Why Windows on a Mac?

Slashdot reports on a bounty for successfully booting Windows XP on an Intel iMac. One wonders exactly why such a hack would be of the least interest. Certainly not because one wants a Windows PC and finds that buying hardware from Apple is cheaper than a Dell purchase. In fact, given the hardware margins on the Intel iMac, Apple would probably be laughing all the way to the bank if this hack were actually used by people.

Wouldn't it be far more interesting to try and hack a Dell/Walmart PC to run Mac OS? That would at least save people a few hundred dollars. (And if the goal was to create a dual Windows/Mac boot, such a hack would probably help too.) On the other hand, Apple probably devoted a lot more attention to making it difficult to pull this off.

Football: Quick Picks

Denver (-3.5) over Pittsburgh:

I just don't see Pittsburgh -- which holds the dubious distinction of being the only team to lose a playoff game at home to New England in the Tom Brady era -- putting together three consecutive good games. They managed to beat Indy despite their best efforts to choke it away; I think they will succeed in choking this time around. Denver's offense also ought to be much better than Indy's proved to be last week.

Seattle (-3.5) over Carolina:

Why?
(1) Bill Simmons picks Carolina.
(2) DeShaun Foster is out. When means that Jake Delhomme is going to throw Carolina out of the game.
(3) Seattle is a strong home team.
(4) Carolina can't win three in a row.

Last Week's Picks: I went 3-1 thanks to Indy's screw-up (which I'd like to think was because of the rust that I had predicted parenthetically in my evaluation of the game).

On Chucking in Cricket

Prem Panicker protests against the new chucking law that bans players for a year when they are caught chucking, instead of merely requiring umpires to no-ball the specific deliveries that are chucked.

I'm afraid I'd have to disagree on that score. What's missing from the equation is the issue of how difficult it is to accurately estimate whether a specific ball is chucked or not (which has finally been defined somewhat objectively as elbow flexion greater than 15 degrees). It's simply impossible with the naked eye.

What's the solution then? Look at the video after the game and evaluate the footage to make the determination. The problem then is that the bowler gains an unfair advantage during that game. Bowlers could then chuck with impunity in critical junctures and get away with it scot-free.

In consequence, there would have to be sufficiently large disincentives that prevent the bowler from attempting to use such a strategy. Banning him for a couple of years fits the bill perfectly as a good disincentive. Hence, the rule we have in place today.

Such an incentive system is hardly uncommon. For example, a ticketless traveler on public transport, when caught, has to pay a fine tens of times higher than the ticket price. Again, the reason is to ensure that the "expected cost of cheating" -- the product of the price when caught, multiplied by the probability of being caught -- is high. When the probability of being caught is low, we need to boost the penalty you pay! (Of course, we might wonder why the probability is low in the first place. The answer is that we can reduce enforcement costs this way! In fact, we can spend less and less money enforcing the law by proportionately increasing the fines more and more until we run into the problem of people not being rich enough to pay up!)

For the record...

Given how predictable NFL playoff games have become, I figured I should take a leaf out of Bill Simmons's book (his picks here) and make my picks against the Vegas line. (For those who don't know what it means, a margin of victory is predicted by the oddsmakers for every game and you get to bet on whether the margin will be larger or smaller. So, if I do better than 50-50 on my bets, I get to make money!)

Seahawks (-9.5) over Redskins:

The end of the road for the offensively-challenged Washington squad. Seattle is going to trample all over them, winning by at least two touchdowns.

Broncos (-3) over Patriots:

This is a close call, given Denver's history of playoff collapses and the Patriots' impeccable record, but I'm siding with the home team. Denver's unstoppable running game is going to dictate play and leave New England's already depleted secondary even more vulnerable to the play-action pass.

Colts (-9.5) over Steelers:

This one is more a pick against the masses than anything else. Everyone has been overpraising Pittsburgh the past few weeks and overconcerned about Indy's finish. (Aside: What's the deal with resting players for the playoffs anyway? My theory is that it causes much more harm than good, taking players out of the winning rhythm that they had worked so hard to build up over the first 15 weeks of the season. How often have we seen the hot wild-card team ride its streak deep into the playoffs? If Indy gets off to a slow start, I would be vindicated.) So, I'll make the safe assumption that the line is a little too generous to Pittsburgh and go with Indy.

Panthers (+3) over Bears:

I'm picking the Panthers to cover. Why?

(1) Carolina played their worst game of the season the last time they were down in Chicago and I just don't see them playing that badly twice in a row. Even though they played well last week and usually blow hot-and-cold with breathtaking periodicity.

(2) What are the odds on Chicago scoring more than 13? A 3-point spread against such an anemic offense seems like a lot.

(3) Bill Simmons has gotten every pick involving Carolina wrong for the last 3 months, and he's picking against them this week.

Starvation on City Streets

No, not of the gastronomic variety. I mean the Computer Science version.

I was driving south on the three-lane Paradise Road in Las Vegas yesterday when I encountered a curious phenomenon: the right (and right-turn) lane was extremely slow and barely crawled along while the traffic on the left two lanes was perfectly smooth. My initial hypothesis was that the intersecting road (Convention Center Drive) must be clogged badly, which in turn must have caused right-turn traffic to back up on Paradise Road. But that hypothesis was soon given the lie when I noticed cross-traffic zipping by whenever it had a green light!

And then it hit me. The fault, I realized, lies not in cross-traffic but in pedestrians! I remembered that a green light for a right turn is coincident with a green light for pedestrians wishing to cross straight on as well. And pedestrians have right of way. So, what happens when you have an immense amount of foot traffic looking to cross at all times, as was the case yesterday with CES in town? Traffic going straight on was fine, as were left turns (which had dedicated signals) but the right turns were completely trumped by the pedestrians. (What of a right turn on red, you wonder. That option was comprehensively ruled out by the volume of cross-traffic from the other two directions.)

Unfortunately my powers of deduction did not help me extricate my car from the jam, trapped as I was in the right lane with traffic whizzing by my left far too fast for me to venture a lane change. And so, it took on the order of 15 minutes to travel 25 yards for a turn.

Moral of the story: We need dedicated right-turn lights!

NFL Picks Update

So, I was wrong on Cincinnati beating Pittsburgh. But at least I can hide behind the injury to Carson Palmer. I'm picking Indy over Pittsburgh in the next round. I'm also going to make Denver the favorite over New England. Their run-heavy, play-action offense matches up very well against New England and, despite the presence of Tom Brady, New England's offense was nothing to write home about in the wild-card game.

On the NFC side, I did have both the Panthers and Redskins winning (I note, however, that I'd failed to update my predictions since Tampa Bay and Carolina swapped places). Next Round: Carolina over Chicago by a whisker and Seattle over Washington by ten.