I had a surprise this weekend. After years of declining value through increasing noise, I'd essentially stopped checking my Technorati stats. What had degraded almost beyond usefulness took a leap into unrecognizable territory last year, as the team started revamping its software. Yet, it now seems that the tinkering has worked: Technorati got its mojo back.
Back in the early days of blogging, all the kool kids "claimed" their blogs on Technorati. In those days, pre-2005, there was no Google Blogsearch. Not everybody had a feed yet, and the diversity of blog software was very uneven in its support for feeds, trackbacks, and blogrolling.
Topical "communities" grew mostly by the blogrolls of well-trafficked sites -- those who knew the ropes led by example, others picked up the techniques as they read. A few commercial ventures started to promote the idea of shared hosting for blogs with similar interests -- leading to the birth of the "science blog" concept, both in the trademarked and more ecumenical versions. In the blogoverse, one ruling credo came to the fore: It wasn't how many readers you had, it was how many inbound links. Links gave an air of reciprocity to the entire enterprise -- you read other people, linked to what you liked, and thereby shared their work with others. Link-love was born.
Technorati did two things well. Most important, the site provided a report of inbound links. For those (like me) who never did get trackbacks to work right, having a third-party site provide link data was the best way to find new blogs, return links to them, add them to the blogroll, and build the community. I checked the site every day, to find the conversations that my posts started elsewhere.
But Technorati's more visible accomplishment was its blog ranking. One numbered list, from one to twenty-four million, listing every blog in order of inbound links. In the early days, the list was almost aristocratic. Latecomers like me could aspire to get the number of links of more established blogs, but it wasn't too easy to climb the charts because they tracked the total number of links, not the number in the last few months. It took an interesting writer to climb the chart. Technorati wasn't the only ranking or link-tracker -- N. Z. Bear's "Ecosystem" memorably gave a cute name to everyone in its tiers of blog ranks, and other upstart rankings would appear over the next several years.
In 2006, mainstream science started to take notice of blogs. Nature ran a news story about the phenomenon, listing 50 "top" science blogs, taken straight from the Technorati rankings at the time. My position on the list, at number 14, proved to be a great advantage as I tried to explain to skeptical colleagues what the heck I was doing. It's one thing for Technorati to say you're number 14 at something, but when Nature agrees -- well, that's scientific. It made a difference.
But in many ways, that was the high water mark of the independent ranking and link-counting engine. The growth of blog "communities" gave new ways to game the system. Repeated blogrolls and aggregators guaranteed some blogs a link count vastly out of kilter with the conversation that they actually provoked. "Link trading" among completely unrelated blogs further distorted the picture. Over time, Technorati started to degrade. The worst of the problem came from spam blogs. As the spambots ruthlessly interlinked with each other, they climbed the rankings and pushed the real blogs down the list. By 2008, most of my inbound links turned out to be either spam comments (not me, I swear!) or automated blogrolls from sites that have nothing to do with science. With all this static, somebody badly needed to adjust the rabbit ears.
New aggregation services, like Postgenomic, Wikio, and others, tried to create topical lists for science. I tried many of them over the years, but each fell short -- either leaving out big chunks of the conversation, or failing to fix problems with incoming feeds. I'm running stock Drupal, but for some reason many third-party sites can't index my feed and won't give me any service to figure out why. Does anybody care? Heck no -- the point of these sites is either to test algorithms or to place advertisements. Technorati was just as mercenary as any of the others, but at least they got it right -- even when I was running Bloxsom with RSS 0.85 cobbled in Perl. But this same accessibility may have made the site's Achilles heel: spambots were pinging ten times as much as the real blogs.
Last year I'd almost given up on finding incoming links. Google gives me a daily digest, but their indexing misses a surprising number of people, even those with well-established sites who write every day. I would sometimes discover that somebody had linked to me weeks before. Often I'd write, or go and put up a comment -- but after days the opportunity for conversation may already have passed by.
There were signs that other people had the same problem. Blogs no longer seemed the conversational tools they once had been -- people were taking true conversations onto Twitter. The social web started to draw people into backchannel conversations, inaccessible to ordinary readers. Fewer and fewer of the established science blogs seemed to explain things from the beginning. For those of us who may not want to "follow" or "friend" each other all the time, the web started to seem like a lonely place. Was there still room for a real correspondence, with revision and thought over time, instead of jotted into 140 characters?
Seeing the last few days of incoming links, all in one place, makes me feel a lot better. It was just chance -- really, looking for some blog coverage of how Sketchbook Pro works on the new iPad -- and there they all were. My inbound links! No spamblogs, no endless list of blogroll links. Now I'm back to discovering new blogs, from people who are engaging with my field. Old friends may have gone over the last few years, or have cut back their online writing as they moved from graduate school into industry or the tenure track. But new ones are rising to take their place. The aristocratic element of the ranking has relaxed -- the new Technorati algorithm is ignoring blogrolls and is instead reading feeds for new links. That gives it an immediate quality, and a blog's rank will ebb and flow with the conversation it provokes.
After all this time, I'm still in the top 14 for science.