I've recently had a look at some of the search terms that have been leading people to my blog, and I find it interesting. For one thing, I note that my occasional use of lines from songs as otherwise-unrelated titles of blog entries is probably leading people astray. They're looking for the song lyrics, and winding up here. To that I say this: I'm sorry that this isn't what you were looking for, but I hope you've enjoyed finding it.
I also see from the searches that most people don't know how to use quoted strings in their searches in order to be more specific about what they're looking for. A search for [tom cruise], for instance, will give many unrelated hits (“Tom and Jane came back from their Caribbean cruise last week.”), while a search for ["tom cruise"] is probably what you want. I saw several searches for [i like coffee i like tea i like java], and variations thereof. To that I note that without quotation marks, there's no point in using the same word twice. That search is exactly the same as [i like coffee tea java].
Along that line, I saw a recent set wherein the same user searched for [recycling waste], [recycling garbage], and [garbage recycling], each five minutes apart, and visited this entry from each of the search results. Apart from the fact that the last two are essentially identical searches (there's a little difference, because Google will give different search rankings to the pages when you re-order the terms), I'd think the user would have noticed that he'd already looked there. The browser even tries to help by highlighting the pages you've already seen.
I have to say, I'm impressed with Google Blog Search! I posted I, the jury right about 8:00 yesterday, and I got a hit there from someone who used Google Blog Search at 8:29 to search for the name of the Connecticut teacher (I won't repeat it here and fool the search engines). That's fast! I guess it's not terribly surprising that Google would index new posts on Blogger that quickly, because Google owns Blogger now. Still, I'm impressed.
Here's a selection of other search terms that I found interesting, for one reason or another:
I wonder what the connection was, in the user's mind, to [religion atheism lawyers money].
I hope my whimsical idea for “Talk Like Dr Cox Day” (and its implementation) was just what the guy who used [how to be like dr cox] was looking for. (And there actually are quite a few searches related to Dr Cox.)
One user seemed to be looking for news with [rich man wins a million dollars in lottery], while another wanted ideas, looking for [i want to win 20 million dollars for free]. Money for nothing, kicks for free, I guess. (Dang, now I've just thrown off another song-lyric search!)
My guess is that this entry is not what the guy who searched for [i totally don't know what that means, but i want it. video] was looking for.
On the other hand, I like to think that I've given these people some useful information:
[long term effects of the john scopes trial]
[bluetooth stereo headphones for cell phone]
If so: “You're welcome.”
I'm sure this guy also found exactly what he was looking for, because my entry points to the full text of the study, but in any case he wins the award for the longest set of search terms:
[a june 2006 study by duke university concluded that the average american today only has two close confidants]
For whoever searched for [santa esmeralda dont let me misunderstood cover model], I didn't say it in the entry that search led you to, but in case you're still reading... the answer is Karen Hafter.
In Every small victory, I mentioned that tennis star Maria Sharapova had been pushing the Wimbledon folks to even up the prize money between men and women. These guys, though, seem to be looking for different aspects of Ms Sharapova:
[maria sharapova sexual fantasies]
[close-up of maria sharapova's lips]
Here's a question everyone needs to know the answer to!: [how do i gently ask a coworker to put their cell phone on vibrate]
Finally, I'll leave you with the most surreal of the bunch, for which I have no explanation: [garlic leadership]
5 comments:
OK, this is either a coincidence or some very odd RSS behaviour. This morning I noticed two new items from you via the RSS feed I have set up in Thunderbird. The first was the item Talk Like Dr Cox Day, which I thought a bit odd since I was sure I remembered it from last year. Sure enough, it was a pointer to that item.
The coincidence is that you mentioned that specific item in your blog entry for today ("Search me").
I've noticed several times before that old items (sometimes very old) will magically turn up again, but only for your blog and for the Molly Ivins (R.I.P.) feed to which I used to subscribe.
It must be 42.
On seeing that lots of searches are hitting the Dr C entry, I decided this morning to update it to point to the December 1st entry (the "implementation").
When an entry's edited, Blogger sticks it at the front of the feed and sets the <updated> tag in the feed. So any time a Blogger user updates an entry, it'll show up at the top of the feed again.
I wish I could stop that, at least for an individual edit ("This is a minor edit, so you shouldn't put it at the top of the feed."), but there doesn't seem to be a way to control that.
The most amazing thing here, at least to me, is that you have the ability to find out so much about who is looking at you.
I never would have guessed that every time I click on a blog turned up by Google that the author would even be able to learn what phrases prompted the read!
What else are you able to find out?
Well, it's all stuff that the browser provides, as part of the protocol when it talks with the web server. A browser can choose not to (or can be configured not to) provide it, but the browsers most people use are happy to. There are "anonymizer" proxies that one can use, which filter that stuff out (or which may even be set up to lie). Some of it comes from a bit of javascript that runs when the page is loaded.
I get your IP address, your browser name and version, your operating system, your screen resolution, the "referring page" (the URL of the page you clicked on to get there), and the search-engine information. Some of that isn't always applicable (the referrer, if you got there with a bookmark, for instance, or the search-engine information if you didn't get there from a search), and most of it can be suppressed, though most browsers don't have the ability built in to suppress them.
From the IP address, I can see who your service provider is, and, to the extent that the service provider localizes its services, where you're located (not your home address or anything, but your state/country, and maybe your city or thereabouts).
If you're worried, use an anonymizing proxy — there are some freely available on the Internet, and they'll hide (or lie about) everything, including the IP address (which is the one bit that the browser can't hide). And configure your browser not to run javascript. That one's dicey, though, because many web sites just won't work if you disable javascript.
If you just want to hide your search terms, don't click on the links that the search engine gives you. Instead, copy the URL to the clipboard and paste it into the address bar (it's just four mouse clicks, instead of one, in IE and FF). That will hide the search-engine information, because the browser won't know you came through a search.
"garlic leadership" obviously refers to the Slobodan Milosevic staking, or possibly the theory that politicians are vampires ... but why it's bringing people to your blog is beyond me!
Post a Comment