Robot books you say?

That’s right, robot books.

 http://education.guardian.co.uk/egweekly/story/0,,2252153,00.html

Well, books put together completely by a computer anyways. I’m still not sure how much I like this idea, but I have to admit it sounds pretty neat. I mean, who wouldn’t want to read the 2007-2012 Outlook for Edible Tallow and Stearin Made in Slaughtering Plants in Greater China?

Some reassuring reading

I was doing some reading last night for class and came across a great excerpt that’s done a lot to calm my nerves. Let me say up front that I’ve been having some trouble adjusting from being a history major to a LIS student. Many of these new concepts we’re learning seem completely foreign to me, as they’re either too abstract or theoretical for my pragmatic mind. So it was with great relief that during my reading of Marcia Bates’s excellent article The Invisible Substrate of Information Science I stumbled upon this: “Often [master's students in information science programs] have considerable difficulties during the first few weeks of the program, because at first it feels alien to think about a resource in terms of the features that matter to the organization and retrieval of it, rather than in terms of mastering its content.” 

Whew! Now I’ve got to start thinking about what my excuse is going to be next month when I still don’t know what’s going on. 

It’s Quiz Time!

Here’s a fun quiz I came across. Maybe my score of 17% explains why I’ve never been a big fan of abstract art. Enjoy!

http://reverent.org/an_artist_or_an_ape.html

Hmmm…

So I know I haven’t been keeping up with this little blog of mine very well, it’s just been kinda hard to think of anything interesting to write about that might actually apply to what’s going in good ol’ 5033. With that in mind I was pleased to have come across an article which lead me to this site. http://www.intrade.com//?request_operation=main&request_type=action&checkHomePage=true#

I’m usually the last person interested in hearing about anything related to markets or trading, but my rabid obsession in this year’s primary has actually caused me to give some real thought to what these prediction markets might mean in terms of giving us some insight into what could happen down the road. Let me say up front that I’m a huge Obama supporter, but looking at this information in an unbiased way I still think it’s very interesting that Obama has almost a 30% lead (at the time of the writing anyways) over Hillary in the prediction markets whereas in most national polls they are running neck and neck.

This leads me to the question “why are the polls and prediction markets so far off from each other?” Polls would seem to be a good, though not perfect, way of telling who might win in an election though often times they can be so far off it makes everyone wonder what went wrong (see New Hampshire). Sure they are just a small sampling of the population, and they are essential filtered through whatever organization is conducting them (by filtered I mean the numbers are usually tweaked depending on who is expected to actually turn out to vote) but one would think that any survey that asks people which candidate they plan on voting for would be a good indicator of how people will, you know, actually vote.

Prediction markets on the other hand aren’t asking people who they plan on voting for, they’re asking them who they think will win, or who they think everyone else will vote for. Now in some ways this seems far less scientific to me, as I would think it’s much harder to try to gauge how your neighbor might vote then it would be to guess how you’ll end up voting. But looking at it in a completely different way I think markets could actually be more accurate than polls. 

As I said before polls are essentially filtered to reflect who pollsters think will actually come out to vote, and this seems to me to be a not-so-good way to predict the outcome. Previous elections might give them some insight into who will turn out, and people are typically asked if they are planning on voting or not, but in the end I think all these little “what ifs” could increase the margin of error in polls by up to 10-30% (Zogby predicted Obama would win California by 13% and he ended up losing to Clinton by 10%).

Unlike polls, prediction markets seem to be more about trying to tap into the vast amounts of knowledge that are the out there. They’re often quite accurate when it comes to business use, and were even pretty dead on in the 2006 elections. Part of me thinks this might also have something to do with the fact people are putting their own money on the line, so they’re actually invested in the outcome, unlike polls where voters might change their answer right after speaking with a pollster and not think twice about it. Ultimately I think prediction markets are unedited and seem to better express our collective wisdom about the future, whereas polls are more like edited snapshots of the past. I’m sure this comparison doesn’t always hold water and someone far more knowledgeable on the subject could shoot a million holes in most of my theories, and so I wouldn’t be at all surprised if I’m totally proven wrong this summer. This primary should be over by then right? Right??

My first post

Despite what you may have heard this is in fact in my first post. I can’t think of much to write at the moment so instead I’ll leave you with a comic I wrote a while back that might help illustrate how my mind’s been feeling lately.