It’s been a while since I’ve done a thorough expose on my search strings. Things have changed a bit since a) I had catastrophic data loss about three months ago, and b) I’ve moved from Movable Type to WordPress (a move that happened along with a switch from one webhost to another). I’ve partially alleviated these problems by wholesale copying my MT archives into my new public_html directory, but the fact remains. Those links still work, but those pages aren’t going to be updated. All the urls are different now.

The reason why my search string collection is so interesting is because I have a text-heavy website. I use a lot of words, and words are the key to Google’s algorithms. I put words together, and Google looks for sets of words, so I get all kinds of hits from people who may or may not be looking just for me, or someone like me. While I started collecting and analyzing these strings as a bit of a joke to entertain my friends, it turns out that this examination is quite enlightening. I find myself mentioning this ongoing research at work, to illustrate a point. It turns out that looking at these strings has twigged me in to some elements of web searching I think I would otherwise have breezed right by. While I often find these strings funny, I respect user searches more now than I used to. It feels as though this work has worked to my advantage in ways I could never have imagined. The fact that people find me when they’re not looking for me is a great gift.

That said, sometimes people are looking for me.

mazar
rochelle
diary of a subversive librarian
rochelle mazar
mazar.ca
random access mazar
utm mazar

It doesn’t disturb me in the slightest that people are using search engines to look for me directly. In spite of all the articles about how problematic it can be for people to discover your blog, I, stats-obsessed as I am, am more aware than most how absolutely public this writing is. And I don’t feel particularly flattered by these searches either. It’s not as if there are hundreds of them. I’m not famous, or anything.

But what’s interesting about these strings is how much a person needs to know about you in order to find you. I have a rather unusual name. As far as I can find out, I’m the only Rochelle Mazar in North America. A French place name paired with a quasi-Ukrainian surname is, I suppose, a bit unusual. You also have the fact that “Mazar” is not a proper surname at all; it was mangled at the border, making it even less likely that there would be a doppleganger out there for me to cope with. So typing in my last name or my first name does, apparently, eventually get to me. My full name gets you there even faster. These people know who I am, and are looking especially for me. Someone even knows where I work and is searching for me that way.

Others don’t appear to have met me personally, but seem to have heard of my blog. They search for it by title (“random access mazar”, and “diary of a subversive librarian”). They’re not looking for me per se, but for my blog. Real person search, virtual identity search. Both end up in the same place. This was a conscious decision I made; I linked up my real self with my virtual self by naming them the same thing.

Since I spent most of my time these days talking about issues relevant to librarianship, it makes sense that search engine algorithms would send some library queries my way.

librarian behind desk
academic libraries future 3 years
radical reference
radical reference librarians
proactive reference public libraries
reader’s advisor* alice walker
trillian library academic use
academic monograph
casual librarian
library anxiety

My suspicion is that these sorts of queries are the product of librarians making their way online. Note the use of a wildcard in the reader’s advisory string. There’s no direct Boolean involved, but a lot of it is presumed; trillian ? library ? academic use might as well be a subject heading. You can see the careful thinking behind these queries; most of them are keywords strung together.

The only one of these strings I think is not the product of a librarian’s search is the first one; that’s clearly a user looking for an image, but using the wrong feature.

I’m still most interested in search strings that show very little processed thought. I’ve recently had discussions about just this; is this the dumbing down of academia, turning to algorithms to parse our search strategies? Or is this another form of scaffolding, letting people think less about creating the perfect search string and more about the topic at hand?

teenagers 2005 what’s important to them?
evaluate about since the invention of internet libraries and textbooks have become obsolete
turn the handle and the couch becomes a bed ikea sofa
xanga can hurt people

What I love best about these strings is how spontaneous they are; can’t you just see an internet user staring at that search box, thinking about their question, adding useless words into the string. No, I take that back; what I love best is that those searches work. That second string, evaluating the idea that textbooks are obsolete in the face of the internet; I have written about that, and this awkward, clumsy search lead them to me. There are enough keywords nestled in that unprocessed thought to get somewhere useful. That’s a powerful search engine at work. This is cyborg searching, an algorithm so responsive it accepts the unprocessed nature of human thought. Tapped right in, plugged into your brain.

I’m writing about search strings; in the tradition of metacritique, of course I get searches from people interested in, well, search strings.

search string apostrophe
search strings
funny search strings in google failure
my favourite search words
weird search strings
string search narrow algorithm

Though, that first string is probably someone hitting the same mySQL problem I had, where apostrophe’s bork the database command, and looking for the script to remove them. I’ve got it. I can hand it off. Just email me.

The rest: I think it’s interesting that someone is looking at how search strings show the obvious failure of the system. We distrust it, we librarians in particular; a machine can never be so smart as to not have a panapoly of errors to giggle about. In some ways I think we want it to fail, again, librarians in particular. If Google fails, we will still feel we have a legitimate place in the world. It must fail, it must be laughable.

Does someone out there have a set of favourite search words? What would make a search a favourite? I think that’s a search doomed to failure. My (somewhat educated) guess is that most people don’t think about their search strings at all. I suspect that getting back a list of their own search strings would be a foreign and off-putting experience for most people. It would be like getting an itemized list of what they had thought about in the last three hours; recognizable, but not nearly as linear as a list would make them seem. Putting stuch things in a list would make the process unrecognizable.

new jersey bar exam and july 2005 and opinions

However, some people really do think hard about their search strategies. Interestingly, while this search is perfectly constructed, as opposed to most of the other strings I’m displaying, the results were obviously poor if this person was led to me.

One of my other favourite things about looking at search strings is seeing the ones that are more statements of belief or feeling rather than actual searches.

i hate reading
wow can’t express by words you look beautiful
start revolution

The total randomness of these strings delights me.

Often my musings about search strings leads me in the direction of thinking that Boolean is dead, that real searching these days is more about throwing out lines of thought and seeing if anything bites; I spend my time illustrating how thoughtless searching is for most users, how they don’t metaconceputalize before turning to Google (or Yahoo or any other search engine). But then I get strings like this:

powered by wordpress inurl:ca

This is a brilliant bit of Google-Fu. Someone wanted to see all wordpress blogs with Canadian domain names. So they typed in the tag line of all wordpress blogs (“powered by wordpress”) and limited the search to domains ending in .ca. This is simply genius. This user got exactly what she was looking for; I bet it’s an interesting list, too.

And then we have the strings which are direct or indirect questions; they are turning to the internet because they have a specific activity or plan in mind. I’ve been thinking of these as the true reference questions of the web. These users are approaching a search engine not as a database but as an answerer or questions.

writing a narrative essay letter to my boss
how to fill out a reference letter for school
why does my computer random boot?
how can i clear my search strings
hide phone number from bots
post a comment blog

I am looking to do a particular thing; the instructions will surely be on the internet.

I’m still keeping an eye on the strings, obviously. I feel that there’s something for me to learn in them. I’ll keep you posted on precisely what that might be.