Classification Systems
We have a tendency to classify things into categories. It helps organize information and find things easily. Most of us are aware of one or the other classification scheme in areas like music, sciences or literature.
One of the most famous classification systems is the Dewey Decimal Classification system (DDC) used in libraries all over the world. But is it “fair”? Look at how it divides the religion space. 70% of the religion space is reserved for Christianity. Islam and Hinduism are clubbed with “associated” religions (like Bahai and Buddhism) and get less than 2% of the religion space (even though, put together, they are followed by more than one third of the world’s population!) So was Dewey biased? Maybe, maybe not. His system was devised to optimize shelf space for books. No prizes for guessing which religion had the most books in the land Dewey lived in…
So you realize classification systems are devised with an aim in mind (like optimizing shelf space). If you have a different aim, then an existing classification system may not work for you.
Contrast the above with the ad-hoc classification that happens in the digital world. Would you classify “Animal Farm” as fiction? Satire? Political? In the digital world, users decide which category an item falls under. Amazon, for example, classifies books based on what readers say, not what the publisher says. Further, the same item can fall into multiple categories. Digital indices like Google factor that in. And even update themselves as categories change over time. The digital world is far more democratic, liberal and open to change than the physical world in matters of classification.
All such systems, physical and digital, are often very good for organizing data that was already out there when the system was designed. But they start degrading as new entries are added after the system was designed; the reason being that the new entries don’t really “fit in”.
But the best classification system I know of comes from the physical world: the Periodic Table in chemistry. Its organization provides so much information about an element’s properties. And even predicts the properties of any new element that might be discovered. Nothing beats a predictive system!
One of the most famous classification systems is the Dewey Decimal Classification system (DDC) used in libraries all over the world. But is it “fair”? Look at how it divides the religion space. 70% of the religion space is reserved for Christianity. Islam and Hinduism are clubbed with “associated” religions (like Bahai and Buddhism) and get less than 2% of the religion space (even though, put together, they are followed by more than one third of the world’s population!) So was Dewey biased? Maybe, maybe not. His system was devised to optimize shelf space for books. No prizes for guessing which religion had the most books in the land Dewey lived in…
So you realize classification systems are devised with an aim in mind (like optimizing shelf space). If you have a different aim, then an existing classification system may not work for you.
Contrast the above with the ad-hoc classification that happens in the digital world. Would you classify “Animal Farm” as fiction? Satire? Political? In the digital world, users decide which category an item falls under. Amazon, for example, classifies books based on what readers say, not what the publisher says. Further, the same item can fall into multiple categories. Digital indices like Google factor that in. And even update themselves as categories change over time. The digital world is far more democratic, liberal and open to change than the physical world in matters of classification.
All such systems, physical and digital, are often very good for organizing data that was already out there when the system was designed. But they start degrading as new entries are added after the system was designed; the reason being that the new entries don’t really “fit in”.
But the best classification system I know of comes from the physical world: the Periodic Table in chemistry. Its organization provides so much information about an element’s properties. And even predicts the properties of any new element that might be discovered. Nothing beats a predictive system!
Comments
Post a Comment