If someone asks you, "Are you short?" or "Are you tall?", what would you say?
Your answer might be that you're short. Or "short-ish", or maybe "tall-ish". While traditional hard categories generally work well, they frequently fall... short... where they get blurry around the edges.
We see this today in our classification of gender. Pretty much since the first computer systems were built, gender has been Male or Female (or "Unknown"). Now we have LGQTB, etc etc.
Pity the poor programmers who have to cater for this. Their challenge extends well beyond the ability to store what has effectively become an arbitrary definition. Then if the logic of their system relies on a clear distinction between female and male, things can get really messy.
It's especially difficult for computers because they need things to fit into neat, tidy boxes. Computers are usually programmed using "Boolean" logic, which follows a formal chain of reasoning such as "If A, Then B; Else C".
Imagine then, how they struggle when the criteria states that something can be "A-ish".
You can see the difficulties if this relates to whether a person is eligible for a pension based on their gender - do they get a pension or not? Or do they sort-of get one?
It can lead to horrendously tangled, loopy logic inside computer systems that nobody really understands.
The need to deal with the soft edges of definitions was recognised early in the history of computing. In 1973 Lotfi Zadeh proposed a theory of fuzzy logic with a range of innovations that included fuzzy pattern recognition and mathematics that could be used in control systems.
These aimed to more closely mimic human reasoning that allows for real-world situations where things don't fit neatly into categories.
Zadeh was born in Azerbaijan in 1921, and soon showed his exceptional talent as a student. In 1942, he graduated from the University of Tehran before moving to the United States where he eventually became a professor of computer science at the University of California, Berkeley.
Zadeh is remembered as a pioneer the field of in signal processing. If you're reading this on an electronic device, it almost certainly uses techniques developed by him.
Fuzzy Logic is also a science program on radio 2XX. Its origins are unclear, but its heritage goes back several decades, which probably makes it one of Australia's longest-running science shows. Sort-of.
2XX will be broadcasting live from the Multicultural Festival in Canberra on Sunday. Drop by and say hello.
Listen to the Fuzzy Logic Science Show at 11am every Sunday on 2XX 98.3FM.
Send your questions to AskFuzzy@Zoho.com Twitter@FuzzyLogicSci
We've made it a whole lot easier for you to have your say. Our new comment platform requires only one log-in to access articles and to join the discussion on The Canberra Times website. Find out how to register so you can enjoy civil, friendly and engaging discussions. See our moderation policy here.
