Wednesday, April 12, 2006

Game of the Name

I was in a chat recently (like a minute ago) when the phrase confirmation bias came up. Now, this is something unknown to me, and sounds a bit like a religious rite, like, I want to confirm you as Eastern Greek Orthodox, or some such. I can't say I know much about confirmation.

These days, I could plead ignorance and merely ask "well, what the heck is confirmation bias" and get an explanation. But we live in the information age, and that information is available. I would Google it but for stuff like this, it's just that much easier for me to head to Wikipedia.

I know, Wikipedia's been getting bashed as of late. People put in some faulty information, especially about other people. People working in Congress were banned from Wikipedia because they kept puffing up the representatives they work for.

But I find, because of the way Wikipedia works, that I'm likely to find what I'm looking for there. And thus I have. In a nutshell, confirmation bias means you look for evidence to back your own claims while ignoring evidence against it.

This happens a lot when people seek victimhood. Why does the media bash Clinton? Why don't they ever talk about Bush's infidelity? They only ever seem to notice negative articles, and never the balance media can provide.

I find academic types like giving names to things. If it sticks, they can point to it and say "I said that!". Thus, confirmation bias. Naming things shouldn't be underrated, because it is useful to name something so people can talk about it. For example, once I define design patterns, we can talk about it. I don't need to explain it every time we talk. It becomes shorthand.

Still, when there's an obsession to give names to things, you realize things have gone too far. Nowhere did I see this trend more than at ETECH, and Bruce Sterling was nearly as guilty as Tim O'Reilly. He wanted people to adopt his term, spimes, referring to objects trackable in space and time so one can query for it. He had referred to something as "everywear", I think. This has been incredibly difficult to find on the web. The closest one I've found has to do with wireless stuff you wear by a company named Charmed.

I'll need to see a transcript of Bruce's talk to make sure. But at least the name isn't nearly as hideous as Web 2.0, a term I heard back in September or so, coined by Tim O'Reilly.

I did like "continuous partial attention" as a phrase to refer to people who are constanly monitoring several forms of communication (email, IM, cell phone, other people, etc), even if the name sounds like a math function. (Nothing beats probably appoximately correct learning or PAC-learning for a great name. Shouldn't that be called "about right" learning? Ah, but in this case, probably refers to probability, which means it's highly probably, and approximately may have a stronger mathematical definition, and correct is relative to some definition of truth).

For some reason, this phrase from Billy Joel's Only the Good Die Young (whose title may have more relevance than imagined, but I digress), which goes:

You got a nice white dress and a party on your confirmation

Have you ever noticed how some words only come up in some contexts? Confirmation tends to be associated with religion. I was thinking of the word "vicariously". How many people think it always goes with "to live vicariously" typically through someone else?

I'm sure there are other phrases like this, but no more come to mind.

No comments: