Search Engine Tinfinger - the Human Omnibus
What is Tinfinger?
Tinfinger is a human omnibus.
What does that mean?
Just what it says: it is an omnibus about humans.
Tinfinger is for you to find Internet content about humans.
So, it’s like a phone book then?
No, we’re not a White Pages.
We’re interested only in famous people, where “famous” means that you’re mentioned in news stories or important blogs. People who participate in major sporting leagues, get their scientific dissertations published in journals like Nature, hold high office in a political party, are prominent members of a religion, hold executive-level jobs at a company, appear in court, create works of art, are associated with one or more technologies, or are just your every day celebrity… all of these people qualify.
What kind of content do you store?
To start, Tinfinger will store lists of links to news articles and blog entries outside our site, and we will enable you, the user, to choose pictures and upload your own articles about your favourite people. We might include other kinds of content later.
So it’s like Google, or Yahoo, or… ?
You can perform a text search on our data by typing words into our search box, like you can at Google or Yahoo or any other search engine.
However, our site is not just about search.
What is it about, then?
Individual humans are classified at Tinfinger into categories, like how books are classified by librarians using the Dewey system, or living things are classified by biologists using the Linnaen system. Experienced Internet users would be familiar with what Yahoo’s directory used to look like and what the Open Directory Project looks like now. Although we use the same basic structure as those directories, the focus of our directory is people, not topics.
Big deal! What’s the difference?
Other search engines require you to type in “keywords” into their search box to give you a list of results. Our directory structure focuses on names. Because people are classified into categories in our database which could also double as keywords in other search engines, you will get some of the same effects in a Tinfinger search as a keyword search on other sites, but we hope that our categories contain better-targeted, fresher content.
What’s on a Tinfinger headline news page?
The basic layout of a category news page is simply a listing of news stories, and will be familiar to users of Google News, Topix.net, Blogniscient or Memeorandum - particularly the last of those four. The headline and a small snippet of text from each story is listed down the page. On our site, each of those stories is associated with a different person who falls within that category. The subject’s name is highlighted in the news snippet with a link to his or her Tinfinger profile page. Below the news snippet, there is a list of recent news articles and blog entries which also mention that person, ranked by relevance and timeliness.
So each person is mentioned in just one category?
Each person is classified into one primary category, yes. However, each category page lists the most popular stories not only about people classified into that category, but also all subcategories below that category. So the Sports category can contain stories about people in the Sports / Golf category, the Sports / Olympic / Winter / Luge category, and the Sports / Football / Soccer category.
But what if I don’t want to bother with all this category stuff?
Each person (and all content associated with them) also has a number of keywords associated with them. These keywords have come to be known as “tags” at other sites. You can bypass the category system and simply search for tags in the search box at the top right of each page.
What if I’m looking for a person whose name appears more than once?
That’s a big issue for Tinfinger. The online encyclopedia site Wikipedia has already had to deal with this problem, which they call disambiguation - that’s a long page! Their solution is to put extra information about the subject in brackets as a suffix to the subject name itself. Our solution is similar: each person has a “truename” in our database, and their profile and other information is all accessed by putting their truename in the URL. If the person’s name only appears once in our database, their truename is the same as their name, with dashes instead of spaces. For duplicate names, things are significantly easier for us because we can already identify people by which category they are in, so that we can easily tell Michael Jackson the pop singer from more than a dozen vaguely famous people with the same name who would be in different categories by appending the name of their category to the end of their name to create a unique truename.
