the data than to the makers of the data."

It is interesting to think about this same challenge from the perspective of the enterprise where relationships in our work oriented 'social network' are as complicated as any in our personal life.   Solutions for improving the 'social web' as outlined by Chris will need to apply across both.