Alden Bates' Weblog

Google BlogSearch was launched today.

I suspect most of the blog search searches will be for terms relevant to bloggers (blogging software, spam, current events, etc). I was amused to find I'm top ranked for Operation Overlord, despite the entry in question being a review of an Unreal Tournament map and not much at all to do with the World War II event, but as most searches for information on Operation Overlord are likely to be conducted on the main index, I doubt it'll matter. The most interesting thing is the freshness of the results (sort by date and you'll see how long ago the latest post was made).

Feeds don't block robots

LiveJournal allows users to block their journals from being spidered, however as their URLs are in the form http://www.livejournal.com/users/journal/, they can't use a robots.txt file to selectively block journals. Instead they use the meta robot tags on each page. Herein lies the problem: there's no supported way to add meta tags to rss and atom. End result: Google has indexed a bunch of journals which the owners wanted kept out of the index. This was a problem previously with other search engines like Technorati, but Google has a much, much higher profile... A reminder that noindex tags are not a guarantee no one will find your entries.

Technorati 1, Google BlogSearch 0, IceRocket 0.5

says Eric Rice. But I'm sure Google will be doing plenty of tweaking before they take blog search out of beta...

One thing I noticed searching on the blog search with Firefox: When I click on a link and get redirected to the end page, FireFox is displaying Google's [G] favicon on the tab and in the URL bar instead of the favicon for the site. Odd bug there.

Edit: Google are working on the LJ problem.