The answer is no, we’re not doing anything different. We have our web search results and if a lot of people link to a particular ‘tweet’, or a particular message that someone did on Twitter, then we think that’s really relevant and we see the content of that page or that status message and we’ll just return it with our normal scoring.
We look at the content of the page, which is probably pretty short, 140 characters or less for the core of the message, we look at the people who linked to it and what they said when they linked to it. But it’s not as if we do anything different in our scoring for Twitter, or do anything strange or anything like that.
We treat then just like any normal webpages so we crawl the web, and as part of crawling the web, we end up crawling a lot of different ‘tweets’ or “twitters” by different people on Twitter.
They just get ranked normally. There’s no special sauce or anything differently going on within our search results for just showing Twitter.
Related posts:
- What are the factors that go into determining of a PageRank of a Twitter page, e.g. it is the followers, backlinks?
- Which search media does return the more reliable information: Google or Twitter?
- I have a server-side script that automatically redirects visitors to a mobile version of a site if they are using a mobile browser. My question is: What are some things to watch out for (if any) when serving different content based on the visitor?
- Any reason why Google search does not treat the @ symbol differently given the rise of Twitter? For example, @mattcutts and mattcutts give me the same results.
- How can I make sure that Google reaches and indexes pages that are on a lower (deeper) level of a website?
