October 31, 2009

Facebook Group Hysteria

Facebook is a place where you can join groups to tell people about things you like and dislike, things that you want and want to avoid. But group-joining is sometimes more than that. Facebook groups sometimes seek to cause change, usually by promising some effect for some critical number of followers. Witness these real Facebook groups:

If this group Reaches 500,000 members I will name my son Batman

Think of the children!

For Every 1,000 that join this group I will donate $1 for Darfur.

It's a worthy cause (er, what was the specific charity again) but I think my clicks are worth more than a tenth of a penny!

I love JESUS! If you love JESUS join this group.

Wait - you wouldn't be exploiting people's religious beliefs just to see how many people will join your group, would you? No.

If a billion people join this group Mark is allowed to get a Labrador

I dunno, Leah. If you don't want Mark to get that Labrador, you might want to set the goal a little higher than a sixth of the world population.

Facebook: SWITCH BACK TO THE OLD NEWS FEED!!!

As far as Internet emphasis goes, three exclamation points is not all that many; however I'm pretty sure this cause doesn't even warrant that many.

Here are some groups that have yet to be created, but they are clearly inevitable:

  • If 5 people join this group, my wife will get off the couch and get me a beer
  • For every pi people who join this group, I will donate the square root of negative one dollars to mathematics education
  • If 666 people join this group, the Lord of Darkness will trade your souls for my 15 minutes of fame on a reality TV show
  • If billions of people join this group, McDonald's will bring back the McDLT!!!!!!!!!!!!
  • JOIN THIS GROUP IF YOU DON'T WANT ME TO PUNCH YOU IN THE FACE
  • If this group reaches 3,000,000 people, unicorns will be real
Posted by James at 10:23 AM | Comments (3)

October 27, 2009

That Song Must Die 10

I have to knock the balloon boy story off the top of my blog. What better way to do this than with a new "That Song Must Die!"

But... oh no! One of these songs is growing on me!

1) "It's Me, Bitches"

This might actually be a decent song if it weren't for the singing/rapping. And the weird whooooooop! noise. Which means I guess I think the drums are OK. Not safe for work language in this one. Don't people get bored with this stuff. "It's me, bitches" is essentially a rap version of "Pop goes the weasel." It even has a kid dancing. To profanity. Charming!

2) Take U to Da Movies

What the...? If you listen to just one Sudanese rapper today, listen to this one. BANGS is his name, though he inexplicably does not wear them in his hairstyle. There is something compelling about the innocent title and refrain, and the electronic tune played on a $20 Casio keyboard from 1983.

Maybe I've been working too hard, but now I feel like a trip to the movies. "YEEEAH." Lord help me, but this song is starting to grow on me. But it is not for me to say which song must die -- that's your job. Tell me: WHICH SONG MUST DIE? And why?

Unrelated Bonus: While searching for "bangs" I found this picture, and I'm pretty sure this girl is made of a space age polymer.

Posted by James at 12:42 AM | Comments (8)

October 17, 2009

Bad Headlines: People Don't Even Need Help

My last post expressed my unhappiness with the idea of headlines being chosen by the number of people who click the headline. my opinion on that is that headlines become part of the story and so click-based headlines is not going to get you more accurate headlines, just more sensational headlines.

But bad headlines are not new, and you don't need robots to create them. I thought I'd post this example from the recent "balloonboy" silliness.

Sheriff: No indication balloon ordeal was hoax

That's the headline.

Here's what it says in the story:

The sheriff's office said it does not believe at this point that the balloon episode was a stunt, but investigators planned to question the family again Saturday.

"The sheriff's office doesn't believe" is different from "no evidence." The sheriff has a responsibility to be very conservative in his accusations. He may find some scant evidence but find it is not enough to move forward with, in which case he would make a statement such as you see here. And perhaps follow up with some more questions. (The additional questions certainly indicate he still believes there is a possibility it was a hoax)

Heck -- people have already seen plenty of evidence that satisfies a much lower criteria than law enforcement must meet. The father is an attention-seeking jerk. The parents are willing to put the whole family on TV hours after this ordeal, waking the kids up at some ungodly hour in the morning mountain time to get on a nationwide morning show. They let their kid vomit on national TV and don't ask to stop the interview; the interviewer had to suggest it. The kid keeps referring to everything as being "for a show" -- indicating that this instruction came from his parents. And then there's the possibility that the family called the media as well at 911.

There is definitely some evidence that it was a hoax. Enough evidence to prosecute? I don't think so. But indications? Certainly. The man has an absurd lack of judgment and is an attention-seeker. That's not evidence, but it's enough for people to judge him harshly. And I don't think that judgment is unfair.

The headline is in conflict with the facts (and even the story itself). Considering the headline as part of the story, the story is self-contradictory. And the least accurate part is in giant, red type.

It may well be that "Sheriff doesn't believe balloon ordeal was hoax" is less compelling than "Sheriff: No indication balloon ordeal was hoax." But, so what? Well, it is a big deal if it means the difference between your story getting clicked and you getting paid.

Is this the tip of an iceberg? I've always taught the kids that commercials were essentially institutionalized (and adult-approved) lying for the sake of making money. But what happens as these principles are applied to our information sources? I guess we know what happens: trading accuracy for dollars becomes a science and sensationalism wins.

Accuracy and your brain lose.

Argue, if you like, that the truth can still be discerned from this story. Fine. Then we can talk about percentages of people who are able to come away with an accurate message, or about the the trade-off being increased cognitive load required to make the news useful vs. dollars. In the latter case, this would be the informational version of shrinking the contents of a box of cereal as a sneaky way to raise the price of the cereal.

I think it's no good.

Posted by James at 9:06 AM | Comments (5)

October 15, 2009

Robots Ate Your Brain: Marketing and Headlines

This Fast Company article proves that the humans have given up and are going to allow the robots programmed by marketing experts to eat our brains. It's about The Huffngton Post using an algorithm to decide what headlines stories will run with.

The idea is simple; for the first five minutes that a story is posted, two separate headlines are randomly displayed to readers. the one that gets the most clickthroughs is considered the better headline, and so that one is shown to all the readers after that.

It's true, a headline is important for bringing readers in. A headline is also important to framing an article and informing people who are just browsing the headlines.

Is optimizing clickthroughs the same as optimizing the value to the reader? Reader value is in how quickly and accurately the headline informs the reader. This algorithm only cares about clicks.

We can guess that what gets people to click: sensationalism. Subtly sacrificing value for attention is one of the ways your world becomes dumbed down, replacing expertise with algorithms. I am not opposed to algorithms that attempt to capture, recreate or enhance expertise, but this one cares nothing about value or meaning.

Posted by James at 3:11 PM | Comments (6)

October 3, 2009

Apple Butter

Apple Butter

My first attempt at canning happened yesterday. I cooked down about 20 apples into 3 pints of what approximates apple butter. It's really a thick, spiced apple filling. I think it would be excellent spread on shortbread, or heated and placed on top of vanilla ice cream.

In any case, I went to the trouble of canning 6 half-pints of this stuff. Gave them the 10 minute boiling water bath and all. But Maggie is still afraid they would breed botulism, so they're sitting in the fridge.

They're flavored with cinnamon, nutmeg, clove and a tiny bit of ginger. Sweetened with caramel squares and some brown sugar. It's tart, strongly apple-flavored and good stuff.

Posted by James at 11:47 AM | Comments (5)

Do Your Disney Research!

I couldn't help but comment on this New York Times review (from Seth Schiesel) of Dead Space Extraction for the Wii:

You know the Disney theme park rides, like the Haunted Mansion, where you climb into one of those little clamshell contraptions, and it moves along a track while swiveling and tilting to show you different scenes? Well, what if you gave all the riders some sort of paintball-style infrared rifle and put targets to shoot at in each scene? Something along the lines of "Shoot the vampire with the silver bullet before he eats Sally Sue." [...]

I want full credit when such attractions start popping up at amusement parks around the country in the next few years.

Wait, you mean like this?

That would be Buzz Lightyear's Space Ranger Spin. It's been around for over a decade.

So I don't expect credit will be coming your way any time soon, Mr. Schiesel. It's good for video game reviewers to leave the house once in a while, or even do their research.

You can even download an online game that says it allows you to collaborate with theme park visitors.

Posted by James at 8:59 AM | Comments (0)