In philosophy, the common definition of knowledge is that knowledge is a 'Justified True Belief'. As long as these three conditions (J, T, and B) are met, you can claim to have knowledge. But in 1963 Edmund Gettier published what is now called the Gettier problem, which seems to show that J, T, and B are not enough. *

You work in the office with a great guy called Mr. Nogot. In the past he has explained network cards and nuclear physics to you, and kept you informed of the donut count in the staff room. He has not lied to you, and you feel justified in trusting his word.

Today Mr. Nogot informs you that he is the proud owner of a new Ford car. At lunch, you see him drive off in a Ford. And while digging through the papers on his desk you come across the ownership papers to a Ford.

Because of all this, you form a justified belief that Mr. Nogot does indeed own a Ford. Because of this, you also form the belief that "someone in this office owns a Ford". (If you were asked who, you would say Mr. Nogot. But if simply asked if someone if your office owned a Ford, the answer would be yes).

But it turns out that Mr. Nogot was only fooling you... It's not his Ford he was driving, and those papers were forged. So you can no longer claim that you knew** that Mr. Nogot owned a Ford (Because it wasn't a justified true belief).

The next day you see your co-worker Mr. Havit in the parking lot, where he is parking his Ford. He, five reliable witnesses, and the parking attendant explain that he has had this Ford for the past year. You are convinced. You now have a justified true belief that Mr. Havit owns a Ford. And, therefore, that "someone in your office owns a Ford".

So... Did you know before this that someone in your office owned a Ford? You did have a justified true belief that someone did. It was justified for Mr. Nogot, and true for Mr. Havit, but it was clearly both justified and true. It just so happens that you were just basing that belief on false evidence.

So is justified true belief enough for knowledge? Even when it's based on false Ford ownership? And if it isn't enough for knowledge, what is required?

* This is not the original Gettier problem. I'm not exactly sure what form that took, but I think it involved Mr. Nogot and a Ferrari.

** You didn't have a justified true belief, and if you didn't have a JTB, you didn't 'know'--that is, at that point you didn't have knowledge. A possible problem here is that not all people have the same definition of 'knowledge', and that not all people use the word 'know' as only identifying an instance of 'knowledge'. In this case we have defined knowledge as a JTB, and we are using 'know' and 'knew' as forms of JTB.