I predict someone will make millions inventing a rubber doohickey thing that you can use to help open jars with.
What would be nice about a predictions-centralizer would be that people could use the results to advocate for getting some of the worst media offenders fired. Normally I dislike the saberization of industries, but political pundits could sorely use some new metrics to combat the "do we get lots of eyeballs when she says outrageous thing" forces.
Ok, some questions:
General architecture:
What would a this site look like in terms of overall architecture?
Is it just a list of predictions, tagged by predictor, subject, timeframe?
Is the key figure a running scorecard once we have enough information?
Universe:
Whose predictions should we be paying attention to, and in what fields? Is this just politics, current events, or Dr. Z laughable sports picks?
Would you also want to track the prediction markets over time? The polls?
Data handling:
Surely this needs to be automated. How? What data mining could one do?
How effective would a simple pull approach be: i.e., "send in your predictions, mighty bloggers." Could getting some kind of institutional affiliation (School of journalism, e.g.), make bona fide paid pundits submit this stuff?
getting some of the worst media offenders fired
It's always straight for the jugular against the poor press. Nothing we do will get Deborah Soloman fired, people, we just have to live with it.
Surely this needs to be automated.
Why would you need that? If it's policing you're looking for, rather than finding the best predictor, I assume you could have people submit pundit's predictions with links. You only care about those pundits/predictors that other people pay attention to, if it's about policing.
To a certain degree, the Long Wager site does this, but it's not for general prediction tracking. A friend and I took a look at it a few years ago, but it ground to a halt. (My fault!) Except for financial predictions like Dow 36000 or binary outcomes in sports and politics, assigning a truth value to an outcome is pretty hard.
6: Dude, if you think that's the jugular, I have an anatomy class I'd like you to take.
As far as the site, I'd think you'd want it to be tiered, with different levels of functionality (ugh, that word) for different groups of people. Personal submissions could be self-posted; web-only famous people could be linked and submitted by others; and beyond-web media figures could be submitted by anyone, but would have to include a video or audio clip or transcript of the prediction.
You'd also want to sort the predictions broadly by category, so that the Iraq picks weren't cluttering up the Afghanistan picks, or the Countdown to Universal Healthcare etc.
assigning a truth value to an outcome is pretty hard.
Eh, this is kind of true and kind of not true. I had an argument two weeks ago in which I said oil was on its way to $150 a barrel. I'd have been happy to slap a timeline on it. You can say the same thing about winning political races, timing of dropping out, which issue people will say is the most important in exit polls in November, when the Supreme Court will next rule on X topic, etc.
IEEE just did a great set of predictions on the singularity, actually. (Universal healthcare made me think of this...)
I can't believe no one has acted on my call in July 2006 for keyboard pants (with crotch mouse).
Oh, wait, THEY TOTALLY DID!
Gimme some money!
Ogged, you're starting to remind me of my grandmother when she was on her deathbed. She began apportioning her possessions to anyone who showed up for a visit. But instead of deaccessioning tchatchkes, you're handing out intellectual property. It's very sweet and all, but you really are leaving the blog, aren't you?
6: Nothing we do will get Deborah Soloman fired, people, we just have to live with it.
You're right, I'm going for their legacy now. I want to gnaw on David Broder's skull. I want children 100 years from now to recoil in horror at the mention of Cokie Roberts' or Tim Russert's name. In 50 years I want a new version of Godwin's Law to spring up that substitutes any mention of the American Political Press from 1993-2008 for the Nazis. I want a statue erected on the mall showing George Will and Fred Hiatt giving George W Bush a blowjob labelled "Never Again". I want High School Civics classes of the future to demonize Roger Ailes, Murdoch, GE and Media ownership*. I want Tom Friedman's descendant to change their names out of shame.
*Sop to JE.
In the long run, all journalists will die.
Whoever did this blog would actually have to read a lot of pundits. Maybe even watch a lot of Fox or CNN or something awful like that. It's not worth it.
I just got home and the big group house across the street is playing contemporary country at their keg party. Sometimes living on Capitol Hill really sucks. When will property values drive these young staffers the fuck out of here?
15: And then we (or some of us) get to gnaw on their skulls.
When will property values drive these young staffers the fuck out of here?
Cutbacks forced by the mortgage crisis will reduce the Senate to one member per state and small population states will have to start sharing Representatives. Staffer layoffs restructuring in order to provide enhanced governance services will follow.
The banner ads for antipsychotics may pay for the whole site
People have neglected to mention impaling.
Google will implement this as an automated service. It will undergo a rocky period in 2015 when Google begins not including the increasing number of predictions that Google will soon become a malignant bigger-than-any-government entity which will use its control of information on the Web to achieve its ends by means fair or foul. The crisis will be resolved when Google uses its power to effectively render anything written by such folks essentially unreachable on the Web. The resulting Internet Comity will last for a 1000 years.
More seriously, groups like Media Matter have people watching tv news all day. They could probably be able to log predictions as they come in.
People have neglected to mention impaling.
YES WE CAN!
Come on, people, Web 2.0. It's a blog, and each day (or so) predictions that have come due pop up as posts, and people get to read how right or wrong so-and-so was. You collect predictions mostly by people sending them in to you, because Drum, Yglesias and Atrios will all link to you when you launch. The only tough part is starting up, because you'll need a backlog of predictions so that there'll be regular content. Filler can be those predictions of life in 2000 from 1934 or something. You can slug things by category and pundit, for easy searching/lookup.
The thing people forget when promoting web 2.0 is that it can sound like a lot of work.
I've wanted to do a sports version of this for years!
They really missed the obvious placement for the trackpoint nub, didn't they.
This sounds ludicrously easy and kind of fun. If you can guarantee that someone will provide links to the predictions as they occur, I will totally set up this blog.
Ideally, each prediction would get two posts, right? One marking the prediction, and one that occurs in the future for the "how did he do?" effect.
Ideally, each prediction would get two posts, right?
Not necessarily. Like I said in the old post, if someone makes a prediction today, you just write a post and schedule it for when the prediction is due. On that day, up it pops, and merriment follows. After you have more of an archive, it might be fun to note predictions as they're made, and also note that so-and-so has been wrong 6 of the last 7 times, or whatever.
I predict the Being and Time reading group will be revived before this blog is set up.
guarantee that someone will provide links to the predictions as they occur
My guess is that there would be a few months (2-6?) before the thing really takes off, when you'd have to work pretty hard to find predictions (work pretty hard meaning "read lots of news stories" and "set up lots of google alerts for things like 'months from now'.")
If mrh isn't going to do it, we will at our place. Seriously, it's a totally great idea. And a great way to make high-profile people haaaaaaaaaate you. What could be more fun.
I'd call dibs, Ari, but I probably won't have much free time to do it before next week, so if you're itchin'... go for it.
Next week? You're kidding, right? It's all yours. Just make sure that we have the URL, okay? And if you decide not to do it, pass that along.
I did just register punditpredictions.com, though.
Not necessarily. Like I said in the old post, if someone makes a prediction today, you just write a post and schedule it for when the prediction is due. On that day, up it pops, and merriment follows. After you have more of an archive, it might be fun to note predictions as they're made, and also note that so-and-so has been wrong 6 of the last 7 times, or whatever.
Couldn't you just do most of this with Google Calendar?
Couldn't you just do most of this with Google Calendar?
Goosing stock prices again, Tim?
Ogged's plan relies on someone making predictions keyed to relatively small-scale time ranges, like a specific day or week.
40: predictions based on well defined intervals are made more or less constantly. "In the next year" or "within six months" or "by the end of the week" etc. I mean, "the next six months" is now proverbial, no?
It's a blog, and each day (or so) predictions that have come due pop up as posts, and people get to read how right or wrong so-and-so was.
The thing that More-Responsible-Than-Me-Greg and I got bogged down in was actually quantifying the predictions. If someone says "Lakers in 6" or "$150 oil by July 4", that's something that you can put a check up or down on. What do you do with "the next six months will be the critical determinant of success or failure in Iraq"? Or "George Bush's war on terror will be vindicated"? (There was a famous and high-larious example of someone predicting a boom in Dan Quayle revisionism in the same issue of TNR that started trying to make a contrarian case for Quayle as a decent politician, wasn't there?)
As a recommendation, the Macintosh press will be a fruitful source for building a backlog of predictions, and Steve Jobs' deranged fanboys can easily be trolled into linking to you. See also: Red Sox fans. You could fund a nice dinner out with Google Ads + Sifu alone!
Well, who are you going to catalogue? The opinions change as the propagandists feel they need to change, and thus the predictions change as though nothing happened.
It would be kinda fun, but wouldn't end the nonsense...
Be sure to always click on an ad when you visit a major news site, Ny Times, Yahoo, etc. It keeps media free, and costs big corporations money, which will cut down the profit margins...
As a recommendation, the Macintosh press will be a fruitful source for building a backlog of predictions, and Steve Jobs' deranged fanboys can easily be trolled into linking to you. See also: Red Sox fans. You could fund a nice dinner out with Google Ads + Sifu alone!
You think I read red sox blogs? Those people are fucking nuts. I watch the team, and avoid everything else. Jeez. I might be from Boston, but I'm not a goddamn lunatic.
Web 2.0 means you get your readers to do the work for you. You award one "point" for every prediction by a major pundit posted by a reader. Readers get to wager their points on the outcome of the predictions.
punditpredictions.com
Cred's dead, baby. Cred's dead.
I did have some fun with bits of this a few years ago, just running Daily Telegraph editorials from the runup to the war, four summers on. There's a lot out there in the way of totally worthless predictions that can be mined. Just go and see what people promised the "surge" would achieve.
I couldn't find a usable prediction in this a.m.'s Washington Post, but in the process of looking, I read an entire David Broder column.
When waterboarding is finally outlawed, it will be a hollow victory unless we can also ensure that prisoners won't be forced to read Broder.
45: I might be from Boston, but I'm not a goddamn lunatic in that particular respect.
You think I read red sox blogs?
"Keys to the Game" is pretty good. Although in general all people of good faith are repulsed by the negativity and ingratitude exhibited by the bottom 20% (50%?) of Boston fandom (after 2 championships!). J.D. Drew -- what did he ever do to anyone? Perhaps now that he's sporting an OPS of .900 people will leave him be.
Oddly enough, I'm in the early stages of pitching a similar idea to a newspaper at the moment.
I've also wanted to gather a rogue's gallery of the most egregious Bush-era media people and Republican operatives. But I'm way too lazy to do it. A lot of people have to be shamed out of polite society, since prosecutions (much less impaling at the hog farm) seem very unlikely. You'd probbly want to fold together scandals, graft, lies, and silly predictions. There could be a whole page just enumerating Friedman units from various sources.
What do you do with "the next six months will be the critical determinant of success or failure in Iraq"?
If you're fortunate enough to be able to engage the pundit in dialog, you ask for clarification -- "How would you define 'success in Iraq'*? What precisely are you predicting will be different six months from today?" When they waffle, you say "well, gosh, you know we really can't give you a link unless there's a specific, measurable prediction." Until the site makes it big, of course, you don't have any leverage there, so you'll have to start with those pundits who do trade in specifics.
* HHAAHHAHAHAHAHAHAHAHAHA
Cred's dead, baby. Cred's dead.
You don't like the name? Well, shit, I'm open to suggestions.
Oddly enough, I'm in the early stages of pitching a similar idea to a newspaper at the moment.
To keep track of promised book reviews?
What do you do with "the next six months will be the critical determinant of success or failure in Iraq"? Or "George Bush's war on terror will be vindicated"?
I don't know why you couldn't list it under "Pointless" or "How Would We Know" or "What Does That Mean?"
And Deadcred.com seems like a good suggestion. Or maybe something playing off the notion of a credit rating service for pundits. You could even downgrade future offerings from said pundit by pseudo-specific amounts.
One rich source of material would be when bloggers claim that they will address a topic in more detail in a later post.
I would also like to see a website that depicts all the people whom various pundits have insinuated were homosexuals, terrorists, élites, or black radicals. Just a list. Then you could have a list of people who are actually homosexuals, terrorists, élites, and black radicals. You know, for comparison.
59: Discovering the network left as an exercise for the reader.
You don't like the name?
Oh, I like it fine, I was just offering a subtitle, or mouseover, or whatever.
Oh! Cool. Your approval is like cool, refreshing water.
cool, refreshing water
Yeah, but it's still a hundred fucking degrees outside here, with the weather.com folks calling for more of the same tomorrow. If they're correct, I predict it will suck a lot.
||
Remember the obesity report card thread? here's an article looking at the dismal statistics of success of such a strategy.
(For the record, I was fair-to-middling on the report cards back in that thread, and I've totally, totally abandoned that position, for many reasons, since then.
|>
||
Okay, Here's the article.
|>
Huh. I can see two possible mechanisms for that: (1) the obvious, that making someone pressured and insane about their weight is likely to make their eating disordered and make them fat or keep them fat, or (2) that the 'oblivious' parents were actually right that their children weren't 'really' overweight -- they might have been at a BMI that looked bad to the researchers, but were on an individually appropriate developmental track (say, getting stocky before a growth spurt), and the parents had a good sense of that. So the parents who thought their kids were overweight were right, and the parents who didn't were also right, and both groups stayed the same over the course of the study.
One rich source of material would be when bloggers claim that they will address a topic in more detail in a later post.
Or that Ogged is they are going on hiatus for X amount of time.
68: Or just that pointing out that someone is fat isn't enough to make him or her lose weight. At least it shows that the problem, if there is one, wasn't that parents just had no idea Junior was chubby and needed to be informed.
||
If you are designing a web application so that a balloon frequently pops up that says "All unsaved changes will be lost. Are you sure you want to navigate away from this page?", you should make sure that it doesn't pop up on a whole lot of pages where there is no save button nor any fields to fill with data to begin with.
I just want this bit of wisdom to get more circulation, because it is clear to me that not all of the relevant people are aware of it.
|>
I confess I don't see the interest in a compare-the-predictions site. People throw that sort of thing out all the time. Admittedly, I didn't read the original thread suggesting the whole thing.
Is it an "Aha! Gotcha!" thing? Or, more seriously, a means to taking down those who are normally taken seriously? I can see that, as events unfold, it would be of interest to see what various people had to say about it back when. Those people would, to my mind, chiefly be major NYT columnists and such. Not really bloggers (who, let's face it, have very little impact on the public imagination).
However. I am not a blog or pundit junkie to the degree many people here seem to be; which is odd, since most people I know think I'm rather a freak in the extent to which I dip into the blogosphere.
We might get LB to write for the promised book reviews blog, of course.
Di, it's mostly motivated by the fact that a lot of famous pundits have been pulling predictions out of their butts for decades, in supreme confidence that when their prediction comes due their light-headed audience will have forgotten all about it. Thomas Friedman is only the most famous.
"Di" = Parsimon. We regret the error.
It's a great idea, but these blog or website "ideas" only seem to become wildly successful if they are spontaneous. I was reading about the guy who started hot or not and it was random. (Not exactly random, he was sitting in a bar chatting with a friend about whether someone was "hot or not" and lo and behold.)
The other day I was thinking about sites which convey in very big picture and basic terms historical concepts. The radio show was talking about Buddhism and I was thinking it's always unclear where it came from and where it went and how it got there. It would be nice to just go to a site and get a graphical representation of some markers in this area in order to fit it into another historic context. Sort of a graphical wikipedia.
Another idea (don't steal this one) revolves around a website that just links to articles that satisfy the "I'm looking for something to read for 5-10 minutes" standard. A site that just contained links to something like this would be great. (Don't everyone rush to register the domain name . . . it's taken!)
Anyway, this falls into the same category.