Google’s location aware searching is one of the features that really makes it a great search engine, but it would seem that I have become too used to it. I just made a $10 mistake because I assumed that Google was returning results ranked based on geography.
I remember my surprise when it was first implemented several years ago. I would start typing some very generic search term, like “Children” and it automatically suggests terms including “Children’s Hospital Boston” and “Children’s Museum Boston”. This is kind of amazing, given that there are children’s museums and hospitals all over the country. Google is geolocating my IP address and then using its vast data on historical searches to figure out that people in my area are much more likely to be searching for those terms that “Children’s Museum San Francisco”.
Passover is coming up, and I just finished the book I was reading, so I started to search for a new book to read over the holiday. My usual trick of buying ebooks and then printing a few chapters to read over the sabbath isn’t very practical for a longer holiday like Passover (four days of no electronics this year), so I decided to reverse it. As some of my friends do, I started by picking a library book that I could read in hardcopy during the holiday and then purchase the ebook for reading the rest of the time.
As luck would have it, my sister-in-law works at a library in Newton, MA, so this is pretty easy for me. I just needed to search their online catalog for a book I wanted that was available, and then she would be able to pick it up for me. I googled “Newton public library”, went to the site, and started punching in titles. I was in the mood for something light, like Game of Thrones, but each book I put in was checked out. Finally, I put in Tinker Tailor Solider Spy, saw it was available, and asked her to get it for me.
She wouldn’t be back at the library until the next day, but it was evening, the library was closed, and I figured it was unlikely anyone would check it out before she got to it. In the meantime, I felt like reading right then, so I decided it was safe enough to purchase the title on my Kindle and get started.
To my surprise, my sister-in-law called me the next morning to tell me that in fact the book was checked out. Curious, I went back to the library’s site, and it was still listed as available. Huh?
I started to look at the website more closely, and in the upper right corner of the home page in very small type, it said “Newton, Kansas”. Oops.
I went back to the google search results page, and the next link below was another Newton library. While it was not clear from the search results, this second result was for Newton, MA.
Library websites all look the same – amateur designs left over from the Web 1.0 days. In fact, neither the Newton, MA nor the Newton, Kansas websites indicate where they are located beyond small address text on the home page, so if you don’t realize that there are two Newtons, it’s easy to get confused.
Why didn’t Google order the websites “properly”? I’m guessing that most of the people who search for a library live in that town, so perhaps it didn’t have good geographic search data for surrounding towns.
On the other hand, when I just tried Bing, the Newton, MA library was the first hit. I’m curious as to what would happen if I ran the search for a computer in Kansas.