intel3.0

Social Media, Local Governance, Public Involvement

2009-04-01T18:38:33Z

With the help of Google and some friends on Twitter, I’v put together a few examples and strategies for using social media tools to increase public involvement in local government and public policy. Additional suggestions and examples are always welcome.

Crowdsourcing to increase public involvement, investment, and solicit actionable ideas: http://tinyurl.com/c23bne
Leveraging a mobile / SMS network as a mobile newsletter, for polling, or petitions: http://publicsphere.worldbank.org/m-government-promise-participation
http://wiki.mobiles.tacticaltech.org/index.php/Enabling_participation
http://www.ejeg.com/volume-4/vol4-iss2/griffin_et_al.pdf
A “plain talk” community dissemination and commentary portal: http://bigcitytalk.org.uk/

And, here are a few other links on the subject of public involvement:

The Government 2.0 BarCamp (sorry, the last meeting was March 29, but you can find all the sessions here)
Government 2.0: the February 2009 Social Media Club SFSV Meeting
Simon Wakeman’s “social media and local government” presenation
The FutureGov collaborative consultancy
Building event attendance: http://www.susanmernit.com/blog/2009/03/case-study-using-social-media.html
Using consultants to develop custom public involvement portals: http://www.newcommreview.com/?tag=social-media-case-studies (see: Nanos Research)

I found many other examples in this space, but they tended to overlap with “good social media strategy” in general for businesses and organizations. Many thanks to the following twindividuals for their suggestions and links: RevzNexus, badgergravling, kristiewells, Polledemaagt, simonwakeman, and jennifered. I’m still collecting some examples, and if I get some good ones I’ll update this post.

Top Magazines and Social Media

2009-03-23T19:15:22Z

This is only a quickly compiled list, but here is a table of some of the major social media tools being used by the top US magazines as listed by their 2006 circulation:

2006 Rank	Magazine	Main Page Feed	Message Boards	Twitter	Facebook	Myspqce	Blogs	Online Community	Other Notes
1	AARP		√					√	Photo sharing, journals, video podcast
2	Reader’s Digest	√		√ (1700 followers)	√ (1500 fans)		Several Channels	√ Via Gather.com	Central site links and portal for all social network links. “Out Loud” podcast series.
3	Better Homes & Gardens						√	√ “Home and family” network
4	National Geographic				√ (192,000 fans)		√		Photo feeds, “Herods Lost Tomb” iPhone app, video podcasts of TV series
5	Good Housekeeping		√		√ (300 fans)		√		Mobile content, video channel
6	Ladies Home Journal				√ (278 fans)			√	Video channel, photo sharing
7	Time	√			√ (<1000 fans)				Podcasts
8	Woman’s Day	√			√ (258 fans)		√	√	Video channel,podcast, mobile content
9	Family Circle	√			√ (<800 fans)			√	Calendar, parents.tv video channel
10	People	√			√ (1000 fans)		√		Mobile content, video channel
12	Prevention	√	√				√		Podcasts
14	Sports Illustrated	√			√ (1760 fans)				Video channel, mobile content, “2009 Bracket Challenge” on facebook
15	Newsweek	√		√ (9300 followers)	√ (1800 fans)		√		“On Air” podcast, video podcast channels
16	Playboy	√		√ (4800 followers)	√ (1M+ fans)	√ (4600 followers)	√		Mobile content; central link and portal for all social network links; podcasts
17	Cosmopolitan		√		√ (287,000 fans)		√		Mobile content, podcasts

I also looked at about a dozen other magazines - and pretty much every magazine has a Facebook presence (some intentional, some fan-created) and online content in a blog format (which implies categories, feeds, comments, and commentor accounts). However, it’s obvious that there’s a wide disparity in the level of community and penetration experienced by each magazine in the various social networking sites.

As I said, this was just a quick compile - so corrections and additions are welcome.

Museums Using Social Media - A List of Lists

2009-03-02T20:12:26Z

Here’s a quick collection of lists of museums active in the social media space.

Museums on Twitter (click here to download the full spreadsheet):

Museums on Twitter	name	followers
brooklynmuseum	brooklynmuseum	12679
thewomensmuseum	The Women’s Museum	1451
sciencemuseum	Science Museum	1373
exploratorium	The Exploratorium	1233
gr_art_museum	GR_Art_Museum	794
knoxart	Knoxville Museum Art	728
woodenboat	Spaulding Center	664
mattressfactory	Jeffrey @ the MF	638
eastmanhouse	EastmanHouse	596
atamnh	AMNH	566
grmuseum	Grand Rapids Museum	560
24_hour_museum	24 Hour Museum	524
thefranklin	thefranklin	514
henryartgallery	Henry Art Gallery	507
rubinmuseum	Rubin Museum of Art	497
cincycac	Contemporary Arts	469
americanart	americanart	416
huntermuseum	Hunter Museum	379
asuartmuseum	Diane Wallace	349
21stcenturyabe	21stCenturyAbe	306
asianartmuseum	Asian Art Museum	291
arc_appalachia	Arc of Appalachia	208
mfsm	MFSM	168
fortsmithcvb	FortSmith CVB	136
sciencemuseummn	sciencemuseummn	108
chemheritage	chemheritage	90
edventure	EdVenture	83
cyberville	Cyberville	59
artglassmuseum	Bergstrom-Mahler Msm	39

And, compiled by other diligent reseazrchers on the web, we have:

A list of museums using Flickr: http://tinyurl.com/c36m9g
A list of the top ten Museums on MySpace: http://museumpods.com/id391.html
A list of the top museums on Facebook: http://www.musesphere.com/Facebook/index.htm
A list of over 2,000 museum blogs listed by Technorati: http://technorati.com/search/museum?language=n&media=blogs

And finally, a couple of nice links discussing the “whys”, “whats” and other considerations for museums thinking about getting into the social media space:

Cosi.Org: Museum Social Media Categories
Museum 2.0: Does your museum need its own social network?
Museum 3.0 (no relation): Engaging with Social Media in Museums
Beth’s Blog: What is a museum without YouTube?

The Top Twitterers’ Network

2009-02-25T23:09:53Z

I’ve been playing around with Twitter.com’s new social graph functions in their API, and for a quick experiment pulled the network formed by all twitterers with 30,000 or more followers (the “twitterati” elite). When I get a chance to do some more in-depth network analysis, I’ll follow up to this post, but here’s a quick look at the network. I identified 45 nodes (twitterers) with 30K+ followers, and 458 edges (follower links) among them. The following graph (sociogram- link opens to full size image) shows their connections, with nodes colored relative to each node’s betweenness centrality (directed).

Betweenness centrality usually connotates control or influence over network resources - which, of course, would be “tweets” in our example. Offhand, I’d have to interpret this as the relative ability of a twitterer to decide whether or not to retweet (pass on) a juicy bit of news or information to the rest of the network.

Of course, we should look at access too - the following graph colors and ranks each node by its relative closeness centrality which usually indicates relative connectedness and access to network resources. I’d interpret this as showing who in the network is most likely to hear about a given bit of news tweeting its way throughout the network.

For betweenness, our top twitterer was stephenfry, followed by zappos. For closeness, our leaders were mchammer and again zappos. Overall, the same list of 8 or 10 twitters lead on both measures.

Keep in mind that the network graphs above and measures I’ve discussed only include connections to other 30K+ twitterers; if we included the millions of followers’ links outside of this core group we wouldn’t really be able to make much visual sense of it. That’s a bigger project that I may tackle some other time!

Some other quick observations: it’s a tight network with no natural clusters. The nature of twitter is such that there really isn’t much cost for following everyone else of note, so pretty much everybody in this core is a degree or two away from each other.

These charts were generated using the wonderful, free yed graphing software.

Review: LimeSurvey Open Source Survey Software

2008-11-24T20:47:33Z

Until recently, I’ve always hand-coded my online surveys, especially those requiring survey logic or special features (snowball sampling, name-generators, etc.). Recently, I decided to try the LimeSurvey open source survey software (formerly phpsurveyor) as it looked like it had matured sufficiently to meet most of my needs. I am basing my review on LimeSurvey 1.80 RC1 (November 2008), using PHP5 and MySql 4+ hosted on a shared imHosted.com web host account.

I used LimeSurvey for two simultaneous surveys, sent to executives and professionals in two different high-tech industry sectors in Florida to support economic cluster analysis and industry development activities. Collectively, about four hundred private sector participants responded to the survey using the LimeSurvey interface, accessing the survey via custom forwarded domain names I purchased just for the surveys. Now that the surveys have completed, here are my reactions:

Strengths:

Skinnable, with a selection of decent out-of-the-box templates.
Exportable, duplicatable survey structures that can definitely cut down labor required to set up repeat or similar surveys.
Versatile dashboard for setting quotas, monitoring survey statuses, and exporting results into a variety of formats incluting native Excel spreadsheets.
Powerful and intuitive “survey logic” controls for setting boolean survey conditions (”…if respondent answers this, then show this question”).
Versatile survey flow options (question grouping, and display of questions by groups, singly, or in combination).
Customizable question structures, and an online library of user-contributed question types.
Save and auto-resume features readily available for users who choose not to complete the survey in one sitting.
Supports custom “success / thank you” landing pages.
Automatically generated PDF and “printable” survey versions.

Weaknesses:

Limited documentation forces you to teach yourself the interface largely through clicking every dashboard icon to see what it does (at least, that is how I ended up teaching myself).
The “auto save” feature breaks on many corporate computer systems, I presume in networks that control browser cookies, sessions, and private data. I had many users step away from a partially completed survey, who upon return were asked to provide a login and password - which could not be generated by the survey administrator, and appears to only be generated by respondents who explicitly “save” their partial surveys.
The question structure customization process is very non-intuitive and not well documented. I never was able to figure out how to program a two-dimensional x by y table of numerical text inputs (I had to be satisfied with breaking it into x separate questions with y rows each).
LimeSurvey happily recorded completely empty survey results (although it does designate such responses as “incomplete”). It would be useful to be able to designate at which point the survey is considered “complete”, or to auto-discard completely empty responses.
There really aren’t any features for identifying or preventing duplicate responses.
The automatically generated “printable” and PDF versions don’t resemble each other or the online survey at all, and are not formattable (at least, not through the dashboard, or via CSS that I was aware). The layout of the PDF version in particular appeared very amateurish (it looked better to create a PDF of the “printable” version).

Of LimeSurvey’s weaknesses, the missing “login/password” problems were the most problematic, and likely to result in incomplete surveys and frustrated survey participants. It was encountered by at least six or seven of my survey respondents (at least, those that took the time to contact me as survey administrator); I don’t know how many more of the incompletes suffered the same problems. It’s unacceptable to require a survey participant to ’start over’ a long or detailed survey, without a “lost password” or other interface to help them resume where they left off. Next time I use it, I am going to take some very explicit precautions to improve the experience of my survey respondents.

Overall, I found LimeSurvey to be a solid and valuable survey tool, and one I intend to utilize again. It more than met my requirements, and had a lot of advantages over my old methods for most straight-forward survey needs. It has a few significant flaws, but it is open-source and I’ll bet that they’ll be addressed before too long - perhaps by version 2.0, which is already under development. I’d strongly recommend the software to anyone with at least a fundamental level of server-side scripting knowledge (installing server-level scripts and databases). Give it a shot!

Estimating Company Revenues

2008-11-07T22:29:24Z

When I was doing industry research for academia (ten years ago), we would often utilize a quick rule-of-thumb for estimating the annual revenues for a company (which, unless they were traded on public exchanges, were rarely available) based upon the number of employees they had. The rule of thumb was that for every employee, the company was expected to make about $150,000 in revenue if they had about average profitability. While this was of course a gross estimation, it was always useful for providing a quick order-of-magnitude guess … and was pretty close to the actual revenue more often than you would expect.

Estimates like this are a very simple form of “productivity” or “profitability” measurement, and are often more valuable for understanding growth and health in an industry than outright employment levels (revenues may actually increase while employment drops, if the industry is investing in capital or other labor-saving strategies).

I thought I’d update that rule-of-thumb for inflation, and see if it could be made more accurate for Florida and for specific industries. With the help of the Florida Agency for Workforce Innovation and the US Census department, I found that number is still pretty accurate overall but varies considerably depending on the specific industry:

Average revenue (receipts) per employee, all nonfarm industries: $155,855
Average revenue (receipts) per employee, by NAICS Industry category:

NAICS	NAICS Industry Title	Average Revenue Per Employee
211	Oil and gas extraction	$653,026
212	Mining except oil and gas	$259,278
213	Support activities for mining	$91,256
236	Construction of buildings	$402,222
237	Heavy and civil engineering construction	$219,837
238	Specialty trade contractors	$102,937
312	Beverage & tobacco product mfg	$488,544
313	Textile mills	$151,282
314	Textile product mills	$101,500
316	Leather & allied product mfg	$90,020
322	Paper mfg	$338,855
323	Printing & related support activities	$118,034
324	Petroleum & coal products mfg	$389,390
326	Plastics & rubber products mfg	$160,830
331	Primary metal mfg	$280,552
334	Computer & electronic product mfg	$234,192
337	Furniture & related product mfg	$113,561
423	Durable goods merchant wholesalers	$634,292
424	Nondurable goods merchant wholesalers	$710,063
425	Wholesale electronic markets and agents and brokers	$1,943,250
441	Motor vehicle & parts dealers	$477,929
442	Furniture & home furnishings stores	$197,921
443	Electronics & appliance stores	$231,990
444	Building material & garden equipment & supplies dealers	$210,899
445	Food & beverage stores	$140,304
446	Health & personal care stores	$162,314
447	Gasoline stations	$301,395
448	Clothing & clothing accessories stores	$124,033
451	Sporting goods hobby book & music stores	$127,705
452	General merchandise stores	$180,443
453	Miscellaneous store retailers	$119,799
454	Nonstore retailers	$327,096
481	Air transportation	$244,773
483	Water transportation	$721,237
484	Truck transportation	$116,283
485	Transit & ground passenger transportation	$70,070
486	Pipeline transportation	$476,142
487	Scenic & sightseeing transportation	$66,803
488	Support activities for transportation	$111,086
492	Couriers & messengers	$106,488
493	Warehousing & storage	$26,172
511	Publishing industries (except Internet)	$164,228
515	Broadcasting (except Internet)	$221,386
516	Internet publishing & broadcasting	$157,703
518	Internet service providers web search portals & data processing	$145,751
519	Other information services	$78,610
521	Monetary authorities - central bank	$810,118
523	Securities intermediation & related activities	$261,880
525	Funds trusts & other financial vehicles (part)	$757,219
531	Real estate	$167,220
532	Rental & leasing services	$166,177
533	Lessors of nonfinancial intangible assets (exc copyrighted works)	$183,898
541	Professional scientific & technical services	$112,145
551	Management of companies & enterprises	$22,063
561	Administrative & support services	$38,344
562	Waste management & remediation services	$161,856
611	Educational services	$81,089
621	Ambulatory health care services	$107,759
622	Hospitals	$103,109
623	Nursing & residential care facilities	$48,241
624	Social assistance	$43,500
711	Performing arts spectator sports & related industries	$115,735
712	Museums historical sites & similar institutions	$63,263
713	Amusement gambling & recreation industries	$68,149
721	Accommodation	$69,209
722	Food services & drinking places	$40,557
811	Repair & maintenance	$87,924
812	Personal & laundry services	$52,599
813	Religious/grantmaking/civic/professional & similar org	$115,972

Results were based upon the 2002 US Economic census for the state of Florida, and adjusted for inflation to 2007 USD. Special thanks to Becky Rust and Pamela Schenker at the Florida Agency for Workforce Innovation, they do more work than they’re appreciated for.

Getting a Grip on Twitter Influence

2008-10-27T19:25:39Z

A few weeks ago, I spent a long day throwing together a new Twitter application, “twinfluence.com“. In about 8-10 hours coding PHP and MySQL with the Twitter API, I launched the site in time to present it at a BarCamp. The primary idea behind it was that simply counting followers was an insufficient way to measure influence via Twitter.

Counting followers treats all followers as equal, and assumes that any tweet message ends with one’s followers - which we all intuit as wrong. Ideally, we would methodically map out the entire network of Twitter and calculate some rigorous social network analysis metrics like clustering, betweenness, equivalence, etc. to really understand the structure of the Twitterverse and the role any particular Twitterer plays in it. Sadly, the limitations of internet communication and the Twitter API make that completely impossible for a “real time” twitter analyzer like twinfluence. I was forced to limit twinfluence’s real-time “crawling” and indexing to the second-order network - followers of followers (and still have to cache those with 30,000+ followers for offline crawling and analyzing). Again, it’s an arbitrary cutoff (Josh Carrico’s blog discusses this nicely, and of course retweeted messages can travel further than that), but extending the horizon to the second order network has helped change the discussion of influence to include whom one’s followers are, and how much audience each brings to the party.

To quote the intro on twinfluence.com’s “about” page:

Imagine Twitterer1, who has 10,000 followers - most of which are bots and inactives with no followers of their own. Now imagine Twitterer2, who only has 10 followers - but each of them has 5,000 followers. Who has the most real “influence?” Twitterer2, of course.

I also wanted to make certain that the metrics were all transparent and well-defined, and some global trend statistics were collected so everyone could get a better idea of what should be the “baseline” expectation as one’s twitter network grows. Twitter Grader (probably the most comparable Twitter tool out there) is an impressive app, but it doesn’t live up to my expectations on those counts.

Twinfluence ranks on Reach (second-order followers), Velocity (how quickly reach is built), Social Capital (average followers’ network size), and Centralization (how much the network relies on a few high-profile followers). Detailed definitions are available on the twinfluence.com “about” page. In two weeks, over 7,000 twitterers have been analyzed, and we’ve learned that as a twitterer’s network grows - (1) they tend to pick up a few high-profile followers, but over time new followers tend to have smaller and smaller networks; (2), their networks decentralize over time; and (3) velocity picks up (snowballs) as network size grows.

Twinfluence has gotten a lot of attention, discussion, and worthy critique. Some (of many) of the blogs that have picked up the discussion:

The latest conversations I’ve had on twitter is that the social capital metric is skewed toward twitterers with small networks. We can probably improve the social capital metric, make it more realistic and useful by devaluing the contribution of followers who themselves have thousands of followers (those probably don’t contribute real “capital” anyway, since they probably aren’t listening). When I hash out a new, workable algorithm we’ll release a new version of Twinfluence. My thanks for suggestions/feedback by ScottAllen, Macartisan, Sigepjedi, CCSeed, Yannleroux, Mlaine, KevinLyons, PatchWalker, RepeatNone and others.

The Art of Case Study: Structuring Unstructured Information

2008-09-30T17:48:55Z

I see the term “case study” used frequently in PR, marketing and management research – but often it is used to describe a very informal and arbitrary descriptive narrative. To me, “case study” should refer to a specific research methodology for making consistent comparisons and measurements of two or more subjects, particularly when combining multiple and unstructured sources of information.

It’s not a complicated tool, but a useful one for conducting background, literature, and internet research in a systemized fashion. The basic process includes:

Ask what dimensions are measurable given the available data, and what will be useful to measure.
For each dimension, establish consistent criteria for measurement, boundaries and limitations, and how it should be measured.
Construct a “template” or outline which will be filled out for each case.

The template should still include a section for a narrative description of what makes each case unique, such as its history and individual characteristics. The most important task is the development of dimension variables and how they are assigned, but the only real rule is that you are consistent. Dimensions can include existing quantitative data (for example, “number of full time employees” for company case studies), or subjectively assigned values. Subjective assignment can be performed using rank ordering by the researcher or another source (for example, “best restaurants in town” as ranked by personal experience), or assignment to a Likert scale (local restaurants scored from 1 to 7 where 1 indicates “Very Bad”, 3 indicates “Average”, and 7 indicates “Very Good”).

A simple example is how Wikipedia handles articles about corporations – it includes narratives, timelines, histories, and that nice little box of “quick facts” in the top right corner which tends to be pretty consistent across different articles. Just like in the Wikipedia example, there’s nothing that says you can’t include multimedia in a case study research project.

Consistent assignment of dimensions will enable the researcher to perform simple descriptive statistics and graphs, and even create combined rankings that weight and combine values from different dimensions. A completed case study report should consist of the following:

Summary / overview
A definition of how dimensions were measured, including sources
Individual cases, all using the same template
A “meta” analysis, that provides descriptive statistics and analyses of case-by-case measurements

A well defined – even if it is short and sweet – case study plan will help you get more useful information from your background research phase, and help you present the results in a more useful and understandable format. Just remember the key word – consistency.

Social Media Case Study: Carnival Cruise Lines

2008-09-22T20:41:05Z

Miami-based Carnival Cruise Lines has been a pioneer in leveraging social media to build customer loyalty and satisfaction. I interviewed Carnival’s Senior Manager for Interactive Marketing Strategy Jordan Corderra and Social Media Strategist Stephanie Leavitt about how they built such a successful and loyal online community.

Company Profile:

Employees: 81,200
2007 Revenue: $13 Bn
2007 Revenue Growth: +10.1%
Headquarters: Miami, Florida
Established: 1972
Type: Subsidiary (Carnival Corporation & plc)
Website: http://www.carnival.com/
Sources: Wikipedia, Yahoo Finance

Carnival Cruise Line’s social media components:

Blog: John Heald’s Blog
Virtual, interactive tour: FunShipIsland.com
Online community/ social network: CarnivalConnections.com. Photo / story sharing and scrapblogging, trip planning, event organizing, Q&A
Twitter: CarnivalCruise
Flickr photostreams

Targets

Travel agents
New customers
Past customers

Measurements of Success:

Site traffic (each of the three listed sites topped one million visits by February 2008, and over two million visits have been recorded for John Heald’s blog)
Bookings (over 20,000 generated by CarnivalConnections.com in 2006)
A 2008 cruise was organized for subscribers of John Heald’s blog, and over 800 participated

1. What were the specific outcomes and objectives you were attempting to reach with the blog, funshipisland.com, CarnivalConnections.com?

Each site has specific goals, but overall the objective is brand positioning. John Heald’s blog is a communication medium that’s used for engagement. We’ve changed the objective a bit since it first launched as it was originally intended to build buzz about our latest ship launch.

Funshipisland.com highlights our unique products/ship features and serves as a prospecting tool.

Carnivalconnections.com focuses on community and helps guests connect with other cruisers and to Carnival. The original objective of the site was to increase bookings by encouraging guests to invite their friends and family to join them on their FunShip cruise. The forums were an added bonus and we launched the Carnival Channel and the various blogs in Nov. 2007 as a way to “get to know” Carnival and the people that work here.

2. What have been your ROI / measurements of success?

Because each site and each component of our social media strategy has different objectives, they each have their own success metrics. We definitely track engagement, satisfaction, and leads/sales. But to have a truly successful social media strategy sales shouldn’t be the number one measure of success.

3. Did Carnival have any “false starts” initiating their social media strategy? Did you utilize any consulting firms to assist your strategy and rollout?

Of course we encountered some “false starts” along the way. We definitely eased into our foray in social media. Carnivalconnections.com launched in Feb. 2006, John Heald’s blog launched in March 2007, Funshipisland.com launched in July 2007 and our Twitter, Facebook, YouTube activity launched shortly after. Our interactive agency at the time, Avenue A Razorfish, helped with Funshipisland.com and Carnivalconnections.com.

As we mentioned earlier, John Heald’s blog started out as a very focused initiative. Once John finished his time onboard the Carnival Freedom we had to rethink the strategy of focusing on a particular ship.

From launching a new site to simply opening a Facebook fan page, each and every one of these initiatives involved in depth discussions with our interactive team. For some brands it doesn’t make sense to participate in every social networking site available, so we definitely tend to ease into new applications. But the great part about social media is that you learn something new just about every day, so you shouldn’t fear it.

4. I see Carnival has some activity on Twitter and Flickr - do you have any significant plans for these or any other upcoming social media platforms?

Our Twitter and Flickr activity plays a role in our overall social media strategy. As for significant plans, you’ll just have to follow us to find out.

5. I recently read about Carnival hiring SAS Marketing Automation ( to integrate and perform predictive analytics for their marketing and customer communication campaigns) - will that tie into your social media efforts, or is that a separate and independent initiative?

We are very excited about that partnership, however the interactive team will not be utilizing their services at this time. It is certainly something we are thinking about for the near future.

6. Do you have any advice for other companies considering building their own online communities?

If your target audience isn’t there, rethink investing time/money into that platform. Just because it works for another brand doesn’t mean it will work for you.

Links:

Analytics Inside MMOs - The Government Wants To Be There First

2008-09-19T15:41:26Z

When Google agreed to provide China censored search capability, it justified its action on the principle that information cannot be contained. The implication was wherever a connection exists, people will explore and adapt their methods of communication to allow the free and unmonitored flow of information.

This week, the Greater Tampa Bay National Defense Industry Association (NDIA) had an event welcoming the joint Coalition forces, and the subject of social media came up between a general and some highly placed defense industry execs. Aware that the US has been monitoring their social media activities, terrorist groups have apparently moved some of their key conversations… into the World of Warcraft and other MMOs.

Most analytics are based upon the consistently structured, open-access, and archived nature of conversational interaction on social media platforms. Conversely, MMOs enable proprietary, unstructured, unrecorded, firewalled, many-to-many communication amongst participants, sharing voice, text, images, and other media, that do not lend themselves even a little bit to external monitoring and analysis. Proof of the effectiveness of this strategy - and Google’s position - was demonstrated when information regarding democracy and civil liberties was shared to and among Chinese citizens via MMOs as early as 2005, effectively bypassing the government censorships of those topics on the internet, as shown by this extract from “Taking New World Notes” by Wagner James Au:

“As of this date [late 2005], the firewall does not evidently block counterrevolutionary speech transmitted within online worlds. So far, Chinese authorities have not cracked down on political expression in MMOs (though they have imposed limits on play time, arguably an indirect abridgement of free speech.) With over a million paying Chinese subscribers in World of Warcraft alone, however, sheer numbers make it inevitable that some will soon directly test the limits of political expression there, too.”

While it was opined that “NSA can get into Blizzard’s systems when it needs to”, we have an analytical and accessibility rift between social media platforms that foster access to conversations among members, and those that don’t. The implications are significant for counterterrorism and intelligence purposes, but also for everything that social media analysis is used for - market and brand research, advertizing and monetizing, public relations… and of course civil liberties. Will government intelligence efforts “crack open” MMOs and lead the way for civil and commercial analytics?

Thanks to Thomas Eskridge of the High Tech Crime Institute and Ken Pohl for their insight and contributions to this discussion. For more discussion on the US government project to mine MMOs for counterterrorism, read “Government to Seek Terrorists in World of Warcraft: The Full Proposal.”