All Things Mathematical
February 16, 2009 9:22 PM Subscribe
Somewhere between 538 and xkcd, Ask Doctor Math is an advice column for practical math questions.
So far he's covered the meaning of "average", the statistics of being middle class (in response to this article at the NYT, previously), the Infinite Monkey Theorem, and the chance of Ruth Bader Ginsburg dying in the next 5 years, among other topics.
So far he's covered the meaning of "average", the statistics of being middle class (in response to this article at the NYT, previously), the Infinite Monkey Theorem, and the chance of Ruth Bader Ginsburg dying in the next 5 years, among other topics.
In my mind I'm trying to picture an appropriate x-y axis on which fivethirtyeight.com and xkcd.com are connected by a straight line with Ask Dr. Math somewhere in the middle and I do believe that I'm failing. If I extend the axis a few orders of magnitude out of scale I seem to have found a solution, but I note that at one extreme end I seem to have a Geocities fetish page dedicated to stalking Ann Coulter's Adam's Apple followed by some trainspotter who has an entire encyclopedia of facts about the design, production, deployment and life-span of highway grade asphalt of all sorts, including porous, permeable and non-permeable asphalts.
Am I supposed to be using more than two dimensions for my grid? There is apparently a tolerable solution involving a plot of a single axis across seven dimensions. If so, I can only note that there's no way in hell I'm solving this on paper in a lecture hall in a GE class, mainly because I'm still two weeks within the drop date and I respectfully submit that the professor is likely dangerously mad from a severe brainworm infestation and/or glue.
posted by loquacious at 10:12 PM on February 16, 2009 [12 favorites]
Am I supposed to be using more than two dimensions for my grid? There is apparently a tolerable solution involving a plot of a single axis across seven dimensions. If so, I can only note that there's no way in hell I'm solving this on paper in a lecture hall in a GE class, mainly because I'm still two weeks within the drop date and I respectfully submit that the professor is likely dangerously mad from a severe brainworm infestation and/or glue.
posted by loquacious at 10:12 PM on February 16, 2009 [12 favorites]
loquacious: Alphabetical order.
posted by CaseyB at 10:44 PM on February 16, 2009 [7 favorites]
posted by CaseyB at 10:44 PM on February 16, 2009 [7 favorites]
I admit I've only read the front page of his blog, but how exactly is this similar to xkcd? Pretty cool stuff nonetheless.
posted by Marisa Stole the Precious Thing at 10:56 PM on February 16, 2009
posted by Marisa Stole the Precious Thing at 10:56 PM on February 16, 2009
I guess what I had in mind were the more mathy comics on xkcd, and how they blend humor with math that I wanted to learn more about. In my opinion, ADM trades a lot of the humor in exchange for being more informative and relevant, but without going so far as 538.
posted by piers at 11:44 PM on February 16, 2009
posted by piers at 11:44 PM on February 16, 2009
Aw, you missed my favorite xkcd.
Hadn't seen Dr. Math before, thanks for sharing.
I do think it's somewhere in between; xkcd is not often explanatory, doesn't really have a central theme (perhaps existentialism?), and is purely entertainment.
538's goal is exposition of statistics, particularly as applied to politics (the central theme). It is entertaining, but I think most people don't read it primarily for that purpose.
So I guess that means Dr. Math explains things somewhat, is partially for entertainment but partially informative, and had a loosely defined theme (?)... hmm, dunno.
posted by nat at 11:56 PM on February 16, 2009
Hadn't seen Dr. Math before, thanks for sharing.
I do think it's somewhere in between; xkcd is not often explanatory, doesn't really have a central theme (perhaps existentialism?), and is purely entertainment.
538's goal is exposition of statistics, particularly as applied to politics (the central theme). It is entertaining, but I think most people don't read it primarily for that purpose.
So I guess that means Dr. Math explains things somewhat, is partially for entertainment but partially informative, and had a loosely defined theme (?)... hmm, dunno.
posted by nat at 11:56 PM on February 16, 2009
I guess what I had in mind were the more mathy comics on xkcd, and how they blend humor with math that I wanted to learn more about.
Don't mind me. I'm just doing my job as a grumpy old coot. Good post.
posted by loquacious at 12:24 AM on February 17, 2009
Don't mind me. I'm just doing my job as a grumpy old coot. Good post.
posted by loquacious at 12:24 AM on February 17, 2009
If 538 is a vector in a hyperspace where each dimension can be used as a measure of what a website is like, and xkcd is another, I don't think you can draw a line between the two that contains Ask Doctor Math.
posted by delmoi at 1:35 AM on February 17, 2009
posted by delmoi at 1:35 AM on February 17, 2009
I'm a real doctor of math.
I'm a complex doctor of math. I have imaginary parts.
Cobordism.
posted by twoleftfeet at 1:46 AM on February 17, 2009 [2 favorites]
I'm a complex doctor of math. I have imaginary parts.
Cobordism.
posted by twoleftfeet at 1:46 AM on February 17, 2009 [2 favorites]
I don't know where people are getting the idea that "somewhere between" means there exists a linear equation defining the line, much less that Dr Math bisects the distance. I'd say a more likely interpretation is that both xkcd and 538 exist as vertices on a directed graph, with Dr Math also a vertex that will be included in any path connecting the other two. That is to say, Dr Math is the coffee shop that you keep passing as you try to walk the bridges of Koenigsberg.
(Firefox spellcheck knew "Koenigsberg" but not "vertices". How about Euler? No, that's OK. Vertexes? OMG, that worked. *bemoans kids on lawn*)
posted by DU at 4:49 AM on February 17, 2009 [1 favorite]
(Firefox spellcheck knew "Koenigsberg" but not "vertices". How about Euler? No, that's OK. Vertexes? OMG, that worked. *bemoans kids on lawn*)
posted by DU at 4:49 AM on February 17, 2009 [1 favorite]
Am I supposed to be using more than two dimensions for my grid? There is apparently a tolerable solution involving a plot of a single axis across seven dimensions.
Remember that "between" is only defined for n-1 dimensional objects in an n dimensional space. Also, n is not 7 but rather the Hausdorff-Besicovitch dimension of the Internet. Thats enough hints for now!
posted by vacapinta at 5:07 AM on February 17, 2009
Remember that "between" is only defined for n-1 dimensional objects in an n dimensional space. Also, n is not 7 but rather the Hausdorff-Besicovitch dimension of the Internet. Thats enough hints for now!
posted by vacapinta at 5:07 AM on February 17, 2009
I wish I had this guy for my stats class back at school. Maybe I would have learned something.
posted by backseatpilot at 5:26 AM on February 17, 2009
posted by backseatpilot at 5:26 AM on February 17, 2009
Nice!
posted by GenjiandProust at 5:49 AM on February 17, 2009
posted by GenjiandProust at 5:49 AM on February 17, 2009
Theorem: 538, xkcd, and Dr. Math lie on a plane in (any) hyperspace, and can be connected by a parabola.
posted by fantabulous timewaster at 6:38 AM on February 17, 2009
posted by fantabulous timewaster at 6:38 AM on February 17, 2009
interesting blog.
the monkey case makes the simplifying assumption that all the letters in the text are independent, which they're not in real English (or any other language). how does this affect the results? real texts are sampled very narrowly from the space of possible letter sequences. Does this make it even less likely the monkeys would ever converge on them? granting that it would still be in massively long time regardless.
posted by cogneuro at 6:43 AM on February 17, 2009
the monkey case makes the simplifying assumption that all the letters in the text are independent, which they're not in real English (or any other language). how does this affect the results? real texts are sampled very narrowly from the space of possible letter sequences. Does this make it even less likely the monkeys would ever converge on them? granting that it would still be in massively long time regardless.
posted by cogneuro at 6:43 AM on February 17, 2009
(Firefox spellcheck knew "Koenigsberg" but not "vertices". How about Euler? No, that's OK. Vertexes? OMG, that worked. *bemoans kids on lawn*)
A couple of people at work who really should know better use "vertice" as the singular.
posted by Foosnark at 6:44 AM on February 17, 2009 [1 favorite]
A couple of people at work who really should know better use "vertice" as the singular.
posted by Foosnark at 6:44 AM on February 17, 2009 [1 favorite]
cogneuro: "interesting blog.
the monkey case makes the simplifying assumption that all the letters in the text are independent, which they're not in real English (or any other language). how does this affect the results? real texts are sampled very narrowly from the space of possible letter sequences. Does this make it even less likely the monkeys would ever converge on them? granting that it would still be in massively long time regardless."
The monkey case makes a major error when he assumes that each block of 360.000 characters is independent and discrete, and thus that the monkeys only produce 48 blocks a day.
1-360.000 is one block which may contain The Hobbit
2-360.001 is another
3-360.002 is another again, etc...
Thus, there's over 17 million blocks a day produced. Which makes little difference, when the overall scale is taken into consideration.
posted by benzo8 at 12:00 PM on February 17, 2009
the monkey case makes the simplifying assumption that all the letters in the text are independent, which they're not in real English (or any other language). how does this affect the results? real texts are sampled very narrowly from the space of possible letter sequences. Does this make it even less likely the monkeys would ever converge on them? granting that it would still be in massively long time regardless."
The monkey case makes a major error when he assumes that each block of 360.000 characters is independent and discrete, and thus that the monkeys only produce 48 blocks a day.
1-360.000 is one block which may contain The Hobbit
2-360.001 is another
3-360.002 is another again, etc...
Thus, there's over 17 million blocks a day produced. Which makes little difference, when the overall scale is taken into consideration.
posted by benzo8 at 12:00 PM on February 17, 2009
At a math camp I went to when I was 17, the instructor threatened to kick a student out of the class for an hour if he used the word "matrice" again.
posted by Hactar at 12:16 PM on February 17, 2009
posted by Hactar at 12:16 PM on February 17, 2009
"the monkey case makes the simplifying assumption that all the letters in the text are independent"
I don't think that matters, does it? Even if the text were all "a"s, there would still be the same probability of matching it, assuming the characters being output were independent and equally likely. I don't know, sounds like a question for the Doctor.
posted by piers at 1:57 PM on February 17, 2009
I don't think that matters, does it? Even if the text were all "a"s, there would still be the same probability of matching it, assuming the characters being output were independent and equally likely. I don't know, sounds like a question for the Doctor.
posted by piers at 1:57 PM on February 17, 2009
loquacious: Alphabetical order.
posted by CaseyB
More like ASCII order, I think.
There are a number of differences between alphabetization as carried out by hand and as carried out by computer. In most traditional publications, numbers are alphabetized as though spelled out -- so 538 would be filed under F. This is one of several quirks that computer software generally doesn't observe, because it would be more complicated to program (or impossible if the context is too generic; 538 should be alphabetized differently if the language isn't English!). Most dictionaries ignore spaces as characters; my phonebook still files "Mc" as "Mac"; etc., but these subtleties are likely to be lost under the influence of computers. You're probably wise to look for "2001" before "Amelie" at Blockbuster.
I think the old style has merit in one respect, not in that it's old, but that it eliminates some ambiguity. All of the conventions I mentioned above are designed to help you find what you're looking for on the first try, even if you've only heard the name and don't know how it's formatted in writing -- so long as you can spell to begin with! So forgive me this bit of apparent pedantry; I don't mean to call anyone wrong, but just to share something I think is kind of neat.
posted by aws17576 at 3:00 PM on February 17, 2009 [1 favorite]
posted by CaseyB
More like ASCII order, I think.
There are a number of differences between alphabetization as carried out by hand and as carried out by computer. In most traditional publications, numbers are alphabetized as though spelled out -- so 538 would be filed under F. This is one of several quirks that computer software generally doesn't observe, because it would be more complicated to program (or impossible if the context is too generic; 538 should be alphabetized differently if the language isn't English!). Most dictionaries ignore spaces as characters; my phonebook still files "Mc" as "Mac"; etc., but these subtleties are likely to be lost under the influence of computers. You're probably wise to look for "2001" before "Amelie" at Blockbuster.
I think the old style has merit in one respect, not in that it's old, but that it eliminates some ambiguity. All of the conventions I mentioned above are designed to help you find what you're looking for on the first try, even if you've only heard the name and don't know how it's formatted in writing -- so long as you can spell to begin with! So forgive me this bit of apparent pedantry; I don't mean to call anyone wrong, but just to share something I think is kind of neat.
posted by aws17576 at 3:00 PM on February 17, 2009 [1 favorite]
If you input all of the works of Shakespeare into a Markov-generator, how long do you figure for the Markov-generator to accidentally put one of the plays back together completely by chance?
posted by Navelgazer at 9:21 PM on February 17, 2009
posted by Navelgazer at 9:21 PM on February 17, 2009
Depends on the token length. If the tokens are words, longer than the present age of the universe. If the tokens are acts, not so long.
posted by fantabulous timewaster at 9:44 AM on February 18, 2009
posted by fantabulous timewaster at 9:44 AM on February 18, 2009
« Older Invasional Meltdown | Godwin comes to the banking crisis Newer »
This thread has been archived and is closed to new comments
someone actually did this experiment once with real live monkeys and found that the monkeys were extremely overly fond of the letter "s" for some reason. Also, they [the monkeys] really enjoyed urinating on the keyboard
The fact that he clarified which of the possible beings enjoyed urinating on the keyboard makes me love him.
posted by Lemurrhea at 9:33 PM on February 16, 2009 [6 favorites]