Elise’s look at weblog tool marketshare is interesting if not the most accurate. I’d much rather see numbers from someone who could programatically actually determine what blogs use, like Technorati or Feedster. Anyway I tried to follow along in the audience and typed “wordpress.org” into Google, which gave me a helpful page with “link to” and “contains term” links, which I assume is Elise’s methodology. Link to returned 288,000, as is in her chart, but contains term gave 674,000, which is radically different than the 5,000 she attributed to WordPress. I sent a note suggesting she look at this number, to which she replied to Google for www.wordpress.org, which I did. The “contain this term” link returned the even more modest “Results 1 – 10 of about 981,” so obviously the chart should be updated to 1 instead of 5 immediately.
Since until very recently (like two days ago) “www.sixapart.com/movabletype/” bore the same relationship to movabletype.org that http://www.wordpress.org does to wordpress.org (a non-favored alias that redirected to the correct site), I think that MT’s share should also be dropped, to 0.15.
Hrm. Blogger::blogspot.com as Radio::radio.weblogs.com (Program that may or may not be hosted::hosting), so either Radio ought to be at 455, or Blogger should be dropped to 1420 (with www) or 2920 (without).
While the Googling method is clearly flawed, if you got your figures from Technorati and Feedster instead it would seriously underestimate the number of diarylands and xangas out there. I know people don’t always like being reminded that the majority of bloggers are teenage girls, but that doesn’t make it untrue.
Phil, that sounds fair and balanced.
Hmm. I just clicked the link in your post and the clicked the “Contain the term” link: ‘Results 1 – 100 of about 5,040 for “+www.wordpress.org”.’
http://www.google.com/search?num=100&hl=en&lr=&safe=off&q=%22%2Bwww.wordpress.org%22
Keep that 5 handy. And keep up the great work! Thanks!
Google so doesn’t like you. I got to thinking about how incredibly low the results for Blogger were, and how that’s probably because Google special-cases its own properties, so I did what the search engine spammers do when they think their link counts are being artificially lowered: went to AlltheWeb.com.
http://www.blogger.com: 13,100K
http://www.movabletype.org: 3,310K
wordpress.org: 2,700K
http://www.typepad.com: 763K
Those number are certainly wrong, too, but they have a much better feel to them.
Slightly related I recently did some very rough stat gathering for the Bloggie nominated sites. WordPress came third in the list of tools behind Blogger and MT, but I could’ve guessed that I suppose.
Be interesting to see where the divide lies next year.
And with the release of WP 1.5 this week, our numbers are bound to jump again over the next few weeks. It will be interesting to see if Elise does yet another update in the future, and revises her statistic gathering methods.
There’s no one covering this, Blog Census is long dead and the people with the info (Technorati, Feedster, Bloglines) don’t share it. I might just end up writing a crawler, it wouldn’t need to store a lot of data. These numbers will also change a lot when WordPress MU is out.