WebProNews

Tuesday was a very busy day for me, with the expo floor, a pretty major interview (more on that as it develops) and lots of busy work for me.

As a result, the one session I actually blogged was "Meet the Blog and Feed Search Engines".

First up was Mark Fletcher, who discusses Bloglines. He says feed content is the fastest growing content segment on the internet. Says they only crawl feeds that are subscribed to, so when you start a new feed, you should subscribe to it yourself. Also mentions the coming blog search from Ask, which he wishes he could demo today. Says there are several apects to blog search, including searching for blogs, searching for citations, saved searches, popular blogs, most popular articles on a daily basis.

Next was Adam Hertz from Technorati. Says Technorati was called the "backbone of the blogosphere", which makes him the chiropractor (nice laugh). Starts with Infoporn, which is a requirement for any Technorati talk. 10% of blogs post once a week. 9% of blogs are spam, 25% of pings are spam. In the last year, English has fallen to second place among posting languages, with Chinese taking first place.

Adam asks how many people use blog search. Not too many raise hands. How is it? Crowd agrees, not so good. They use authority to rank popular blogs, and to restrict searches to popular blogs. Also, started using tags a year ago, and 90 million tags already.

Next up, Chris Tolls from Topix.net. Says that Google and Yahoo does a good job with the static reference web. However, there's this incremental web, where things are constantly changing, and you are only interested in what is going on now. He says what's different between them and Technorati, is they automatically tag each piece of content based on topic, so that if a blogger talks about many things, it'll show up in the right place, while Technorati asks people what their blogs about, and when they go off topic, the system breaks.

Topix has started some forums on the news they have, and in just a month, they're up to 4,000 posts a day. This proves people want to talk back, they want to interact with the news, even if they don't have their own blogs.

Says Yahoo has a lot more page views than Google, but makes the same amount of money. The reason: Search has a $50 eCPM, compared to $4 eCPM for content.

Next up was Chris from Feedster. He talks about the "Good Old Days" of Feedster, 18 months ago, when they were indexing less than a million blogs. Shows a clip from Rocketboom where they cover the new look of Feedster. Says they never expected this business to grow so quickly (probably why they're having so much trouble with growth), and they'll be at 30 million feeds next quarter.

He mentions they started the Feedster 500, but they'll soon be adding a way to search that algorithmically in the future.

Next is Vinod from Google, who wasn't supposed to be here ("roped into it" at the last minute), and explains Google Blog Search. He says that one challenge is because some blogs are about fresh news, but others are about communication and creating a social network, and showing recent results doesn't work for everything.

Last up is Jeremy Zawodny from Yahoo. He says in 1997, all everyone cared about where eyeballs, and we suffered flashy, ugly, annoying advertising, and Yahoo made a lot of money. "Those were the good old days". People used to visit lots and lots of websites every day. Now, they pull the information down to them instead, through a feed reader. He says the web is growing faster than Yahoo, so Yahoo is a smaller percentage of the web than it used to be.

Jeremy says that once people hit a critical mass of bookmarks, there's a point where they switch over to feed readers. People also don't follow random links and read lists of links like they used to; instead, they type random things into search engines and check that out. Jeremy notes that, as a pilot, he gets lots of magazines on flying, and those things are filled with advertisements. However, he reads those ads, and checks them out, because they are interesting and useful.

He shows Yahoo's blog listings in their news search. It isn't a stand-alone blog search, just integrated, because they felt there was no need for another one. Looks at meme-trackers, like del.icio.us, which can be good for 20,000 hits on their popular pages.

During the Q&A, Chris from Topix talked about spam. One guy in the audience assumed that, because RSS is a pull system, there shouldn't be spam. Chris: "Have you used a search engine lately?" Nice laughs. He says that there are so many link farms out there, that the Google approach to counting links doesn't work anymore. Jeremy says anytime the incentive becomes high enough, people will try to find a way to get to the top.

Adam says that they try to look at the business model to detect issues. One thing they do (and this is surprising, and brilliant) is scan the Google AdSense id to detect patterns. Vinod says Google has unique issues because they have both the blog search and the content, through Blogger.

Someone asks what is the big difference between hosting a blog on BlogSpot or on their own server, and Adam says the only thing they use that for is spam detection, because for a while, much of what they got from BlogSpot was spam. Jeremy says that once people realized that blogs, due to design, rank better on search engines (BLOG = Better Links On Google), it created an explosion of splogs.

The discussion segued into China and censorship. Vinod says Google has no blog search in China, and there's a reason for that. Adam notes that the Chinese people leverage tags to stay under the radar, with groups choosing innocuous, almost random tags to links items on a subject. Chris from Topix says they have no Chinese content, partially because they are owned by three newspapers who have zero business in China.

Add to document.write("Del.icio.us") | DiggThis | Yahoo! My Web

Technorati:

Nathan Weinberg writes the popular InsideGoogle blog, offering the latest news and insights about Google and search engines.

Visit the InsideGoogle blog.