Posts Tagged ‘ twitter ’

Twitter and the KM Context

[W]e came across the word “twitter,” and it was just perfect. The definition was “a short burst of inconsequential information,” and “chirps from birds.” And that’s exactly what the product was.
– Jack Dorsey (Sarno, 2009)

Twitter, the popular microblogging tool from which users post updates in 140 character increments, recently celebrated its five-year anniversary. In the world of fly-by-night Web 2.0 applications, that makes it a well-established and time-tested social technology. What has contributed to Twitter’s success? Why is it such a popular tool?

As its co-founder, Jack Dorsey, suggests in the quotation above, Twitter is a place where users can publish short bursts of information and share it with a larger community. It is “the best way to discover what’s new in your world”, reads the website’s about page ( Still, users unfamiliar with the platform or dubious about this claim might wonder precisely how this tool can be productive. After all, Dorsey’s endorsement is not exactly inspiring: what good is information if it is inconsequential? What makes Twitter such a powerful tool, from both a knowledge management or business perspective and the broader context of information-sharing is how it operates in real-time. It allows members of communities of practice to track relevant news and share important events as they happen. This crowdsourcing approach to information means that users who follow other users publishing information relevant to their community of practice can keep their finger on the pulse—an extremely valuable commodity in a world that is increasingly knowledge-centric. Similarly, these users can participate in a live, public conversation within a global network of peers, encouraging an ongoing exchange of knowledge. More importantly, the simple premise of “following” (or, in other words, subscribing to user feeds) allows complete personalization, while creating links between users that shape community networks organically, rhizomatically.

Another advantage of Twitter is that it is highly scalable. Twitter has an API (Application Programming Interface) that allows customized software to be built around the basic platform. In this way, users can log in to their account using third-party software like TweetDeck, which allows them to organize and view tweets in a variety of ways. In addition, this characteristic also allows the development of widgets to publish tweets on websites and blogs. Viewed as much as a disadvantage as an advantage, the 140-character limit on updates forces users to state a single idea clearly and concisely. This limitation was originally due to the average character limit for text messages from cell phones, which had been considered by the founders as the principal technology for using the service. Soon after the service went public, however, most smart phone models no longer had that limitation on text messages. By then users had discovered that the character limit was the ideal length for short status updates; the limitation distinguishes Twitter from other casual blogging services such as Tumblr, which, no doubt, helped promote the service as a brand. While sometimes inconvenient for users with longer, more elaborate messages, the difference makes Twitter unique as a social media tool.

A definite disadvantage of this technology, as with many social media technologies, is the public nature of updates and the murky notion of intellectual property. Twitter is perhaps more volatile in this sense than other, similar technologies like blogs or wikis, which require more thoughtful consideration before publishing. The brief nature of tweets make it easy for users to submit whatever they happen to be thinking or seeing, regardless of legal considerations such as intellectual property or copyrights, and updates are published immediately without the opportunity to review or delete before they go live. This can be problematic for users, particularly high-profile users; one dramatic example, though certainly not the only one, would be a tweet that resulted in the termination of one CNN correspondent. In 2010, Octavia Nasr was fired for publishing an update expressing regret over the death of the Grand Ayatollah Mohammad Hussein Fadlallah, a leader of Hezbollah. Twitter poses a problem for e-Discovery that courts around the world have not yet come to terms with.

To provide a nuts-and-bolts explanation of how Twitter works and to help understand its practicality, it is useful to consider the following scenario: You are interested in motorcycles, and want current information about promotions, events, and people in your area related to that interest. You create an account on, and search the website for likeminded users. Scanning through user profiles, you decide to follow users representing several motorcycle dealers in your city, a couple motorcycle riding clubs, a national motorcycle news magazine, and a number of individuals who identify themselves as “motorcycle enthusiasts”. You begin receiving these users’ updates (or tweets), and begin to learn about the local motorcycle community. After a few days of reading tweets, you learn that there is going to be a bike show and that several of the users will be attending. You are unable to attend the bike show yourself, but you get to experience it through the tweets of your fellow users, who describe the event and post pictures of different models on display. You are able to engage some of these users, asking them questions about the event as it is taking place. You also discover that there is a hashtag that Twitter users are using to identify tweets about the event, and by searching all tweets that include that hashtag you discover several more users to follow. In this way information is exchanged, and you develop relationships with other members of the community that you might otherwise not have had. Now consider this same scenario in a different context: you have recently opened a motorcycle shop. Using the tool in the same way, Twitter becomes a valuable social tool for promoting yourself or your company, in addition to acquiring and sharing useful information.

Knowledge management (KM) resides in an interesting interdisciplinary space, somewhere between sociology, philosophy and economics. In his 1962 article, “The Economic Implications of Learning by Doing”, Nobel-prize winning economist Kenneth Arrow clearly states the necessity for organizational practices that manage the learning process; the economics of KM are concerned with breaking down and quantifying this process. In The Tacit Dimension (1966), Michael Polanyi describes the concept of “tacit knowing”; knowledge that deals with the implicit nature of human experience, skill and action is considered tacit, while knowledge codified and transmittable through language is explicit. Polanyi’s epistemological model serves as the fundamental principle of KM, distinguishing knowledge from the concepts of information and data. The sociological underpinnings of KM provide us with the a sound basis for understanding “knowledge” as a concept and its notably various manifestations, while also giving us a framework for making sense of how knowledge circulates within communities and through individuals. The seminal work of Emile Durkheim lends KM a primary concern with the “social facts”—the observable behaviours at the root of human interaction. Rather than relying on theory, KM is preoccupied with studying how people actually share, learn, and use knowledge. KM arose from these disciplinary cornerstones in the early 1990s, when an increased emphasis on the creation, dissemination and utilization of organizational knowledge in professional and scholarly literature identified a growing need for a systematic approach to managing information and expertise in firms. Laurence Prusak identifies three social and economic trends that make KM essential in any organization today: globalization, ubiquitous computing and “the knowledge-centric view of the firm” (1002). Prusak’s description of globalization in particular emphasizes the necessity to stay current; information technology has resulted in a “speeding up” of all elements of global trade, as well as an increase in the “reach” of organizations. Twitter is a technology that can facilitate this necessity.

There are any number of examples that demonstrate how Twitter fulfills the requirements of KM that I have described. In terms of leveraging group and individual interactions based on “social facts”, we can consider the role Twitter has played in the recent revolution in Egypt. Protesters on the ground in Cairo were publishing updates about the conflicts they faced, giving the crisis international exposure it might otherwise not have had. Following the government’s failed attempt to block Twitter—evidence in itself as to the effectiveness of Twitter for spreading a message—there was overwhelming support from around the world for the protestors against President Mubarak’s regime. This global support, along with the grassroots reporting of Egyptian demonstrators, certainly contributed to Mubarak’s ultimate resignation from office. This example shows how the knowledge of individuals in a particular context spread to other communities, and how this in turn inspired a global movement—based on the ever-expanding network of interactions through this particular social tool. The “social fact” inherent in Twitter is how human interaction manifests around these short bursts of highly contextual information, and how communities take shape by engaging in the same and other closely related contextual spaces.

An example of how Twitter facilitates the transfer of tacit knowledge might be the way events are recorded and experienced through it. Take, for instance, the recent SXSW Conference and Festival in Austin, TX, a yearly event that is recognized worldwide as a showcase of music, films and emerging technologies; a Twitter search for “#SXSW” reveals a host of users recording their experience through a variety of media—text describing talks, shows and screenings combined with links to photos, videos, and websites that together form an image of the event. These individuals’ experiences might not otherwise be expressible without a tool like Twitter that facilitates the blending of online multimedia. Moreover, the combined force of a community of users sharing these experiences at the same time can provide a comprehensive panorama of what they are hearing, seeing, and learning. In this way, Twitter allows tacit knowledge to be codified for mass consumption.

Measuring the impact of Twitter and how knowledge circulates through the network is not a simple task. Perhaps the most effective way to do so that we have today is the application of web and text analytics to social media. There are several companies that have recently achieved success in this area, based on textual data (e.g. lexical analysis, natural language processing, etc), user data (e.g. demographics, geographic data), and traffic data (e.g. clickstream, page views, number of followers/subscribers, replies and retweets, etc) mined from social media websites. Canadian-based Sysomos has used MAP (Media Analysis Platform) to provide an in-depth analysis of how people, products and brands are effectively marketed through Twitter and other social media tools. One reviewer describes MAP as follows:

MAP can, for example, tell you that the largest number of Twitter users who wrote about the Palm Pre come from California and Great Britain, as well as who the most authoritative Twitter users who tend to tweet about the Pre are (MAP assigns a score from 1 to 10 to every Twitter user, based on the number of followers, replies, retweets, etc.). Of course, you can then also compare these results with results from a query for ‘iPhone,’ for example. (Lardinois, 2009)

MAP, in fact, was used for an analysis of users during the crisis in Egypt. Some of the visualizations of this data are available online[1] . A recent study comparing social media monitoring software identified five key categories that need to be considered to appropriately measure the effectiveness of a social media tool (FreshMinds Research, 2010):

  1. Coverage – Types of media available based on geographic coverage.
  2. Sentiment analysis – The attitude of the speaker/writer with respect to the topic, based on tone.
  3. Location of conversations
  4. Volume of conversations
  5. Data-latency – The speed at which conversations are collected by a tool, based on the frequency of its web crawlers and the length of time it takes the monitoring tool to process the data.

As the researchers who undertook the study indicate, the possibilities for such data, from both a qualitative and quantitative perspective, are “huge”. Social media monitoring allows us to examine any number of factors in the learning and communicative process as it is manifested through social media technologies, “from category choices to the lifestyles of different segments”, on an individual or at an aggregate level (ibid.). The research group also identifies areas in which social media monitoring needs to improve—particularly within the realm of sentiment analysis. The monitoring tools are not sophisticated enough to provide an accurate measure. While Twitter in itself can be thought of as an organizational practice for knowledge-sharing, the application of monitoring tools can be thought of as Arrow’s organizational practices for managing knowledge. Based on the analysis that such monitoring tools—like Sysomos’ MAP—can provide, organizations and individuals can make more effective use of Twitter.

It is clear that Twitter can be a huge benefit for the effective creation and dissemination of knowledge, if used correctly. Organizations that are prepared to invest the time and energy in a sound social media plan to improve KM would be remiss not to include a presence on Twitter. On the other hand, this technology poses many risks for organizations, particularly in the realm of e-Discovery. The fact that content published to Twitter resides on the website’s servers, and not in the hands of the organization must play an important factor in any organization’s KM assessment. Twitter is perhaps more useful for NFP organizations that have a mandate for advocacy and public promotion (take, for instance, SXSW). It also is useful for individuals with either a professional interest in promotional or informational knowledge-sharing (such as consultants, agents, performers, journalists and salespeople) or as members of an existing community (like our motorcycle enthusiast). The professional and the social are not easily distinguished on Twitter, which can be both a benefit and a curse for users, as we have seen. Finally, while the information shared on Twitter might seem “inconsequential” to some, to others it can be very valuable. It is this value that KM needs to harness, in order to effectively make use of Twitter.

[1] Visualizations for the Twitter data related to the crisis in Egypt can be found at For a compelling overview of the sort of data Sysomos has analyzed with respect to Twitter, an indispensable resource is their report “Inside Twitter: An in-depth look inside the Twitter World”, 2009:


Arrow, K. (1962, June). The Economic Implications of Learning by Doing. Review of Economic Studies 29(3), 153-73.

Durkheim, E. (1982). The Rules of the Sociological Method, Ed. S. Lukes. Trans. W.D. Halls. New York: Free Press.

FreshMinds Research. (2010, May 14). Turning conversations into insights: A comparison of Social Media Monitoring Tools. [A white paper from FreshMinds Research,] Retrieved on March 22, 2011 from

Lardinois, F. (2009, June 4). Pro Tools for Social Media Monitoring and Analysis: Sysomos Launches MAP and Heartbeat. Retrieved on March 22, 2011 from

Polanyi, M. (1966). The Tacit Dimension. London: Routledge & Kegan Paul.

Prusak, L. (2001). Where did knowledge management come from? IBM Systems Journal, 40(4), 1002-1007.

Sarno, D. (2009, February 18) Twitter creator Jack Dorsey illuminates the site’s founding document. Part I. Los Angeles Times. Retrieved September 24, 2010 from


The Commonplace Book—extinct form of critical reading and sensemaking?

I found Robert Darnton’s chapter on the Renaissance tradition of the commonplace book an interesting insight into how people made—and make—sense of what they read.  It made me wonder about how this tradition of reading has changed over time.  Darnton suggests that today’s reader has learned to read sequentially, while the early modern reader read segmentally, “concentrating on small chunks of text and jumping from book to book” (169).  The implication is that, from this transformation of practice, we have lost a critical approach to reading.  The commonplace book, Darnton describes, was a place where early modern readers collected bits and pieces of texts alongside personal reflections about their significance (149).  This activity was a hybrid of reading and writing, making an author of the reader, and serving as a method for “Renaissance self-fashioning”—the grasping for a humanist understanding of the autonomous individual (170).  Arguably, in adopting a sequential mode of reading and forgetting the practice of the commonplace book, we have lost a useful tool for making sense of the world and of ourselves.

At the beginning of the chapter, Darnton makes a curious allusion to the present reality, the Digital Age.  He writes: “Unlike modern readers, who follow the flow of a narrative from beginning to end (unless they are digital natives and click through texts on machines), early modern Englishmen read in fits and starts and jumped from book to book.” [Emphasis is my own] (149). Clearly he is referring to hypertextual practice, the connective structure of texts on the Web that are joined through a network of inter-referential links, and provoke a non-sequential mode of reading.  The Web has initiated a number of changes in how we read, write, create and make sense of texts.  Hypertextuality is certainly one them, but I think Darnton only touches upon the tip of the iceberg with this passing reference.  While the commonplace book as genre might be extinct, new hybrid forms of critical reading/writing have taken its place.  Take, for instance, the blogging phenomenon.  Many people today write blogs on a vast variety of subjects.  Most represent critical responses to other media—articles, videos, images, stories, other blog posts.  They are the commonplace book of the digital native.  The difference is that the digital native’s commonplace book is accessible to all, and (more often than not) searchable.  Consider also the phenomenon of microblogging in the form of Twitter.  As an example, I am going to look at my own Twitter feed ( – I have attached a page with specific examples).  In 140 character segments I carry on conversations, post links to online documents and express my reactions to such texts.  It is, in fact, perfectly possible to consider a 21st century individual’s Twitter feed analogous to the early modern reader’s commonplace book.  These activities represent a far more complex mode of reading than Darnton assigns the contemporary reader.  It is a type of reading that is at times segmental, at times sequential, but is remarkable because of the interconnectivity of sources and the critical engagement of the reader that it represents.  What is most interesting is that, rather than emphasizing the notion of the autonomous individual, these digital modes of reading/writing emphasize collectivity and community—what could be described as a “Posthuman self-fashioning”.


Darnton, R. (2009).  The Case for Books. New York: PublicAffairs.  219p.


I have not include the appendix of selected tweets that was submitted along with this assignment, but I’m sure you’ll get the gist by viewing my Twitter page:

Twitter and the challenge of electronic records

Twitter, like email and instant messaging, is an emerging form of electronic record that poses many risks and challenges in terms of records management.  It is perhaps more risky and more challenging than its precursors, and the management issues are complex on a level that most individuals and organizations are currently not prepared to deal with.  However, thanks to examples such as Paul Chambers’ “obviously facetious” bomb threat [1], the (unsurprising?) outing of John Baird [2], and the CNN journalist who was fired for expressing regret about the death of the Grand Ayatollah [3], we are slowly becoming aware that tweets are public records, visible to all, and can have serious consequences for individuals and organizations when used inappropriately.

In a large organization email and IM records are stored on a server typically maintained by the organization itself, which means that the organization has some control over the disposition of the records.  Twitter, on the other hand, is a service provided to individuals and businesses to share news, updates, advertisements and reflections with a broader community.  Individual tweets are stored on the service provider’s (Twitter, Inc.) servers, and it is under no obligation to the organizations of individuals who use the service.  Twitter only indexes tweets for six days [4], meaning that older tweets cannot be searched (although at least one third-party company has since begun to save and index tweets to make them searchable over longer timelines. [5]) In April 2010, the complete collection of all public tweets, since the service first became available to the public in 2007, was provided to the Library of Congress in order to create a publicly accessible archive [6].  That project is currently under development.  The bottom line is that organizations and members of organizations must be vigilant about what they publish on Twitter, because once it’s been published there is no taking it back.

Founder Jack Dorsey said the following about the inspiration for “Twitter”, which strikes an odd contrast with the negative fallout from the cases I previously mentioned:

[W]e came across the word “twitter,” and it was just perfect. The definition was “a short burst of inconsequential information,” and “chirps from birds.” And that’s exactly what the product was. [7]

Unfortunately for Paul Chambers, Glen Murray and Octavia Nasr, information available for public consumption with some form of permanence, whether intended as such or not, does not always remain ‘inconsequential’.  In the case of Paul Chambers specifically, we witness an example where the record of the tweet is maintained as sufficient evidence in a court to convict.  It serves as a warning for those of us who have a heavier digital footprint through the use of social media, and reminds us that electronic records continue living long after we have forgotten about them.


[1] Booth, R. (2010, September 24) Twitter joke trial: bomb threat ‘obviously facetious’. The Guardian. Retrieved September 24, 2010 from

[2] Green, J. (2010, September 18) Baird’s dig at elites a bit rich, Murray says. Ottawa Citizen. Retrieved September 24, 2010 from

Capstick, I. (2010, September 18) John Baird is really gay.  MediaStyle. Retrieved September 24, 2010 from

It’s sort of a running joke in Canadian politics that Conservative MP and current Government House Leader John Baird is a closet homosexual; that is, he has been widely known to frequent gay bars and has even been “outed” previously to the press by a fellow party member.  While not openly gay (he has refused to discuss his sexuality with media), a number of members from opposing parties and the press alike have pointed to a wealth of circumstantial evidence to the fact.  The controversy lies primarily in the Conservative party’s official position on same-sex marriage and the incongruities of the PC platform with LGBT culture and values.

Ian Capstick is suggesting in his article that the tweet written by Ontario Liberal cabinet minister Glen Murray is implying something about John Baird’s sexuality, as well as indicating that he happens to be one of the “Toronto elites” Baird recently accused of forcing the Long Gun Registry (a policy the Conservatives are attempting to reject).  This puts Murray in a mildly uncomfortable position, but he will not face any serious repercussions.  It is really just one more episode in this mostly insignificant narrative about John Baird.

[3] Anonymous. (2010, July 9) CNN journalist fired for controversial twitter message. Retrieved September 24, 2010 from

Octavia Nasr was fired after backlash from her tweet expressing regret and respect for Grand Ayatollah Mohammad Hussein Fadlallah, a leader of Hezbollah, following his death on July 4, 2010.

[4] This information was acquired from a Twitter representative in response to a trouble ticket (#1044971) I filed on June 24, 2010.  From @dino: “Hello, Yes, our search currently only indexes about the most recent 6 days of Tweets. Longer-term or historical search is a feature request we see a lot, but there’s no current timeline for this feature. I’m sorry about that! Thanks, dino”


[6] Raymond, M.  (2010, April 14) How Tweet It Is!: Library Acquires Entire Twitter Archive. Library of Congress Blog.  Retrieved September 24, 2010 from

[7] Sarno, D. (2009, February 18) Twitter creator Jack Dorsey illuminates the site’s founding document. Part I.  Los Angeles Times.  Retrieved September 24, 2010 from

Living on a fish island…

((Note: I started writing this post last Friday and then just forgot about it.))

I had the strangest dream ((last)) Wednesday night.  I dreamt I was in a bubble under the sea.  Atlantis.  The décor was that of a 5-star hotel.  The Hilton Aquatique.  The soundtrack for my dream was Rush’s Limelight.

Ever since that dream I can’t stop listening to Rush. ((Well, I’ve moved back to Iron & Wine, now))

Let’s do a closing of the tabs, à la Neil Gaiman.

#1- So Archie is tying the knot.  As one twitterer put it, “it’s a sad day for lovers of crappy comics everywhere”.  Let’s be honest though.  Archie isn’t complete failure, he has some decent academics, some better athletics, he could probably make it into a nice college.  But the Andrews are your typical middle-class family, and Riverdale’s been hit just as hard as anywhere else in the US by the current economic downturn.  The Lodges, on the other hand, pretty much own Riverdale.  Archie’s just being practical.  His fiancée’s father has the connections and the money.  Not only could he get placed in a fancy school, but he might finally trade in that old jalopy for a reliable, sporty-looking vehicle.  A hybrid.  Or an SUV.  Whatever.  He’s making the smart choice.

#2-Bill #44 Twitfest.  In general, I think Twitter is a great medium within which to introduce items for political debate.  Note the following observations made by various provincial politicians: Continue reading

‘New York’ Talks Twitter

Biz Stone and Evan WilliamsThis article in New York Magazine is worth a read, whether you’re a Twitter devotee, or a microblogging cynic (as a former member of the latter and a current member of the former, I can say it presents a pretty balanced picture of what Twitter is all about.) Thanks to mastermaq for tweeting it.

New York | How Tweet It Is

… From Twitter’s initial public debut as the best way to find the parties at South by Southwest in 2007, we’ve gone from hackers taking over Barack Obama’s and Britney Spears’s feeds to Republican operatives spending their post-election-malaise retreat bragging about who had more followers, to the Mumbai attacks, when users trapped in the Oberoi Hotel were transmitting messages that chronicled the ongoing madness. Twitter executives are proud of the Mumbai aftermath; Forbes called it “Twitter’s moment,” and Stone’s face lights up when it’s mentioned. “Twitter is not about the triumph of technology,” Stone says. “It’s about the triumph of the human spirit.”

I posted about Twitterers taking advantage of Twitter to communicate during the T.O. blackout a couple weeks ago. It’s a minor example compared to Mumbai, or the most recent unrest in Gaza (which has also developed a vast community of Twitterers). But for something less than two years old, and such a basic premise as fostering a culture of sharing in bite-sized bits that fit in a single text message, it’s surprising the kind of impact it has had so far, and continues to have. There’s a lot of people who study these forms of social media that look at Twitter and see something that should have been obvious– to the point of being mundane– revealed to be the next big evolution in how to interpret and disseminate information.

The reporter, Will Leitch, goes on to talk about how a Twitterer named Krums was the first to witness the crash of US Airways Fligh 1549 as it happened, and how Twitter became ground-zero for that particularly news item even as the New York Times and other news agencies scrambled to cover the story. Krums got a shot of the plane crash and his first thought was to share it on Twitter. And interestingly enough, as Leitch later discovers, Twitter didn’t even notice a spike in traffic.

“That’s only for huge shared experiences, like the inauguration, or Mumbai.” Twitter had unleashed something … and its executives were completely unaware, as its system worked on its own, without them.

#tcot presiding over Twitter, #tpot eating digital dust

Top Conservatives on Twitter

To Build The Conservative Community on Twitter

  1. Find the following strategy that works best for you. For some people that means following everyone on #TCOT. For others, a more selective strategy of just following people who you find most interesting may work better.
  2. Make a point of tweeting conservatives on the list who you don’t know, but you think might be interesting.
  3. Use the “#TCOT” tag before tweets you think might be of interest to the entire community.
  4. Tell your conservative friends who are not on Twitter to join now.

I find it both frightening and awesome that #tcot (a microblogging tag for “Top Conservatives on Twitter”) beats out any other hashtag used on Twitter by above 10,000 uses. 10,000! It tops out at 30,177 tags, with #mumbai a distant second at 19,492, and #gaza (which I’d have assumed would be on top) in fourth at 13,664.(see: )

Are liberals just not using Twitter? Have they not figured out how to use hashtags? …It appears #tlot (Top Liberals on Twitter) and #tpot (Top Progressives on Twitter) have been suggested, but so far the bandwagon remains a lonely place. (I’m partial to TPOT myself, but then I’m also partial to ridiculous puns; I’d be somewhat concerned, though, about an abundance of pot/kettle/black jokes at the expense of any forward-thinking progressives.)

Still. Something must be done. The right is trailblazing the microblogging scene while the left is caught resting on its laurels. This cannot stand!

Twitter, tagging, and the T.O. blackout | Blog Watch | Twitter shines in T.O. blackout

Tonight, I’m watching the power of Twitter in action. At around 10 p.m., the power went out in west Toronto. People started Twittering about it almost immediately.

Eventually, one Twitter-er, Ryan Coleman, suggested a hash-tag, a Twitter channel, for updates on the blackout:

propose #darkTO for power outage tweets.

Twitter users know a good tag when they see one…

This is really pretty neat. It beats getting people phoning the operator because they’re reaching a busy circuit for the Hydro outages line.

What’s really amazing is that these Twitterers all had to be either outside of the blackout area, running a backup generator, or in possession of an aircard w/laptop and/or (most likely) a cell. (Presents some interesting insights into how Twitter is actually getting used.)

For more about the outage in Toronto: