He is a name many of you know. If you don’t know him, he literally wrote the book on computer science and operating systems. Several dozen of them, really. And when I say ‘the’ book, I mean the book that is used in your basic theory class. The kind that people call by the author’s last name. He’s also responsible for a small operating system named Minix, which inspired a young finn named Linus Torvalds to start his own operating system, Linux. This certainly explains how he’s been able to keep bringing more servers and pipeline online as people attack his site… and it gives instant crediblity to his claims of being under attack. Basically, this man knows his math. He knows his theory, and his implementation. He also knows what he knows, and he knows what he doesn’t know.
Basically, if I wanted to pick one man to analyze a completely complex and messed up environment, reduce it to structured principles, it would be him.
Interesting! I skimmed his little “out-of-the-closet” thing earlier today, and was befuddled as to why he thought any of it would be surprising. I had no idea that he was really a bigshot in the world of computers.
Excellent question. One to which I have little idea of the answer. He has admitted that he’s in favor of Kerry, but he’s been scrupulous in keeping the site non-partisan. I can’t really see why anyone would want his site eliminated.
I’ve been following that site closely since spring, and in teh last few weeks I’ve been taking it with a huge shaker full of salt. Not because I doubt Tannenbaum’s methodology, but because I’ve lost all faith in political polling at all.
Normalizing the data, push polls, confounding factor’s such as the cell phone voters, all make it nearly impossible to get a clear picture of what teh trends are from any small sample. Unless one side is running away with the thing, polls are useless. I don’t know who will win tomorrow. Zogby doesn’t know. Rassmussen doesn’t know. Tannenbaum has taught me a valuable lesson: don’t pay any attention to close polls.
Karl Rove, thank goodness, is no Patton. More of a Nixon. Well. More of a CREEP.
The debate is, the man is an american, but an expat, the man supplies a tool that we are using to judge this election, one that has been recognized as good, independently. Then we suddenly learn that not only is he a name, but he is an expert in the field of complex systems, that he really does understand polling on a level most pollisters do not… I speak as an expert in market research here, having been involved in it for most of my life. (Been involved with such places as Bruskin Goldring, DDC, uhm. JD Power, at a position where I interact with the senior staff and those who design the polls and those who implement them.)
How does this change the view of his site, knowing his really ironclad reputation for A: speaking his mind, and B: integrity.? Personally, I respect it more, now.
Okay, you got one of my jokes. As far as learning not to trust the polls, well… yeah. That’s exactly my point. Mr Tannenbaum sees what is, and then explains it. Well. He’s been a teacher for a long time.
At a rather stunning 55% for Kerry compared to 40% for Bush. It would be interesting to know if they are weighting the results of that poll into their standard polls somehow.
A cell phone poll that didn’t contact anyone over the age of 29? Why was that?
Anyway, here is what The Zogby, himself, says about this cell phone issue - (and a question raised about the missing Amish vote in these telephone polls.)
It might be that the reason Zogby says “the nature of the poll is outside polling industry standards” is because those polled were subscribers to some sort of MTV thing. I mean, if Bush got 40 whole percent, I’m amazed.
As far as learning not to trust the polls, well… yeah. That’s exactly my point. Mr Tannenbaum sees what is, and then explains it. Well. He’s been a teacher for a long time.