Top Clothing Queries
2004
1. bikini
2. mini skirt
3. prom dresses
4. lingerie
5. little black dress
6. poncho
7. t-shirt
8. sports bra
9. red dress
10. low-rise jeans
1. Sex. 2. Sex. 3. Sex. 4. Sex. 5. Sex. 6. Rain! 7. Clothes. 8. Working out so I'll look good so I'll get more Sex. 9. Sex in red. 10. Sex.
Although the top results for poncho are for those stupid hippie ponchos that are slightly assymetrical that are all the rage nowadays, and not something that you could, say, go hunting in. Oh, well.
Hmm, as far as I can remember, I never searched for any of those terms during 2004. In fact, other than night vision, I'm not sure I've ever searched for those terms or phrases alone. Did they count searches that include those terms in addition to other terms? If not, who the hell searched for "cnn"?
I feel like I just beat a kitten to death... with a bag of puppies.
There's a lot of questions that this data prompts. For instance, mp3 is number 10 on the list of popular queries, but not number 1 on the list of tech queries. If you take a close look at the data, you'll see several anomolies like that.
Google doesn't present their results like Lycos does: these things are not labeled Top 10 Search Requests, but rather "Popular" Queries. I know for a fact that Google screens the sex out of their results. It is unclear whether this data undergoes further post-processing, but it wouldn't surprise me. It's closer to editorializing than just presenting the facts.
It would also be interesting to see the size of the tail here. For example, how often do people search for "britney spears" compared to searching for anything related to food? Are the top ten queries a thousand times more popular than the next hundred or only two times more popular?
Definitely not a rigorous presentation. Note that they also took out the browser and OS data which I thought was the most interesting part.
Dwindlehop wrote:Definitely not a rigorous presentation. Note that they also took out the browser and OS data which I thought was the most interesting part.
Dwindlehop wrote:Definitely not a rigorous presentation. Note that they also took out the browser and OS data which I thought was the most interesting part.
This is a result of the whole browser market share flap with browsers like firefox pretending to be IE because of stupid sites that actually check and say "no soup" if you're non-IE.
Yeah, but that's BS. If nothing else, you can present the data as "here's the number of browsers reporting themselves as IE." It's still an interesting number.
I agree that it would be nice to have the raw results out there for people that understand the caveats associated with the data collection. Unfortunately, the people who write news stuff don't know the difference so they editorialize the results for Google. There's not much difference from Google doing the editorializing and someone else except that at least Google's in a better position to understand the results to do the editorializing.
George wrote:If not, who the hell searched for "cnn"?
I had a roommate (stupid drama major) who asked me how to find something on the internet, with the computer I had set up in the living room. So I told him to try google. I looked over a few minutes later, and he was at Yahoo, doing a search for Google. Eventually, he actually got to Google and figured out to type in the actual thing he was searching for.