It’s so bloody hot here in Western Massachusetts – where it never gets hot, and therefore none of us have air conditioning – that I can’t seem to think. I can, however, still count, which means that I can engage in one of my favorite passtimes: quantitative research.
Specifically, I’ve spent the day taking a close look at the blogs that have linked to Global Voices over the past month, from July 10th – August 9th. This is an unusually interesting slice of time for our site – our new design, which radically increased the number of posts we put up per day – went live on July 12th. And it’s been one of our highest traffic periods, including the end of the debate over the Muslim blogosphere’s reaction to the 7/7 bombings and our reblogging of the Committee to Protect Blogger’s story on the BBC and death threats to an Afghan blogger.
The bottom line: Intelliseek’s Blogpulse service sees 1029 links to http://cyber.law.harvard.edu/globalvoices, 259 of which (25.2%) have been posted in the past month. Blogpulse sees an additional 37 links to http://www.globalvoicesonline.org, 23 of which (62.2%) were posted this past month. In total, 26.5% of links to Global Voices have been posted in the past month. That’s interesting, as 36.7% of our traffic since inception occurred in July. One interpretation is that we got a lot of traffic in July to comment threads, but fewer people linking to the site than we might have expected.
(A side note: a search for “http://cyber.law.harvard.edu:8080/globalvoices” – the server which runs our wiki – turns up 1097 links. I’m guessing that Blogpulse interprets that search as the 8080 server as well as the main server, and finds 58 additional links to the anonymous blogging guide, or other popular pages on the wiki. By this count, there are 1134 total links to the vast Global Voices empire.)
These 282 links come from 152 different blogs – the average blogger who links to us does so 1.86 times in a month. Showing us lots of love over the past month: Beth Kanter, with seven posts (I linked seven times as well); Fons Tunistra’s China Herald has five posts; David “Oso” Sakasi and David Weinberger, with four posts; heterotopias.org, Le Chialeux de Salon and Desipundit with three. Instapundit linked to us twice in this time interval, but those links drove a disproportiate number of additional linkers.
Blogpulse believes that Global Voices is currently the 186th most cited blog in existence. What’s more extraordinary about this is that they seem to believe we were as high as the 91st most cited blog around July 30th, when the BBC/Afghan blogger storyy got so much attention.
By way of comparison, our friends at Worldchanging.com have 2252 citations over a much longer lifespan. Blogpulse sees them as ranking 96th in terms of citations and generally experiencing a rank between 72nd and 95th… Our buddy and great supporter Joi Ito ranks 114th, ranging between 102 and 146 over the past month. In other words, we’re now getting as many monthly citations as a well-respected A-list blogger. Not bad for seven months online.
It’s unsurprising that most posts linking to an English-language blog would be in English. And, indeed, 83.4% of the posts that link to a Global Voices post are solely in English. (Many post are in another language, but include English excerpts or phrases.) But there are also posts in Chinese, Danish, Finnish, French, German, Italian, Persian, Portuguese, Spanish and Taiwanese. Spanish is our leading alternative language, with nine posts; Chinese follows with seven.
I’ve attempted to identify the nation a blogger is either posting from or focusing on. This isn’t always possible – in about 25 cases, I have nothing to go on, other than the language of the posting individual. But most bloggers identify themselves, either on a bio page or with a badge proclaiming membership in a local blogosphere. Using this information, I’m able to make the guess that we have bloggers in the following countries linking to us over the past month: Argentina, Armenia, Australia, Bahrain, Bangladesh, Belgium, Brazil, Bolivia, Cambodia, Canada, China, Cuba (probably blogging from the US), Denmark, France, Germany, Greece, Hong Kong, India, Iran, Iraq, Ireland, Italy, Japan, Jordan, Malaysia, Mexico, Nepal, The Phillipines, Saudi Arabia, Singapore, South Korea, Spain, Taiwan, Trinidad and Tobago, the UK and the USA (including an English blogger posting from within the US prison system.) Of 211 posts examined, 38.4% were apparently from the US, 48.8% apparently from outside the US, and 12.8% unclassified.
A couple of weeks back, Rebecca and I both noticed a number of links from blogs that self-identified as conservative. This raised a question for us: were we getting linked disproportionately from the right side of the political spectrum? This proves to be a harder question to answer than the location of a blogger – I ended up deciding that I wouldn’t use any personal knowledge I had of a blogger’s politics, just indications they made on their sites or profiles. (Fortunately, David Weinberger’s “Americans Against Bush” button allowed me to identify him as representing the left, while “Right Wing Hate Machine” made it pretty easy for me to characterize him as representing the right.) Over the period I was considering, we recieved 10 links from 9 self-described “right” blogs, and 10 links from 5 self-described “left” blogs. (Self-described is a key word here – Jeff Jarvis is responsible for two of those links, and he acknowledges that leftists don’t always accept his self-characterization as a “liberal”.) Three linking bloggers identified themselves as moderates, or linked extensively to popular political blogs on the left and right. While there were more individual right-wing bloggers than leftists, the posts from folks on the right were concentrated around July 29th and 30th, and primarily focused on bashing the BBC, while posts from the left were more diverse in timing and topic.
(Before anyone draws any conclusions like “Conservatives are more global than liberals”, let me say two things. One, this is a small sample – 20 data points in total. Two, it’s quite likely that the “self-identification” rule is skewing the results. I know many of the people who’ve linked to us and would characterize them as belonging to the “left” even if there’s no indication on their blogs. A better method to answer this question would be a reader survey that asked readers to self-identify their politics. We may try this in the next month or so. One way or another, though, these results challenge my assumption (as an avowed liberal) that our content would appeal primarily to readers on the left. Very interesting, and worth further thought.)
I come out of my day’s worth of research with a sense that Global Voices is working, in a deep, profound way. Two of our major goals when we started the project this past December: create a space for global conversation, and have an influence on the existing blogosphere, ensuring that blogs aren’t just about US politics and technology. That blogs from 35+ countries and almost a dozen languages are pointing to us suggests that we’re starting to create a global space; that Blogpulse thinks we’re one of 200 of the most cited blogs suggests that we’re starting to have that influence on the blogosphere. It’s not unreasonable to image that we might be one of the hundred most cited blogs by the end of 2005, a goal that would probably have a truly transformative effect on the blogosphere as a whole.
Thanks to everyone who’s linked to, read or been influenced by the links Global Voices has posted over the past six months. Please keep tuning in. We really do intend to change the world of blogging to make it more global, more interconnected and more diverse… and so far, we’re doing it.
In the spirit of transparency, the worksheet I used to make today’s calculations is available here. If I’ve mischaracterized your language, national origin, or politics, feel free to let me know in the comments section, or via email – ethanz AT gmail DOT com.
Pingback: …My heart’s in Accra » Blown Away by Blogpulse
Pingback: John Palfrey
Fascinating – a great case study/example for nonprofits that want to analyze or come up with metrics for their blogging.
I have two possibly stupid questions:
Did you get the raw data completely from blogpulse’s free services?
THe worksheets — is the data extracted from blogpulse? And, if yes, how did you extract?
Pingback: >> mind the __ GAP* ? » on mapping invisibility and ‘bridge bloggers’