Man Bites Blog

A couple days ago, Jeff Atwood posted the following puzzle:

Let’s say, hypothetically speaking, you met someone who told you they had two children, and one of them is a girl. What are the odds that person has a boy and a girl?

Consider your answer carefully, without doing a web search, or reading the comments to this post. Don’t cheat — but be prepared to explain your reasoning, because the solution might surprise you.

The 800+ message comment thread that ensued is an interesting study in how people go about resolving cognitive dissonance.  I dare not even look at the other 800+ message thread from when Atwood posted the solution.

The problem is interesting, because it forces the reader to overcome two separate intuitive hurdles.  The first is that past events do not impact the probability of future events.   If I flip a fair coin nine times and the first nine times are heads, what are the odds that the tenth will also be heads?  50% — the first nine flips have no impact on the probability of the tenth flip.

So the typical reader has been “primed” to read this problem and think, “Aha! — you can’t fool me!  The sex of the first child has absolutely no impact on the sex of the second child.  The answer is 50%!”

But the question is about the aggregate outcome, not an individual outcome.   And the aggregate outcome of one boy, one girl is twice as likely as the aggregate outcome of two girls, so the answer is 2/3.

This is a puzzle, not a trick question.  But when presented with a puzzle like this, we tend to think it is a trick question, that we are being given extraneous information that is a red herring, and hunt for the flash of insight that will give us a shortcut to solving the problem with little further effort.

Except in this case, the flash of insight is the red herring.  If we allow it to blind us, and don’t actually think through the math of the situation, lay out the possibilities, etc., we’ll end up with the wrong answer.

I submit that the real world has a lot more problems like this, that are solved by old-fashioned elbow grease than by a flash of insight.  But there’s enough problems that are solved by a flash of insight that we sometimes feel stupid when we pick up our pencils and start grinding through the math.  But that’s usually what it takes.

Apple has always been a niche player.  Indeed, the “Mac vs. PC” ads make this explicit.  There’s regular computers, and then there’s Macintoshes.  They did not set out to dominate markets to the point where, like Coke or Kleenex, their brand name becomes synonymous with the category.   They wanted to be better!

But then the iPod happened, which in the mind of many consumers is synonymous with “MP3 player.”   To the point where the general term “podcast” is derived from the name of Apple’s product.  And when iTunes took off, it became synonymous with “online music store.”

But, as Jeff Atwood points out, this comes with some responsibility.  When Amazon became synonymous with “online bookstore,” when online writers referred to a book, they would link to Amazon’s site for it, where people could buy it.

So people started doing the same thing with iTunes when referring to songs and podcasts.  The problem is, if the reader doesn’t have iTunes installed, he is taken to an alert advising him of that fact.

Atwood thinks this makes Apple a bad web citizen, I’m inclined to think it’s consistent with Apple’s too-cool-for-school brand image.  It’s not surprising it would put a virtual velvet rope up between you and its online store.

But Apple can’t have it both ways, either.  It can’t position itself as the dominant player, and still pull off its niche-type behavior.

alastair has a post taking on Jeff Atwood of Coding Horror, and Atwood replies.

Alastair challenges Atwood’s place as an authority on coding, and cites some bold statements Atwood has made on his blog .  Atwood responds that he never claimed to be an authority, and that his bold statements are an example of the SmackDown method of learning, and his desire to hold strong opinions weakly.

The problem is that, for a blogger like Atwood, with thousands of RSS subscribers and daily page views, there is no such thing as holding an opinion weakly (to ape Atwood’s typical style of putting his own opinions in boldface), or at least of expressing an opinion weakly.  A blog post that reaches thousands of readers is by definition strongly holding an opinion strongly.  It inhibits, rather than encourages discussion.

To take an extreme example the Bush Administration couldn’t claim that it was just throwing out the idea that Saddam has WMD as some sort of academic exercise to test their theory and prompt discussion.

In the blogosphere, there is a bit of a conflation between popularity and expertise (ahem), and we forget that what makes a blog popular is completely different from what makes a writer authoritative, particularly in niche fields like programming.    Dennis Forbes makes a good point in a comment on alistair’s post:

It really is irrelevant whether Jeff feels himself an authority figure or not (it is debatable either way). What really matters is whether the community, or a loud enough subsection of it, sets him up as an authority figure.

Exactly.  If, “but Jeff Atwood says PHP sucks” is seen as a winning argument, then it matters what Atwood writes on his blog. 

In essence Atwood’s blog is such a powerful weapon that his initial smackdown ends the fight.  It may be the case that a consensus emerges in the comment thread against Atwood’s original conclusion, and Atwood may even come around to that conclusion, but none of that is featured as prominently as Atwood’s original post.

There’s a fine line between taking a bold stand to provoke discussion and bullying other people with your opinion.  Atwood’s posts often come off as the latter rather than the former.  Perhaps that is all that is possible for a prominent blogger.