Still on the ABC News.com legal beat: Call me a bad libertarian but there are few things I hate more than misplaced privacy arguments.
"I don't believe that the police should be able to collect somebody's private, very intimate information, their DNA, just because they have a hunch that it will be useful."
Ah, nothing says "this is my heart and soul" like a strand of hair or a bit of discarded spittle. I'll buy the case that people shouldn't be compelled to give samples, but if they happen to do it anyway, make the most of it.
Similarly, if your ID maps to a criminal warrant and you happen to give someone your ID, guess what? Given the complete and utter lack of harmful side effects (it's a very simple flow chart: either there's a warrant out against you, and it gets served, or there isn't and life goes on), and the degree to which it actually inconveniences you (i.e. zero), "probable cause" shouldn't be relevant to whether someone bothers to run a query.
This story brings up another reason to oppose the death penalty: Victims (of crimes other than homicide) or witnesses will be more reluctant to report a crime if they know the perpetrator and don't want to see that person executed.
My main reason to oppose the death penalty (but with no motivation whatsoever to act on this conclusion) is that capital cases invariable lead to this convoluted appeals process that wastes millions (not an exaggeration!) of dollars worth of people's time that could have been better spent on something that actually improves society rather than dithering over one particular person's fate.
(Cat pictures, that is.)
Three of the first four (this, this, and this) are each better than anything else I've seen there in the past few days.
(From The Onion AV Club)
I don't think this was an accident. (NSFW?)
Both Hillary and Agnetha have on entirely the wrong shade of lipstick for their respective contexts.
The best one on today's chart.
It's not immediately obvious why.
Anyway, Jane Eyre runs for president. Spot-on.
That what follows could even be called "a mild disappointment" reflects an astonishing idea, that one could hold Wikipedia (of all sources) to high expectations for breaking news coverage. On the big stories, though, Wikipedia tends to be both thorough and even-handed.
That said, I learned nothing new from the San Francisco zoo tiger attacks page, and in fact they listed one allegation (slingshots) that has since been denied without being corroborated.
Since I tend to lots of sympathy for Tatiana and very little for the dead guy or the brothers, I was hoping to have my bias tempered; instead, if anything, one can easily infer that the most recent editors of this page have a worldview similar to mine.
I haven't watched this and don't plan to.
One National Review On-Line writer suggests that Thompson "taped a 17-minute video in which he makes his case more calmly, deliberately, and and with incomparably greater respect for the issues than has any of his opponents. And? For a lot of Iowa Republicans, that’s all they needed."
Really? Maybe my attention span is criminally short but I can't see myself making it through, much less being won over by, 17 minutes of a guy talking at a camera.
This has a "can't be right" intuitive feel to it but I report what I see:
According to Gateway Pundit, "[f]or the last three months of 2007, a Venezuelan was twice as likely to lose his life to violence as an Iraqi." (emphasis removed)
Sources appear to be iCasualties.org and El Universal.
My first guess at a refutation would involve whether Iraqi deaths reported as thoroughly as Venezuelan.
In the grand scheme of things they matter far less than their proponents think -- or at least, they ought to matter a lot less. (Power to the people: we have significantly better access to the players and games of our time than ever before -- and even to a very large body of classic games; more important, we have much better access to enough analytic tools that we don't need a HoF either to be the gatekeeper or to prevent the deserving from being forgotten.)
That said, I have a few brief points to make. Feel free to argue with me, as long as you understand that I'm right, that it's not even close, and that I feel no need to discuss any of these further:
Bert Blyleven = yes
Tim Raines = yes
Jim Rice = no
Don Mattingly = no
Barry Bonds (when the time comes) = yes [with as many disclaimers as you see fit: disclaim him until the cows come home, but to omit him completely would be antithetical to any conceivable purpose of a HoF]
Anyone else is either uncontroversial enough to catch my eye, or borderline enough that apathy wins out for me. (This includes McGwire and Sosa: My blink-of-an-eye reaction would be yes to both, but I'm not so married to either of them that I'd find excluding them unjust, even if I found the reasons to exclude them laughably contrived.)
P.S. Here's a frightening collection of actual arguments made by actual voters.
You may have already heard this (I expect someone to start complaining about it in his on-line sports column soon) but the Patriots have four West Coast road games in 2008. This is just catching up: For the pre-set part of each team's schedule (14 games out of 16), New England hadn't been given "at Oakland / at San Diego" since 2002, and hadn't gotten "at San Francisco / at Seattle" since even earlier.
(As you either already realized or didn't care about: For the pre-set games, at least whether a given game is at home or on the road, the first two and last two teams in each division (alphabetical by place name) are basically travel partners. This adds more variance than strictly necessary to road game travel length, since "at Oakland / at San Diego" is a lot further for an East Coast team to go than "at Denver / at Kansas City," which in turn is further than not having any AFC West road games at all.)
In other news, to be elaborated on either when you care or when I get around to making a clear concise explanation, an NFL schedule is sort of like a Sudoku variant (at the level of abstraction where people like Jason Z. compete) -- if, by "schedule," you're content to mean "eight pairs of slates, within which each team has a road game and a home game, that you can then arrange in the order that suits you."
A half-decent explanation of this is after the fold, including why my fairly straightforward way to get those eight-pairs-of-slates actually has 128 solutions (if you don't n-tuple count some permutations) most years, but 64 this year.
Off-the-cuff first draft of how to do something pretty neat (at least if you've read this far you'd probably think it's neat):
NOTE: Unless/until the NFL votes to extend the current system, this works only through 2009. 2010 onward all bets are off.
NOTE 2: This will be gobbledygook unless you're playing along as we go. If I were less lazy I'd attach screen shots.
1. Learn which teams have which opponents in the season you care about. Most sports outlets publish this right after Week 17 of the previous season, but you can also figure it out through a combination of this site and the final standings. Paste whatever you need to paste into a convenient reference source.
2. Make an 8-by-4 grid based on NFL divisions and previous season order of finish. Group pairs of divisions together, specifically the two divisions in the same conference whose teams all play each other. (Examples: 2008 AFC East vs. AFC West; 2007 AFC East vs. AFC North.)
Now you have a coloring puzzle: Divide each group of 8 into two groups of 4, such that each group has one team per ordinal standings rank but at most one team per travel partnership (travel partnership = e.g. "Buffalo and Miami" or "Oakland and San Diego"). There will always be at least one way to do this, sometimes two.
3. Make a 32-by-16 grid for NFL teams and schedule weeks. (Group the teams by division.) For clarity, thicken the borders so that this resembles an 8-by-8 grid of 4-by-2 grids.
4. Pick an arbitrary color (let's say blue) and an arbitrary conference (let's say AFC). Over the first four pairs of weeks, shade teams in such a way that:
A. Each team has shaded cells for two pairs of weeks (one pair among the first four, one pair among the next four)
B. In a given week, the group of eight teams with shaded cells amounts to two of the groups from step 2. (For example: {BUF, NE, CIN, PIT, JAX, TEN, KC, OAK}.) [If you already understand what we're trying to accomplish, you'll already notice that because of the 2007 NFC standings, the 2008 NFC will need a slight twist, e.g. {DAL, PHI, MIN, DET, CAR, NO, SEA, STL}. Do you see why?]
C. For a two-week period in which eight particular teams are shaded, in one of those weeks those teams will all play division games; in the other, they'll all play games they don't have in common with the rest of their division. (e.g. Pit at NE, Buf at Jax, Hou at Oak, KC at Cin)
D. In other words, the cells with this particular color take care of all the non-common games (two per team), as well as home-and-home for one division rival per team.
5. More generally, in each two-week section there will be five groups of teams (color-code them for your convenience!).
A. 8-team group, as mentioned in step 4
B. 4-team group, all in the same conference as group A, who play intra-conference games against each other. (Example: if the 8 specific teams mentioned above are already group A, then this could be {Ind @ Cle, Ten @ Bal; Bal @ Ind, Cle @ Ten}. Or it could be {Den @ NYJ, SD @ Mia; Mia @ Den, NYJ @ SD}.)
C. 4-team group, some entire division whose non-conference opponents are all in the same division as two of the teams from group B. These teams will all have division opponents two straight weeks, specifically the teams they DON'T face when they're part of "Group A" themselves. Note: You can't actually fill in who plays whom yet. That is, you can't immediately decide that this is "Dal @ Was; NYG @ Dal" rather than "Dal @ NYG; Was @ Dal."
D. 6-team group: two leftover teams (same division) from the same conference as Group A, plus the entire division that faces those two teams in non-conference games. (Example: {Denver, San Diego, AFC South}.) The former two teams will have non-conference games each week, one against the "first half" of the other division (alpha by place name) and one against the "second half." That will leave gaps filled by division games so that all six teams have a home game and a road game. (Example: {Den @ Car, Atl @ SD, TB @ NO; NO @ Den, SD @ TB, Car @ Atl) Note: Same caveat as for part C.
E. 10-team group: Sort of like Group D, except that in the division where half the alphabet has non-conference games, the other half will have conference [non-division] games instead of just facing each other.
6. You'll want to color-code this so that each team is either on the little end of "Group D" exactly once, or on the little end of "Group E" exactly once. To accomplish this, despite everything explained in step 5, you probably just want to look at whichever conference doesn't have "Group A" teams for a given two-week slate, and think of what the different divisions are doing.
(One division is all in-house; one has non-conference games + division games; one has non-conference games + non-division games; one has non-division games + division games.)
At first it would be tempting to claim that there are 24 ways (4!) to arrange the color-coding for those divisions in those weeks, but then you should immediately notice why it's really just 8.
7. As you might notice, opponent assignment and home-road considerations for Group A and Group B are independent of any other group. But opponent assignment and home-road for Groups C-E all depend on each other. In fact as it turns out, once you have the color coding set, you're down to two flavors. You can completely solve for either of those flavors by starting with a "Group C" color (entire division with two weeks in a row in-house) and making a binary choice. (e.g. the aforementioned "Dal @ Was; NYG @ Dal" versus "Dal @ NYG; Was @ Dal.")
Getting from these 8 pairs of slates to a 17-week actual schedule is an exercise for the reader, mainly because there's no elegant way to space things out for byes. (I'd turn one slate in to Weeks 1-2; then 3-4, 5+7, 8-9, 10-11, 12-13, 14-15, and 16-17. Then group a few games each from weeks 3-5, 7-10 to fabricate Week 6. But there's no elegant algorithm for this.)
This NY Times article summarizes the contrast between how two Democratic presidential candidates generally approach economic issues.
Candidate A:
believes in the promise of narrowly tailored government policies, like focused tax cuts.
has more faith that government can do what it sets out to do, which is a traditionally liberal view.
subscribes to the conservative idea that people respond rationally to financial incentives.
Candidate B:
draw[s] heavily on behavioral economics, a left-leaning academic movement that has challenged traditional neoclassical economics
[believes that] a simpler program — one less likely to confuse people — is often a smarter program.
For a quick 10 points, which is which? (And with which archetype do I agree more, and by how much?) [Answers in extended entry.]
With which archetype do you agree more, and by how much? [Answers in your comments, should you choose.]
A = Hillary, B = Obama.
Completely aside from my opinions of those specific candidates, I side with B by a wide margin.
The team that has led the NFL in passing yards has never won the Super Bowl.
Well golly gee, I wonder why. (Hint: It has nothing to do with the effect of high passing yardage, but rather the main cause.)
(More information about the revolutionary "Kneel to Win!" theory. Anyone interested in correct use of sports statistics should take the lessons of this article to heart.)
A couple weeks ago the Sunday NY Times business section had a long article about Google vs. Microsoft. Part of the conflict is a paradigm collision between people who see their productivity happening mainly on a local machine, versus those who see it happening mainly on-line.
(Of course that's not just a Google-Microsoft thing. You can add Apple to the "local machine" paradigm and a variety of web companies to the other side.)
Anyway, I think one of the best pieces of concrete evidence in favor of web-based computing is Microsoft Outlook (and everything wrong with it). [N.B. I have never used Leopard, so I have no idea how good Apple's e-mail client is apart from my high esteem for the developer(s) in question.]